Epoxidation of oleic acid catalyzed by PSCI-Amano lipase optimized by experimental design

Article abstract of DOI:10.1016/j.molcatb.2012.03.011

The present work focuses on the oleic acid epoxide production by using PSCI Amano Lipase as biocatalyst in the reaction. An experimental design (central composite design - CCD) adopting surface response was applied to this purpose. Reactions were performed in a shaker equipment and different variables were investigated, such as temperature (25-55 °C), enzyme load (10-20 wt% of oleic acid mass), hydrogen peroxide load (0.1-0.2%) and reaction time. PSCI-Amano enzyme showed its best behavior as biocatalyst after 3 h of reaction at 55 °C, 10% enzyme load, 0.2% hydrogen peroxide and, applying 150 rpm as stirring. On these conditions, the epoxide yield was around 88%.

Full text of DOI:10.1016/j.molcatb.2012.03.011

Journal of Molecular Catalysis B: Enzymatic 81 (2012) 7–11
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Epoxidation of oleic acid catalyzed by PSCI-Amano lipase optimized by
experimental design
Flávia de Abreu Corrêaa,b, Felipe K. Sutilia,b, Leandro S.M. Mirandaa, Selma G.F. Leiteb,
Rodrigo O.M.A. De Souzaa, Ivana C.R. Leala,c,∗
a Biocatalysis and Organic Synthesis Group, Centro de Tecnologia, Bloco A, Universidade Federal do Rio de Janeiro, CEP: 21941-909 Ilha do Fundão, Rio de Janeiro, Brazil
b Escola de Química, Laboratório de Microbiologia Industrial, Bloco E, Lab-11, Universidade Federal do Rio de Janeiro, Ilha do Fundão, Rio de Janeiro, Brazil
c Curso de Farmácia, Campus Macaé, Pólo Universitário, Universidade Federal do Rio de Janeiro, CEP: 27930-560 Macaé, 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:
The present work focuses on the oleic acid epoxide production by using PSCI Amano Lipase as biocatalyst
in the reaction. An experimental design (central composite design – CCD) adopting surface response was
applied to this purpose. Reactions were performed in a shaker equipment and different variables were
investigated, such as temperature (25–55 ◦C), enzyme load (10–20 wt% of oleic acid mass), hydrogen
peroxide load (0.1–0.2%) and reaction time. PSCI-Amano enzyme showed its best behavior as biocatalyst
after 3 h of reaction at 55 ◦C, 10% enzyme load, 0.2% hydrogen peroxide and, applying 150 rpm as stirring.
On these conditions, the epoxide yield was around 88%.
Received 15 September 2011
Received in revised form 12 March 2012
Accepted 16 March 2012
Available online 28 March 2012
Keywords:
Central composite design (CCD)
Epoxidation
© 2012 Elsevier B.V. All rights reserved.
Lipase
Oleic acid
1. Introduction
The most used procedure to produce epoxidized fatty acid esters
currently consists of two steps involving alcoholysis of triglyc-
Oleochemicals are hydrocarbons derived from vegetable oils
closely related to petrochemicals and well suited for transforma-
tions by the chemical industry [1]. Vegetable oils and fatty acids
can be used in cosmetics, lubricants, chemical additives, detergents,
pharmaceuticals, polymers, and other products [2].
erides/fatty acids using KOH as catalyst followed by epoxidation
of peroxyacetic acid esters or peroxyformic (Scheme 1) [7].
The formation of byproducts occurs due to the high medium
acidity. In addition, corrosion and production of large amounts of
salts when acids are neutralized are one of the problems associated
with this type of reaction [4].
Chemo-enzymatic epoxidation reaction often provides a more
selective and environment-friendly alternative to the Prilezhaev
epoxidation process [8,9]. In the chemo-enzymatic epoxidation
reaction, the enzyme normally catalyzes the peracid formation
from the corresponding fatty acid and hydrogen peroxide [10]. Then
the peracid spontaneously transfers oxygen to the double bond
forming the epoxide (Scheme 2) [11].
Lipases or triacylglycerol hydrolases are an important group of
biotechnologically relevant enzymes with immense applications in
food, dairy, detergent and pharmaceutical industries. Lipases are
also defined as glycerol ester hydrolases that catalyze the hydroly-
sis of triglycerides into free fatty acids and glycerol. They can also
catalyze esterification, acidolysis, interesterification, alcoholysis
and aminolysis in addition to the hydrolytic activity on triglyc-
erides. Lipases are produced from microbes, specifically bacteria
and, they play a vital role in commercial ventures [12,13]. They
represent a broadly employed renewable biocatalyst in lipids trans-
formation.Therefore, this work aims at performing the epoxidation
of oleic acid using lipase as biocatalyst and the Central composite
Vegetable oils and their unsaturated fatty acids can be converted
into epoxies which are useful intermediates in organic synthesis
by participating in many reactions due to the high oxirane ring
reactivity [3]. Among their important applications there is the func-
tion as plasticizer for polyvinyl chloride (PVC). Plasticizers increase
flexibility, workability or distensibility of plastics, hence render-
ing them suitable for diverse applications [4]. One of the most
important plastics additives currently adopted is the epoxidized
soybean oil (ESBO) which has a stable market of approximately
100,000 tons/year [5]. Global demand for plasticizers is projected
to grow to 7.6 million tons per year until 2018. The main market
is the Asia-Pacific region, with China holding on to its dominating
position with 65% share [6].
∗
Corresponding author. Present address: Rua Aluisio da Silva Gomes no. 50, Uni-
versidade Federal do Rio de Janeiro, Campus Macaé, UFRJ, Pólo Universitário, Granja
dos Cavaleiros CEP: 27930-560, Macaé, Brazil. Tel.: +55 21 25627807/22 27962563.
E-mail address: ivanafarma@yahoo.com.br (I.C.R. Leal).
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2012.03.011

8
F.d.A. Corrêa et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 7–11
Acid
O
Peracid
O
Triglyceride
KOH
O
R
O
H
R
OH
O
O
OH
OH
4
4
O
Epoxystearic acid
Oleic Acid
Scheme 1. Schematic presentation of alcoholysis of triglyceride using KOH as catalyst followed by epoxidation of oleic acid by peracid.
design (CCD) in conjunction with Response Surface Methodology to
identify important variables to the process. The Amano PS-CI was
for the first time used to catalyze the epoxidation reaction by using
aqueous solution of hydrogen peroxide as oxidant.
fused-silica
capillary
columns
RTx-5MS
(30 m ×
0.25 mm × 0.25 ␮m) from Restek Corporation (USA). The col-
umn temperature was 60 ◦C for 1 min and then increased to
280 ◦C by 15 ◦C/min, and maintained for 10 min. The carrier gas
used was helium (He) and the flow was 1.1 ml/min. The injec-
tion temperature was set to 250 ◦C and the split ratio was 20.
The ion source and interface temperature were 250 and 300 ◦C,
respectively. Free fatty acids were transformed into more volatile
silylated derivatives in the presence of pyridine and N-methyl-N-
(trimethysilyl)trifluoroacetamide (MSTFA). GC–MS samples were
prepared by addition of 100 ␮L MSTFA at 100 ␮L of reaction. After
15 min, these reactants were dissolved on 2 ml ethyl acetate. 1 ␮L
of this sample was then injected into a GC–MS equipment.
2. Experimental
2.1. Lipase and materials
The enzyme used in the present study was the PSCI-Amano
Lipase from Burkholderia cepacia immobilized on ceramic and pur-
chased from Sigma–Aldrich. Hydrogen peroxide (percentage given
as 30% H2O2 (w/w) in water) and all chemicals (ethyl acetate, oleic
acid, and N-methyl-N-(trimethylsilyl)trifluoroacetamide) were of
analytical grade.
2.4. Iodine value (IV)
2.2. Enzymatic epoxidation reaction
Iodine value (IV) was obtained using the Wij’s method [14].
Aliquots of the samples (1.0 g) were weighed directly into the Erlen-
meyer and dissolved in solvent mixture (cyclohexane and glacial
acetic acid 1:1). Wij’s reagent (25 mL) was added and the flask was
capped, mixed thoroughly and kept in the dark for 1 h. At the end
of reaction time, 20 ml of 10% KI and 150 ml of distilled water were
added. The blank was prepared simultaneously on the analysis.
The mixture was titrated with 0.1 M sodium thiosulfate solution
until the yellow color almost disappears. A few drops of 0.5% starch
solution were added and titration continued until the blue color
disappears after vigorous shaking. The volume of sodium thiosul-
fate solution spent in the titration was used to calculate the iodine
value of samples. The result was calculated using the following
expression:
Oleic acid (2 mmol) was dissolved in ethyl acetate (5 ml) and
selected quantities of H2O2 were further added. The reaction was
initiated with the addition of immobilized PSCI-Amano lipase from
B. cepacia. Reactions were performed on orbital shaker (MARCONI)
at 150 rpm with controlled internal temperatures for 3 h. Subse-
quently, the immobilized biocatalysts were removed by filtration
and the solvent was distilled using a rotary evaporator. The reac-
tion products were analyzed by gas chromatography coupled to
mass spectroscopic detector (GC–MS) and calculated for its iodine
value (item 2.4). We have also performed the experiment by using
the ethyl oleate (∼85% purity) as substrate in both, the best and
worst condition found for oleic acid, in order to corroborate the
mechanism proposed (Scheme 1).
12.69 × C × (V1 − V2)
Iodine value (IV) =
2.3. Analytical procedure
m
2.3.1. GC–MS analysis
The GC–MS analyses were carried out on a Shimadzu CG-
MS2010. Chromatographic separation was carried out on
where C is the concentration, in mol/l, of sodium thiosulfate solu-
tion standard; V1 is the volume, in mL, of sodium thiosulphate used
in titration of blank; V2 is the volume, in mL, solution of sodium
H2O2
H2O
Lipase
Oleic Acid
O
Oleic Peracid
O
O
O
H
OH
R
R
O
O
7
7
OH
OH
4
4
O
Epoxystearic acid
Oleic Acid
Scheme 2. Schematic presentation of the chemo-enzymatic epoxidation of oleic acid.

F.d.A. Corrêa et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 7–11
9
Table 1
Table 3
Real and coded (+ superior level, 0 intermediate, − inferior level) values for the
Effect of parameters estimates of CCD 23 for enzyme studied.
variables evaluated in the epoxidation reactions.
Variables
Effect
p-Value
Factors
−1
0
+1
Mean
Temperature
Amount of enzyme
Peroxide concentration
Temperature × Amount of enzyme
Temperature × Peroxide concentration
Amount of enzyme × Peroxide concentration
58.00
12.75
−0.75
37.25
6.25
6.25
−7.25
<0.0001*
0.0004*
0.4552
Temperature (◦C)
25
10
0.1
40
15
55
20
0.2
Amount of enzyme (%)a
Peroxide concentration (%)
0.15
0.0004*
0.0166*
0.0166*
0.0124*
a
By weight of oleic acid.
*
Statistically significant at 95% of confidence level.
thiosulfate titration of the sample spent; m is the mass, in g, of the
sample aliquot.
It can be checked in Table 2 that the best results are among
entries 5 and 8, turning evident the positive effect of the hydro-
gen peroxide concentration variable in the epoxide conversion.
This positive effect occurs within the analyzed range 0.1–0.2%. It is
important to highlight that previous tests performed in our labora-
tory showed that peroxide concentrations upper to that inhibited
the activity of the immobilized B. cepacia lipase. Several studies
have already reported the occurrence of this increased effect on
epoxide conversion with increasing concentration of hydrogen per-
oxide in the reaction medium [19,20]. H2O2 has been reported to
oxidize methionine residues of various enzymes and, residues at
the surface of the protein seem to be most susceptible. The oxi-
dation of surface residues may not affect the enzyme activity but
side-chain to side-chain transfer reactions were already described,
transferring the initial site of oxidation to readily oxidized amino
acids [21].
Table 3 shows the estimative of the effects for the experimen-
tal design 23 and p values. Variables with p < 0.05 were significant
in the process. Only the variable concerned to the amount of
enzyme was not significant in the reaction within the range studied
(10–20%), once it has presented a p > 0.05 (0.4552). Added to that,
the lowest lipase concentration (10%) did not harm the reaction rate
once the best result could be obtained at this percentage (entry 6,
Table 2). Focusing on the industrial applicability, this result is con-
siderable relevant once it implies directly in operating costs due to
the smaller amount of biocatalyst required.
The variable temperature promoted positive effect (12.75,
Table 3). By increasing the temperature the product formation rate
is also enhanced within the range studied. It is corroborated by
entries 6 and 8 (Table 2) which achieved the highest conversions
(88 and 83, respectively). However, the continuous increase in tem-
perature, there may be a gradual inactivation of the enzyme to total
inactivation, caused by protein denaturation by heat [22].
The experimental data have been adjusted to the proposed
model and its adequacy was performed by the analysis of variance
and parameter R2. Eq. (1) represents the mathematical model of
epoxystearic acid conversion depending on the variables.
3. Results and discussion
The reaction time is not generally considered as a variable in
the case of experimental designs since it causes the greatest influ-
ence on the system and on other variables, mistaking eventual
responses. However, we recognize that it is very significant for final
conversion rates. Based on that, we followed the literature data to
select the reaction time and fixed that on 3 h [15,16].
The central composite design (CCD) consisting of three vari-
ables and varying in two levels was used to identify the important
variables for the epoxidation reaction of oleic acid catalyzed by
immobilized PSCI-Amano lipase from B. cepacia. Temperature,
amount of enzyme, and peroxide concentration were considered
critical variables (independent) and therefore assessed in planning
[17,18]. The CCD 23 was applied with triplicate central points for
calculating the experimental error. The three variables and their
real and coded levels for the enzyme evaluated are shown in Table 1.
3.1. Central composite design
The experimental designs and data analysis were carried out
using the software Statistica 6.0 (Statsoft Inc., USA), according to
the significance level established for obtaining the mathematical
model. The variance explained by the model is given by the multiple
determination coefficients, R2. The significance of regression coef-
ficients and associated probabilities, p(t), were determined by the
Student’s t-test; the model equation significance was determined
by the Fisher’s F-test.
Table 2 shows the 11 treatments considering the three variables
and the percentage yield conversion for each experiment. The first
eight treatments were used to determine the mathematical model
and refer to statistical design. Treatments 9–11 represent the trip-
licates of the central points for obtaining the experimental error.
Table 2
Experimental design and oleic acid epoxidation conversion rates in different tem-
perature, enzyme load and hydrogen peroxide load.
Y = 58.00 + 18.62H2O2 + 6.37T − 3.62E × H2O2 + 3.12T
Entriesa
Reaction
Enzyme
load (%)b
H2O2
(%)c
Conversion
(%)d
× E + 3.12T × H2O2
(1)
temperature (◦C)
where Y is the percentage yield conversion, T, E and H2O2 are
the coded values of temperature, amount of enzyme and perox-
ide concentration, respectively. Statistical testing of the model was
performed by the Fisher’s statistical test for ANOVA (Table 4).
Table 4 of analysis of variance (ANOVA) shows the model valid-
ity by F test and the residue shows the magnitude of experimental
error. The F calculated (92.41) was higher than the F tabulated
(F5,5 = 5.05), showing the experimental model validity. The good-
ness of a model can be checked by the determination coefficient
(R2). The determination coefficient (R2 = 0.98) implies that the 98%
sample variation for epoxystearic acid production is attributed to
the independent variables and can be accurately explained by the
model.
1
2
3
4
5
6
7
8
9
−1
−1
−1
37
34
34
50
72
88
61
83
59
59
61
+1
−1
+1
−1
+1
−1
+1
0
−1
+1
+1
−1
−1
+1
+1
0
−1
−1
−1
+1
+1
+1
+1
0
10
11
0
0
0
0
0
0
a
Numbers were run in random order.
b
c
Enzyme load (%, relative to the weight of oleic acid).
Hydrogen peroxide load (%, relative to the mol).
Analyzed by GC–MS.
d

10
F.d.A. Corrêa et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 7–11
Table 4
Variance analysis for validation of mathematical models (ANOVA).*
Factor
Sum of squares
Degrees of freedom
Mean square
F calculated
F tabulated
p-Value
Regression
Residuals
Lack of fit
Pure error
Total
3361.63
36.38
33.70
2.66
5
5
3
2
10
672.32
7.27
–
–
–
92.41
5.05
–
–
–
–
6.37E−5
–
–
–
–
–
–
–
–
3398.0
*
Confidence level 95%.
Fig. 1. (a and b) Response surface (a) and counter curve (b) for the epoxidation reaction catalyzed by the immobilized Burkholderia cepacia lipase. IM in function of the
temperature and peroxide concentration.
Fig. 1a and b shows that increasing the concentration of hydro-
gen peroxide and increasing the temperature results in optimal
response, around 90%. Above the rated temperature (55 ◦C) it is sug-
gested that it can lead to enzyme denaturation. Likewise, increasing
the peroxide concentration above 0.2% the lipase activity seems to
be inhibited.
The interaction between reaction temperature and peroxide
concentration was evaluated by Sun et al. [17]. It was observed
that the combination of high substrate content and mild temper-
atures (35–50 ◦C) led to increased formation of epoxides (>4.5%).
However, once the substrate ratio is decreased and temperature
increased, the formation of epoxide does not increase. This is prob-
ably due to the lack of enough H2O2 to perform the epoxidation of
3.2. Iodine values
The double bound conversion, as related to the iodine values,
correspondent to the entries 1–11 are given in Table 5. As can
be seen, the iodine value (IV) is in accordance with the conver-
sion obtained, once it has shown smaller values in reactions which
furnished higher product conversions (entries 6 and 8). Contrary,
the entries 1–3 presented the minor conversion rates, and, hence
higher iodine values (IV) were attributed. The iodine values as
related to the initial IV (IVo) are also listed below in Table 5.
4. Conclusion
C
C bond [17].
We worked on different parameters which found to be impor-
tant for the PSCI Amano lipase activity. The enzyme showed its best
behavior as biocatalyst for epoxidation reaction after 3 h at 55 ◦C,
10% enzyme load, 0.2% hydrogen peroxide and, applying 150 rpm
as stirring. On these conditions, the epoxide yield was around 88%.
An experimental design (central composite design – CCD) adopting
surface response was applied to measure the variables interference
and, finally, the mathematical model was satisfactorily validated.
We could also demonstrate by the ethyl oleate experiment that, on
the conditions adopted, the carboxyl group seems to be exhibiting a
very important role for the obtainment of high epoxide conversion
rates.
We finally analyzed the conversion of ethyl oleate epoxide
considering the best (55 ◦C; 3 h; 10% E; 0.2 H2O2) and the worst
variables (25 ◦C; 3 h; 10% E; 0.1 H2O2) proposed by the math-
ematic model for oleic acid peroxide. At these two different
conditions the ethyl oleate epoxide was detected, but merely in
minor amounts (≤10%), which can be justified by the impurity of the
substrate.
Table 5
Calculated iodine values for the reaction products.
Entry
aIV
bIV(x)
Conversion (%)
Acknowledgements
1
2
3
4
5
6
7
8
9
3.52
3.03
2.90
2.43
1.50
0.62
1.85
0.74
2.39
2.10
2.00
0.30
0.40
0.42
0.52
0.70
0.88
0.63
0.85
0.53
0.58
0.60
37
34
34
50
72
88
61
83
59
59
61
This study was supported by grants from Fundac¸ ão Carlos Cha-
gas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ),
Conselho Nacional de Desenvolvimento Científico e Tecnológico
(CNPq), Coordenac¸ ão de Aperfeic¸ oamento Pessoal de Nível Superior
(CAPES).
References
10
11
[1] R. Hatti-Kaul, U. Törnvall, L. Gustafsson, P. Börjesson, Trends Biotechnol. 25
(2007) 119–124.
[2] C.T. Hou, Nat. Biotechnol. 26 (2009) 2–10.
a
IV: Iodine value by the Wij’s method (EN 14111:2003).
IV(x) = [(IVo − IV)/IVo]; IVo: initial iodine value.
b

F.d.A. Corrêa et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 7–11
11
[3] C. Orellana-Coca, D. Adlercreutz, M.M. Andersson, B. Mattiasson, R. Hatti-Kaul,
[12] R. Gupta, N. Gupta, P. Rathi, Appl. Microbiol. Biotechnol. 64 (2004) 763–781.
Chem. Phys. Lipids 135 (2005) 189–199.
[4] C. Orellana-Coca, J.M. Billakanti, B. Mattiasson, R. Hatti-Kaul, J. Mol. Catal. B:
Enzym. 44 (2007) 133–137.
[5] M.R. Klaas, S. Warwel, Ind. Crop Prod. 9 (1999) 125–132.
[6] Ceresana Research. Market Study: Plasticizers (UC-2205), vol. 1, 2011, 346 pp.
(Chapter 1).
[7] L.P. Witnauer, H.B. Knight, W.E. Palm, R.E. Koos, W.C. Ault, D. Swern, Ind. Eng.
Chem. 47 (1955) 2304–2311.
[13] B. Joseph, P.W. Ramteke, G. Thomas, Biotechnol. Adv. 26 (2008) 457–470.
[14] EN 14111 Method, BSI British Standards, 2003.
[15] L.A. Lerin, M.C. Feiten, A. Richetti, G. Toniazzo, H. Treichel, M.A. Mazutti, J.V.
Oliveira, E.G. Oestreicher, D. Oliveira, Ultrason. Sonochem. (2011).
[16] M.I. Rodrigues, A.F. Iemma, Planejamento de Experimentos e Otimizac¸ ão de
Processos, second edition, Casa do Pão Publishing, 2005, p. 342.
[17] S. Sun, X. Ke, L. Cui, G. Yang, Y. Bi, F. Song, X. Xu, Ind. Crop Prod. 33 (2011)
676–682.
[8] W.S.D. Silva, A.A.M. Lapis, P.A.Z. Suarez, B.A.D. Neto, J. Mol. Catal. B: Enzym. 68
(2011) 98–103.
[9] M. Guidotti, E. Gavrilova, A. Galarneau, B. Coq, R. Psaro, N. Ravasio, Green Chem.
13 (2011) 1806–1811.
[10] A.J. Kotlewska, F. Van Rantwijk, R.A. Sheldon, I.W.C.E. Arends, Green Chem. 13
(2011) 2154–2160.
[11] U. Törnvall, C. Orellana-Coca, R. Hatti-Kaul, D. Adlercreutz, Enzyme Microb.
Technol. 40 (2007) 447–451.
[18] R. García, M. Martínez, J. Aracil, J. Ind. Microbiol. Biotechnol. 28 (2002) 173–179.
[19] V.V. Goud, A.V. Patwardhan, N.C. Pradhan, Bioresour. Technol. 97 (2006)
1365–1371.
[20] C. Orellana-Coca, S. Camocho, D. Adlercreutz, B. Mattiasson, R. Hatti-Kaul, Eur.
J. Lipid Sci. Technol. 107 (2005) 864–870.
[21] M.J. Davies, Biochim. Biophys. Acta 1703 (2005) 93–109.
[22] A.O. Triantafyllou, E. Wehtje, P. Adlercreutz, B. Mattiasson, Biotechnol. Bioeng.
54 (1997) 67–76.