DoE oriented reaction optimization on the lipase-catalyzed monostearin synthesis

Article abstract of DOI:10.1016/j.molcatb.2011.07.007

Recognition that trans and saturated fats have negative health effects drive researchers to develop alternative systems that can structure liquid oils into semi-solid plastic pastes for food applications. Monoacylglicerols (MAG) can be used as a promising molecule to achieve this structuring so we have optimized a biocatalytic batch process to the esterification reaction between 1,2-O-isopropylidene glycerol and stearic acid, catalyzed by Lipozyme RM IM, using response surface methodology (RSM) in a laboratory setting with 95% of conversion after 4 h.

Full text of DOI:10.1016/j.molcatb.2011.07.007

Journal of Molecular Catalysis B: Enzymatic 72 (2011) 313–318
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
DoE oriented reaction optimization on the lipase-catalyzed monostearin
synthesis
Ivaldo I. Juniora,b, Felipe K. Sutilia,c, Selma G.F. Leitec, Leandro S. de M. Mirandad, Ivana C.R. Lealb,
Rodrigo O.M.A. de Souzaa,∗
a Biocatalysis and Organic Synthesis Group, Chemistry Institute, Federal University of Rio de Janeiro, CEP22941-909 Rio de Janeiro, Brazil1
b Faculdade de Farmácia, Federal University of Rio de Janeiro, Campus Macaé, CEP22941 909 Rio de Janeiro, Brazil2
c Escola de Química, Federal University of Rio de Janeiro, CEP22941-909 Rio de Janeiro, Brazil3
d Instituto Federal de Educac¸ ão Ciência e Tecnologia do Rio de Janeiro, Campus Maracanã, CEP20270021 Rio de Janeiro, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 30 May 2011
Received in revised form 8 July 2011
Accepted 12 July 2011
Available online 21 July 2011
Recognition that trans and saturated fats have negative health effects drive researchers to develop
alternative systems that can structure liquid oils into semi-solid plastic pastes for food applications.
Monoacylglicerols (MAG) can be used as a promising molecule to achieve this structuring so we have
optimized a biocatalytic batch process to the esterification reaction between 1,2-O-isopropylidene glyc-
erol and stearic acid, catalyzed by Lipozyme RM IM, using response surface methodology (RSM) in a
laboratory setting with 95% of conversion after 4 h.
Keywords:
Biocatalysis
Lipases
© 2011 Elsevier B.V. All rights reserved.
Monostearin
Design of experiments
1. Introduction
yield, poor quality, dark-colored and burned-tasting characteris-
tics [6,7]. To overcame this issues the use of enzymatic process can
A better understanding of the role of dietary lipids in the
coronary heart disease (CHD) continues to give us insights into
metabolic effects of individual fatty acids and their impact on sur-
rogate markers of risk [1]. The recognition that trans and saturated
fats have negative health effects drive researchers to develop alter-
native systems that can structure liquid oils into semi-solid plastic
pastes for food applications because the traditional processes for
manufacturing spreads, margarines, and shortenings uses hydro-
genation and/or saturated fats to achieve the desired structuring
[2,3].
In this way, monoacylglicerols (MAG) can be used as a promis-
ing molecule to achieve this structuring by utilizing the properties
of gel phases (alpha gel and coagel) [4,5]. These MAG are commonly
produced based on alkaline-catalyzed chemical glycerolysis of nat-
ural oil and fats at high temperatures (220–250 ◦C) and elevated
pressure under nitrogen atmosphere leading to products with low
lead to an environmentally friendly approach, employing enzyme-
catalyzed synthesis of MAG by selective hydrolysis or alcoholysis
using 1,3-regiospecific lipases [8–10], esterification of glycerol with
fatty acids [11,12], and glycerolysis of fats or oils [13].
In this work we have optimized a biocatalytic batch process to
the esterification reaction between solketal and stearic acid using
response surface methodology (RSM) in a laboratory setting [14].
The lipase-catalyzed esterification [15] has been investigated as a
potential substitute to the traditional chemical glycerolysis, since
lipases as biocatalysts demand milder reaction conditions which
minimize energy costs, allow a better reaction control and conse-
quently provide higher-quality products [16]. RSM is a statistical
tool for developing and optimizing processes with one or more
responses influenced by several variables. The RSM advantage is
that it allows the user to gather large amounts of information from a
small number of experiments. The RSM use also enables to observe
the effects of individual variables and their combination of interac-
tions on the response.
∗
Corresponding author. Fax: +55 2125627001.
E-mail addresses: rodrigosouza@iq.ufrj.br, souzarod21@gmail.com
2. Experimental
(R.O.M.A. de Souza).
1
Fax: +55 2125627001.
Fax: +55 2125627807.
2
Heptane was purchased from Tedia Co. (R,S)-1,2-isopropylidene
glycerol from Sigma–Aldrich as well as all chromatofigureic
3
Fax: +55 2125627005.
1381-1177/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2011.07.007

314
I.I. Junior et al. / Journal of Molecular Catalysis B: Enzymatic 72 (2011) 313–318
100
90
80
70
60
50
40
30
20
10
0
into more volatile silylated derivatives in presence of pyridine and
N-methyl-N-trimethysilyltrifluoroacetamide (MSTFA). All GC–MS
measurements were carried out in duplicate (Dizge & Keskinler;
2008) using a DB 5–HT (Agilent, J & W. Scientific®, USA) capil-
lary column (10 m × 0.32 mm × 0.1 ␮m). The quantifying was done
based on calibration curves with internal standards. The GC–MS
samples were prepared by dissolving 0.1 g of the final product on
1 ml of n-heptane. 100 ␮l of this solution and pyridine solutions
of butanetriol (1 mg/ml) and tricaprine (8 mg/ml), used as internal
standards, were added on a flask forward by an addition of 100 ␮l
of MSTFA. After 15 min, these reactants were dissolved on 8 ml de
n-heptane. 1 ␮l of this sample was then injected into a Shimadzu
CG2010 equipment.
TL IM
RM IM
Novozymes 435
PS IM
79
71
77
10
0
10
20
30
40
50
60
Reacꢀon Time (min)
2.2. Lowry–Tinsley analysis
Fig. 1. Initial screening for immobilized lipases. Experimental conditions: 60 ◦C,
250 rpm and 1% (w/v) of enzyme (Lipozyme TL IM, Lipozyme RM IM, PS Amano
IM and Novozym 435).
The esterification rate was also measured using a modification
of the Lowry and Tinsley assay [17]. The depletion of fatty acid was
monitored as follows: 0.30 ml of the reaction solution, including the
buffer solutions was added to a tube containing 0.6 ml of n-heptane
and 1 ml of cupric acetate-pyridine (5%, w/v; pH 6.0). The final
solutions were vigorously mixed for 30 s in vortex, and the upper
organic phase was measured by a UV/visible spectrophotometer at
715 nm. Each reaction was analyzed in triplicate, and content con-
version, calculated according to the percentage difference for the
absorbance shown by the stock solution.
100
90
80
70
60
TL
RM
Cal B
PS Amano
50
40
30
20
10
0
41
25
2.3. Batch procedure
23
1
A stock solution containing 1,2-O-isopropylidene glycerol and
stearic acid in n-heptane in proportions of 2:1 (150 mM and 75 mM
respectively) was prepared. In 2 ml cryotubes, were poured 1 ml of
reaction medium, followed by the addition of appropriate enzyme
(1%, w/w or w/v). The cryotubes were then incubated in shaker at
250 rpm and 60 ◦C. The esterification rate was measured by a mod-
ification of the Lowry and Tinsley assay, at time intervals between
0 and 60 min. The depletion of fatty acid was monitored as follows:
0.30 ml of the reaction solution was added to a cryotube containing
0.6 ml of n-heptane and 1 ml of cupric acetate–pyridine (5%, w/v;
pH 6.0). The final solutions were vigorously mixed for 30 s in vortex,
and the upper organic phase was measured by a UV/visible spec-
trophotometer at 715 nm. Each reaction was analyzed in triplicate,
and content conversion, calculated according to the percentage dif-
ference for the absorbance shown by the stock solution. The results
were confirmed by GC–MS analysis.
0
10
20
30
40
50
60
Reacꢀon Time (min)
Fig. 2. Initial screening for free and lipases. Experimental conditions: 60 ◦C, 250 rpm
and 1% (v/v or w/v) of enzyme (Thermomyces lanuginosus, Rhizomucor miehei, Can-
dida antarctica B and PS Amano).
standards. Stearic acid (>98%) was purchased from Vetec Ltd.
Lipases were purchased from Novozymes and Amano.
2.1. GC–MS analysis
The GC–MS analysis was performed by using modified method
from EN 14105. Free fatty acids and solketal were transformed
a
b
100
90
80
70
60
50
40
30
20
10
0
93
94
94
94
100
91
93
93
93
91
88
90
80
70
60
50
40
30
20
10
0
79
72
0
1
2
3
4
5
6
0
1
2
3
4
5
6
Reacꢀon Time (h)
Reacꢀon Time (h)
Fig. 3. Conversion rates of 1,2-O-isopropylidene glycerol stearate (2:1, 60 ◦C, 250 rpm and 1%, w/v) by using Lipozyme RM IM (a) and Lipozyme TL IM (b) as biocatalysts.

I.I. Junior et al. / Journal of Molecular Catalysis B: Enzymatic 72 (2011) 313–318
315
100
a
b
100
80
60
40
20
0
80
75
82
82
82
80
79
81
81
82
81
79
75
60
40
20
0
0
1
2
3
4
5
6
0
1
2
3
4
5
6
Reacꢀon Time (h)
Reacꢀon Time (h)
Fig. 4. Conversion rates of 1,2-O-isopropylidene glycerol stearate (1:1, 60 ◦C, 250 rpm and 1%, w/v) by using Lipozyme TL IM (a) and Lipozyme RM IM (b) as biocatalysts.
Table 1
screening with 4 commercial immobilized lipases (Lipozyme TL
IM, Lipozyme RM IM, PS Amano IM and Novozym 435) and 4 com-
Real and coded values (+ higher level, 0 intermediate, − lower level) for the inde-
pendent variables, 24−1
.
mercial free lipases (Thermomyces lanuginosus, Rhizomucor miehei,
Candida antarctica B and PS Amano). A stock solution containing
1,2-O-isopropylidene glycerol and stearic acid (in a molar ratio
1:2) in n-heptane was prepared in order to avoid pipetting errors.
In 2 ml ependorfs, were added 1 ml of stock solution, followed
by the addition of appropriate enzyme (1%, w/w or w/v) and the
reaction analyzed each 5 min until 15 min. After that, the reaction
was analyzed at 30, 45 and 60 min. The results are summarized in
Figs. 1 and 2 for the immobilized and free enzymes, respectively.
As Lipozyme TL IM and Lipozyme RM IM are less studied in the
synthesis of MAG and lead to good results in the initial screening,
we decide to go further with these two enzymes evaluating the
reaction time needed for full conversion of substrates. As shown in
Fig. 3a and b, after 4 h of reaction it is not possible to detect any
considerable improvement on conversion rate for both Lipozyme
RM IM and Lipozyme TL IM immobilized enzymes. As from 1 to 4 h
a significant raise on reaction conversion is still clearly observed,
we decided to use 4 h as a standard for total reaction time in our
work.
Variables
−1
0
+1
60
1
100
250
Temperature (◦C)
40
0.1
50
50
50
0.55
75
150
Amount of enzyme (%)a
Substrate Concentration (mM)
Stirring (rpm)
a
By weight of total system.
2.4. Statistical analysis
The experimental designs and results analysis were carried out
using the software Statistica 6.0 (Statsoft, Inc., USA), according
with the significance level established to obtain the mathemati-
cal model. The significance of the regression coefficients and the
associated probabilities, p(t), were determined by Student’s t test;
the model equation significance was determined by Fisher’s F test.
The variance explained by the model is given by the multiple deter-
mination coefficients, R2.
Considering MAG as a food ingredient in industry process,
minor costs with starting materials such as the reagent 1,2-O-
isopropylidene glycerol represent a relevant and crucial step. Thus,
we also evaluate the ability of Lipozyme TL IM lipase and Lipozyme
RM IM lipase to catalyze the esterification reaction of stearic acid
and 1,2-O-isopropylidene glycerol in a molar ratio of 1:1 under
the same experimental conditions mentioned in Fig. 3, saving 1,2-
3. Results and discussion
3.1. Initial screening of commercial available free and
immobilized lipases
In order to evaluate the best biocatalyst for the synthesis
of 1,2-O-isopropylidene glycerol stearate, we perform an initial
Table 2
Experimental factorial design and results of FFD for enzymes studied.
Entry
Variable levels
Conversion (%)a
T (◦C)
E (%)a
S (mM)
St (rpm)
RMIM
TLIM
1
2
3
4
5
6
7
8
9
−1 (40)
+1 (60)
−1 (40)
+1 (60)
−1 (40)
+1 (60)
−1 (40)
+1 (60)
0 (50)
−1 (0.1)
−1 (0.1)
+1 (1)
−1 (50)
−1 (50)
−1 (50)
−1 (50)
+1 (100)
+1 (100)
+1 (100)
+1 (100)
0 (75)
−1 (50)
+1 (250)
+1 (250)
−1 (50)
+1 (250)
−1 (50)
−1 (50)
+1 (250)
0 (150)
0 (150)
0 (150)
70
85
90
88
43
64
80
79
72
72
73
70
78
91
84
48
65
86
80
72
73
72
+1 (1)
−1 (0.1)
−1 (0.1)
+1 (1)
+1 (1)
0 (0.55)
0 (0.55)
0 (0.55)
10
11
0 (50)
0 (50)
0 (75)
0 (75)
a
Measured by Lowrey–Tinsley and confirmed by GC–MS.

316
I.I. Junior et al. / Journal of Molecular Catalysis B: Enzymatic 72 (2011) 313–318
Table 3
Estimated effect for variables studied for of 1,2-O-isopropylidene glycerol stearate production.
Variable
Effect
RMIM
p-Value
TLIM
78.6625
RMIM
TLIM
Mean
80.2625
<0.0001*
0.0197*
0.0149*
0.0027*
0.0029*
0.1912
<0.0001*
0.0050*
0.0676
Curvature check
Temperature (T)
Amount of enzyme (E)
Substrate concentration (S)
Stirring (St)
−12.6917
−11.8983
7.6450
1.6150
17.8600
−17.3850
−1.8400
20.3150
−10.4650
−0.6350
<0.0001*
0.0017*
0.2880
*
Statistically significant at 95% confidence level.
Table 4
catalyzed by Lipozyme TL IM and Lipozyme RM IM. This first part
of the study required 22 experiments, eleven trials for each lipase.
The variables studied in the fractional factorial design (FFD)
were temperature (T), amount of lipase (E), substrate concentra-
tion (S) and stirring (St). These parameters were chosen based on
earlier studies of our group. The reaction time was not considered
as a variable in the present experimental design, since that a reac-
tion kinetic study was previously held to establish the best time
in which good conversion rates were achieved. The variables with
the respective levels used of fractional factorial design (FFD) for
Lipozyme TL IM and Lipozyme RM IM lipase are presented in Table 1
and the experimental design with corresponding results are shown
in Table 2.
As shown in Table 2, in most cases, Lipozyme RM IM and
Lipozyme TL IM present similar results independently of the vari-
ables studied. Only in two experiments is possible to observe
a considerable difference in conversion in between these two
enzymes (entries 2 and 7, Table 2). The best result was achieved
when using 1% (w/w) of enzyme at 40 ◦C (50 mM) and 250 rpm
which leads to 90% of conversion (entry 3, Table 2). But, it is impor-
tant to highlight that working on lower catalyst concentration
(0.1%, w/w) good conversions (85%) could also be obtained as can
be seen in Table 2, entry 2. The Lipozyme RM IM enzyme showed
high efficiency at low concentrations since the use of 0.1% (w/w) of
this enzyme has lead to good conversions.
Real and coded values (+ level, 0 intermediate, − lower level) for the independent
variables, 22.
Variables
−1, 41
−1
0
+1
+1, 41
Temperature (◦C)
Substrate concentration (mM)
50
50
51.4
53.6
55
62.5
58.5
71.3
60
75
O-isopropylidene glycerol, which is the limiting reagent in the
reaction.
Again, a stock solution containing 1,2-O-isopropylidene glyc-
erol and stearic acid in n-heptane (1:1 proportion) was prepared
in order to avoid pipetting errors. In 2 ml cryotubes, were added
1 ml of reaction medium, followed by the addition of appropriate
enzyme (1%, w/w) and the reaction analyzed every hour, until 6 h.
The results are summarized in Fig. 4a and b.
The results presented in Fig. 4 shows that with 1:1 proportion of
substrates, we still obtained good conversion rates with Lipozyme
TL IM and Lipozyme RM IM, although they were slightly lower than
the conversions presented in a 2:1 ratio (Fig. 3). However, it is still
advantageous in terms of industrial and economic development,
working with equimolar proportions of reagents, avoiding waste
and decreasing costs in the final product.
3.2. Initial evaluation of reaction parameters using fractional
factorial design (FFD)
The estimated effects and the p values are shown in Table 3
where is possible to see that temperature had a positive effect (7.64)
within the range studied. An increase in temperature causes a sig-
nificant increase in product conversion, as can be seen in Table 2,
entries 1 and 2. When the temperature goes from 40 ◦C (Table 2,
entry 1) to 60 ◦C (Table 2, entry 2), under the same concentration
of substrate (50 mM), the conversion increase from 70% to 85% for
In order to optimize the reaction conditions we have proposed
fractional factorial design (FFD) carried out in order to compare
the Lipozyme TL IM and Lipozyme RM IM as biocatalysts for the
synthesis of 1,2-O-isopropylidene glycerol stearate.
A two level factorial design was adopted for a preliminary study
designed to identify the lipase with the best performance in syn-
thesizing 1,2-O-isopropylidene glycerol stearate. First, we carried
out a fractional factorial designs 24−1 for each biocatalyst, to deter-
mine the variables of most influence for the esterification reaction
Table 5
Experimental factorial design and results of CCRD for Lipozyme RM IM.
Entry
Variable levelsa
Conversion (%)b
T (◦C)
S (mM)
1
2
3
4
5
6
7
8
9
−1 (51.4)
−1 (51.4)
+1 (58.5)
+1 (58.5)
−1.41 (50)
+1.41 (60)
0 (55)
0 (55)
0 (55)
0 (55)
0 (55)
−1 (53.6)
+1 (71.3)
−1 (53.6)
+1 (71.3)
0 (62.5)
75
69
95
68
70
86
81
69
74
74
74
0 (62.5)
−1.41 (50)
+1.41 (75)
0 (62.5)
0 (62.5)
0 (62.5)
10
11
a
Coded variables and real values.
Measured by Lowrey–Tinsley and confirmed by GC–MS.
Fig. 5. Pareto chart from CCRD showing the significance of experimental variables:
T, temperature; S, substrate concentration; L, linear effect; and Q, quadratic effect.
b

I.I. Junior et al. / Journal of Molecular Catalysis B: Enzymatic 72 (2011) 313–318
317
Table 6
Variance analysis for validation of mathematical models (ANOVA).a
Factor
Sum of squares
Degrees of freedom
Mean square
F calculated
F tabulated
p-Value
0.0011
Regression
Residuals
Lack of fit
Pure error
TOTAL
812.27
106.42
103.55
2.85
3
7
5
2
10
270.75
15.20
17.80
4.34
918.69
a
Confidence level 95%.
Lipozyme RM IM and from 70% to 78% for Lipozyme TL IM since
stirring was not significant in the process, presenting a p > 0.05
(Table 3). In Table 3 we can also observe that the amount of enzyme
presented a high estimate effect (17.86).
this amount of biocatalyst. Variables along with their coded and
uncoded values are given in Table 4.
In the CCRD, both selected variables were varied at five lev-
els resulting in 11 experiments, including four factorial points and
three central points allowing to check the curvature (Table 5). To
fit a second order model, four extra points with the same distance
from the central point were added at the matrix for this design.
The results obtained for the production of 1,2-O-isopropylidene
glycerol stearate are presented in Table 5.
The results presented in Table 5, shows that excellent conver-
sion can be obtained by the optimization of reaction conditions
affording the desired product in 95% of conversion (entry 3, Table 5).
We have also obtained the quadratic effect which shows that
variables temperature and substrate concentration and the inter-
action between each other were significant in the process. These
results are shown in Fig. 5.
Fig. 5 shows the effects for CCRD estimatives were similar to
FFD. The variables temperature, substrate concentration and the
interaction between each other were significant in the process.
Analyzing the figure a negative effect of substrate concentration in
the range studied is observed, perhaps due to the low homogene-
ity of the medium generated by increasing the concentration of the
substrate, which in turn undermines the contact of the substrate
with the active site of the enzyme. Another plausible explanation
may be that there is a small enzyme inhibition by high concen-
trations of substrate. In experiments 1, 2 and 3, 4 (Table 5) there
is a clearly negative effect of this variable, since when increas-
ing the concentration of substrates, the conversion of product also
decreases.
Thus, the enzyme Lipozyme RM IM was chosen to continue
the optimization of reaction variables in the CCRD because of
higher average conversion results presented in Table 2. Besides the
fact that amount of enzyme presented a high estimate effect, it
remained constant by 0.1% (by weight of total system) on CCRD
to enable the industrial applicability. As for the enzyme Lipozyme
RM IM was possible to obtain 85% of conversion using only 0.1%
of this biocatalyst, the process becomes cheaper than using 1% of
this enzyme to obtain 90% of the final product. The temperature
and concentration of substrate also had significance in the process
and the CCRD was further applied in order to determine their opti-
mal values maximizing the conversion rates. It was also observed
in Table 2 the need for curvature in the model by including axial
points for presenting a p < 0.05.
3.3. Central composite rotatable design (CCRD) for the production
of 1,2-O-isopropylidene glycerol stearate
Following the first factorial design mentioned previously, a cen-
tral composite rotatable design (CCRD) was employed to obtain
the optimum conditions for esterification synthesis. The study was
carried out using Lipozyme RM IM, which showed the best per-
formance in the previous experiments. The reaction parameters
involved were the temperature (T) and substrate concentration
(S). The stirring was fixed at 250 rpm according with previous
design and also the amount of enzyme was fixed at 0.1% aiming
an industrial applicability, since good results were obtained with
The experimental data have been adjusted to the proposed
model and adequacy was performed by the analysis of variance
and parameter R2 and statistical testing of the model was done by
Fig. 6. Response surface (a) and contour curves (b) for the esterification reaction of the enzyme RMIM as a function of temperature and concentration of the substrate.

318
I.I. Junior et al. / Journal of Molecular Catalysis B: Enzymatic 72 (2011) 313–318
the Fisher’s statistical test for ANOVA. Eq. (1) represents the math-
ematical model of the conversion of 1,2-O-isopropylidene glycerol
monoester in function of the variables.
conversion from initial 71% to final 95%. The mathematic model
proposed suggested a satisfactorily representation of the process
and good correlation among the experimental results and the the-
oretical values predicted by the model equation were achieved.
Y = 74.77 + 5.94T − 7.65S − 3.89T · S
(1)
Acknowledgements
where Y is the percentage yield conversion, and T and S are
the uncoded values of temperature and substrate concentration,
respectively. Statistical testing of the model was done by the
Fisher’s statistical test for ANOVA (Table 6).
Table 6 represents the analysis of variance (ANOVA) which
shows the validity of the model by F test and residue that shows
the magnitude of experimental error. The calculated F (17.80) was
higher than the tabulated F (F3,7 = 4.34), showing the validity of the
experimental model. The goodness of the model can be checked by
the determination (R2). The determination coefficient (R2 = 0.90)
implies that the sample variation of 90% for ester production is
attributed to the independent variables and can be accurately
explained by the model.
Fig. 6 shows that the decrease in substrate concentration and
temperature increase results in an optimal response. But there is a
limit to the increase in temperature which avoids denaturation of
the enzyme source.
The monostearin synthesis was accomplished by 1,2-O-
isopropylidene cleavage using boric acid as standard procedure
described over literature [18].
We thank CAPES, CNPq, FAPERJ and FINEP for financial support.
References
[1] B.D. Flickinger, P.J. Huth, Curr. Artheros Rep. 6 (2004) 468–476.
[2] J.W.E. Rush, P.S. Jantzi, K. Dupak, S.H.J. Idziak, A.G. Marangoni, Food Res. Int. 41
(2008) 1065–1071.
[3] F.A. Kummerow, Arthereos 205 (2009) 458–465.
[4] H.D. Batte, A.J. Wright, J.W. Rush, S.H.J. Idziak, A.G. Marangoni, Food Res. Int. 40
(2007) 982–988.
[5] H.D. Batte, A.J. Wright, J.W. Rush, S.H.J. Idziak, A.G. Marangoni, Food Biophys. 2
(2007) 29–37.
[6] A. Corma, S. Iborra, S. Miquel, J. Primo, J. Catal. 173 (1998) 315–321.
[7] F. Temelli, J.W. King, R.G. List, J. Am. Oil Chem. Soc. 73 (1996) 699–706.
[8] K. Holmberg, E. Osterberg, J. Am. Oil Chem. Soc. 65 (1998) 1544–1548.
[9] M. del Mar Munio, L. Esteban, A. Robles, E. Hita, M.J. Jimenez, P.A. Gonza-
lez, B. Camacho, E. Molina, Process Biochem. (Amsterdam, Neth.) 43 (2008)
1033–1039.
[10] M. Tuter, H.A. Aksoy, Biotechnol. Lett. 22 (2000) 31–34.
[11] C.C. Akoh, C. Cooper, C.V. Nwosu, J. Am. Oil Chem. Soc. 69 (1992) 257–260.
[12] M. Munio, A. Robles, L. Esteban, P.A. Gonzalez, E. Molina, Process Biochem.
(Amsterdam, Neth.) 44 (2009) 724–730.
[13] B. Cheirsilp, P. Jeamjounkhaw, A. H-Kittikun, J. Mol. Catal. B 59 (2009) 206–211.
[14] E.R. Gunawan, M. Basri, M.B.A. Rahman, A.B. Salleh, R. Rahman, Enzyme Microb.
Technol. 37 (2005) 739–744.
[15] (a) M. Heidt, U. Bornscheuer, R.D. Schmid, Biotechnol. Technol. 10 (1996) 25;
(b) A.C.O. Machado, A.A.T. da Silva, C.P. Borges, A.B.C. Simas, D.M.G. Freire, J.
Mol. Catal. B: Enzym. 69 (2011) 42.
4. Conclusion
In conclusion we have used DoE as an important tool for
reaction optimization of esterification reaction between 1,2-O-
isopropylidene glycerol and stearic acid with both Lipozyme RM IM
and Lipozyme TL IM. After initial evaluation, Lipozyme RM IM was
chosen for further development leading to an increase on reaction
[16] F. Camacho, A. Robles, P.A. González, B. Camacho, L. Esteban, E. Molina, Catal.
Appl. A. 301 (2006) 158–168.
[17] R.R. Lowry, I.J. Tinsley, J. Am. Oil. Chem. Soc. 53 (1976) 470.
[18] R. Hess, U. Bornscheuer, A. Capewell, T. Scheper, Enzyme Microb. Technol. 17
(1995) 725.