Enzymatic Esterification of Oleic Acid with Aliphatic Alcohols for the Biodiesel Production by Candida antarctica Lipase

Article abstract of DOI:10.1007/s10562-013-1044-0

Biodiesel can be obtained by esterification reactions of free fatty acids with enzymatic catalysts (lipases). In this study, the immobilized Candida antarctica lipase was employed in enzymatic esterifications of oleic acid with aliphatic alcohols (methanol, ethanol, n-propanol, n-butanol). Some features that influence the enzymatic esterification reaction, such as amount of biocatalysts, reaction time, hydration level and biocatalyst turnover were evaluated. The products were determined by GC-FID and ¹H NMR analyses and these analytical methods were compared. The enzymatic catalyst (C. antarctica lipase) was efficient providing high yields of biodiesel (above 90 %) in less than 24 h to ethanol, n-propanol and n-butanol, whereas for methanol, the enzyme was inactive after ten cycles of reaction. Two new quantitative easy methods were also developed to quantify esters produced by ¹H NMR based on the α-CH₂ protons of oleic acid and esters. The quantification method used in the enzymatic reactions by ¹H NMR showed effective with small differences in comparison with GC-FID analyses. Graphical Abstract: C. antarctica lipase was employed in enzymatic esterifications of oleic acid with aliphatic alcohols (methanol, ethanol, n-propanol, n-butanol) for the biodiesel production. A new method was developed to quantify esters produced (biodiesel) by ¹H NMR based on the α-CH2 protons. [Figure not available: see fulltext.]

Full text of DOI:10.1007/s10562-013-1044-0

Catal Lett
DOI 10.1007/s10562-013-1044-0
Enzymatic Esterification of Oleic Acid with Aliphatic Alcohols
for the Biodiesel Production by Candida antarctica Lipase
Isac G. Rosset
Elisabete M. Assaf
•
Maria Cec ´ı lia H. T. Cavalheiro
Andr e´ L. M. Porto
•
•
Received: 19 March 2013 / Accepted: 15 June 2013
Ó Springer Science+Business Media New York 2013
Abstract Biodiesel can be obtained by esterification
reactions of free fatty acids with enzymatic catalysts
1 Introduction
(
lipases). In this study, the immobilized Candida antarc-
The utilization of biodiesel in diesel motors is very
attractive, in view of the environmental aspects and be a
renewable energy source. It can be obtained by either the
triacylglyceride transesterification present in vegetable oils
and animals fats or esterification of free fatty acids [1].
Alkaline catalysis is most widely used in industrial
processes of transesterification and sodium hydroxide is the
most utilized catalyst because it is cheaper than enzymes or
heterogeneous catalysts. However, in many cases, this type
of catalysis results in saponification reactions, consuming
the catalyst, decreasing the efficiency of the process and
hampering the biodiesel purification and glycerol separa-
tion [2]. Acid catalysts are usually utilized when the con-
tent of fatty acids is high, but they have showed some
inconveniences, like corrosiveness, fewer yields than
alkaline catalysis, slow reactions and need of high tem-
peratures [3, 4].
tica lipase was employed in enzymatic esterifications of
oleic acid with aliphatic alcohols (methanol, ethanol,
n-propanol, n-butanol). Some features that influence the
enzymatic esterification reaction, such as amount of bio-
catalysts, reaction time, hydration level and biocatalyst
turnover were evaluated. The products were determined by
1
GC-FID and H NMR analyses and these analytical
methods were compared. The enzymatic catalyst (C. ant-
arctica lipase) was efficient providing high yields of
biodiesel (above 90 %) in less than 24 h to ethanol,
n-propanol and n-butanol, whereas for methanol, the
enzyme was inactive after ten cycles of reaction. Two new
quantitative easy methods were also developed to quantify
1
esters produced by H NMR based on the a-CH protons of
oleic acid and esters. The quantification method used in the
1
enzymatic reactions by H NMR showed effective with
2
small differences in comparison with GC-FID analyses.
An alternative to these processes is the use of enzymes,
more exactly, lipases for the biodiesel production. There
has been a huge interest in the utilization of lipases as
biocatalysts to convert vegetable oil and animal fatty into
biodiesel. Their utilization for the biodiesel production is
advantageous, as different solvents (hydrophilic and
hydrophobic) can be utilized, the enzymes are versatile,
robust and many lipases show considerable activity to
catalyze not only transesterification reactions of oil and fats
Keywords Biocatalysis ꢀ Hydrolases ꢀ CALB ꢀ
NMR quantification
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-013-1044-0) contains supplementary
material, which is available to authorized users.
[
5], but also fatty acid esterification, which is not possible
when using conventional alkaline catalysts (Fig. 1), if the
enzyme is immobilized and does not lose activity, it can be
reused [6–11]; relative thermostability and tolerance to
short chain alcohols [12–16].
I. G. Rosset ꢀ M. C. H. T. Cavalheiro ꢀ E. M. Assaf ꢀ
A. L. M. Porto (&)
Instituto de Qu ´ı mica de S a˜ o Carlos, Universidade de S a˜ o Paulo,
Av. Jo a˜ o Dagnone, 1100, Santa Angelina, CP 780, S a˜ o Carlos,
SP CEP 13563-120, Brazil
The main advantages in the biodiesel production are that
the enzyme shows high chemoselectivity and mild conditions
e-mail: almporto@iqsc.usp.br
123

I. G. Rosset et al.
O
O
O
O
O
O
2 Experimental
.1 Materials
OH
OH
OH
O
O
O
R1
R2
R3
RO
RO
RO
R1
R2
R3
2
bio(catalyst)
bio(catalyst)
+
+
3 R OH
+
(
(
1)
2)
Methanol, ethanol, n-propanol and n-butanol (HPLC grade)
were purchased from Tedia (USA), and n-hexane and ethyl
acetate (PA grade) were obtained from Quimis (Brazil) and
used without further purification. Immobilized lipase type B
O
O
Ò
from C. antarctica (CALB; Novozym 435 ) was kindly
HO R1
+
H O
2
R
OH
R
OR1
donated by Novo Nordisk Bioindustrial Ltda (Arauc a´ ria, PR,
Brazil). The GC-FID analysis was carried out in a model GC
2010ShimadzugaschromatographequippedwithanAOC20i
auto injector, a FID and a SelectTM Biodiesel for Glycerides
UltiMetal column (15 m 9 0.32 mm). The chromatographic
conditions employed were carrier gas nitrogen at 60 kPa, split
ratio = 1:20, injection volume = 1.0 lL, injector tempera-
ture = 270 °C, detector temperature = 350 °C, oven tem-
perature program initially at 100 °C and increased at 15 °C/
min to 320 °C, remaining at 320 °C for 6.8 min and total time
of analysis = 23.5 min. Tricaprine (Aldrich) was used as
internal standard. A Brucker (Germany) AC200 spectrometer
Fig. 1 (1) Transesterification and (2) esterification of triacylglyce-
rides and fatty acids
(consumes less energy in the process-temperatures around
35 °C) [17]. As triglycerides and fatty acids are natural
substrates for lipases, and these accept a structural diversity
of substrates, which is advantageous because it allows the use
of different sources of raw materials. Free fatty acids are the
key parameters to determine the viability of the transesteri-
fication process of vegetable oil into biodiesel. They are
industrially obtained from triacylglyceride hydrolysis and
when oil has higher acidity ([1.0 %), it hampers the effi-
ciency of the biodiesel conversion if this oil is used for
transesterification [18]. Therefore, the researches on the
enzymatic processes of fatty acids are of great importance to
the optimization of biodiesel production employing
biocatalysts.
1
13
was used for H and C NMR spectroscopy, operating at 200
and 50 MHz, respectively, with TMS as the internal standard.
The NMR experiments were conducted and the chemical
shifts (d) were expressed in ppm and the coupling constants
(J) in Hz. The multiplicities were presented as s (singlet), d
(duplet), t (triplet), q (quartet) and m (multiplet). The orbital
shaker used was Tecnal (Brazil) model TE-421. The reaction
was monitored by TLC employing aluminum-backed pre-
coated silica gel 60 F254 layers (Whatman, UK) with a 98:02
n-hexane:ethyl acetate like eluent.
Residual fatty acids are undesirable in biodiesel, i.e., the
transesterification and esterification processes must be
complete so that these components do not cause any
operational damage, like deposits in the engine, filter
obstruction or fuel deterioration [19]. Therefore, the quality
control of biodiesel is very important for its commerciali-
zation [20]. Gas Chromatography (GC) is the main method
to quantify esters of fatty acids produced, but the Nuclear
Magnetic Resonance (NMR) has come to prominence,
mainly because this technique does not require sample
derivatization [21]. Several studies have shown the quan-
tification of biodiesel is possible by employing NMR
analysis [22]. Recently, we published a study showing the
efficiency of NMR in the biodiesel quantification from
soybean oil with enzymatic catalysis [23]. However, the
literature lacks studies of the NMR quantification of
products generated by esterification reactions from fatty
acids.
2.2 Standards of Esters
First, the esters of oleic acids were prepared by acid catalysis
1
to be employed as standards in GC-FID and H NMR. The
oleic acid (1 eq., 9.51 mmol, 3.0 mL), alcohol (6.0 mL),
20 mL of toluene and concentrated sulfuric acid (1.0 % w/w
in relation to oleic acid) were added in a two-neck flask with
a Dean–Stark system. The system remained under reflux for
15 h and the reaction was monitored by TLC. The mixture
was then diluted with a saturated solution of NaHCO3
(3 9 20 mL) and the remaining alcohol was removed by
low pressure. The products were dried with anhydrous
MgSO and purified by flash chromatography with silica gel
4
as the stationary phase and n-hexane:ethyl acetate (98:02) as
1
the efluent. The products were characterized by IR, H and
In this study, we performed the enzymatic esterification
of oleic acid with Candida antarctica lipase with several
aliphatic alcohols. The quantification by GC-Flame Ioni-
zation Detector (GC-FID) and the development of two
1
3
C NMR analyses.
2.3 Enzymatic Preparation of Esters from Oleic Acid
1
simple methodologies by H NMR to quantify the esteri-
fication reactions for the biodiesel production were also
developed.
For the enzymatic esterification reactions, the following
parameters were evaluated: amount of biocatalyst, reaction
1
23

Enzymatic Esterification of Oleic Acid
a
b
A
B
c
1
Fig. 2 Example of a signal overlap on the H NMR spectra. a a-CH
ethyl oleate and oleic acid (50:50). Expansions were between 2.20–2.40 ppm
2
protons of ethyl oleate; b a-CH
2
protons of oleic acid; and c mixture of
Fig. 3 Triplet for the deployment of resonance signals according to the n?1 rule
Fig. 4 Methods of
quantification of biodiesel by
H NMR employed in the
esterification reaction of oleic
a
b
1
acid. a Equation for direct
quantification; b use of
0
analytical curve (A
= a ? b, A
1
= a ,
= a ? b ? a )
0
A
2
3
123

I. G. Rosset et al.
time, hydration level and enzyme turnover. All experi-
ments were run in Erlenmeyer flasks on an orbital shaker
experiments was 5.0 %. After the reactions, the products
were extracted with n-hexane, evaporated and analyzed by
1
GC-FID and H NMR.
(130 rpm) at 32 °C. After the reactions, the immobilized
enzyme was filtered in a qualitative filter paper and the
products were washed with distilled water (1 9 10 mL),
2.3.3 Hydration Level
concentrated by low pressure, dried with MgSO and
4
1
analyzed by GC and H NMR.
The presence of water in the reaction medium was evalu-
ated. In this experiment, several mixtures of water and
alcohols were prepared: 0.5, 1.0, 1.5, 2.0 and 4.0 % in
relation to the volume of alcohols. The reactions were
conducted in Erlenmeyer flasks on a shaker (130 rpm), at
32 °C for 24 h employing 5.0 % of enzyme. After the
reactions, the products were extracted and analyzed by
2
.3.1 Amount of Biocatalysts
To evaluate the amount of catalyst in the reaction, oleic
acid (1 eq., 17.7 mmol, 0.5 g), the respective alcohol
(
methanol, ethanol, n-propanol and n-butanol, 1.5 mL) and
1
the enzyme in several proportions (0.1, 0.25, 0.5, 1.0, 2.5
and 5.0 % w/w in relation to the oleic acid) were added in
Erlenmeyer flasks. After 24 h, the reactions were extracted
with n-hexane, evaporated and analyzed by GC-FID and
GC-FID and H NMR.
2.3.4 Turnover of the Enzyme
1
H NMR.
Successive reactions with the same immobilized enzyme
were carried out to evaluate the turnover capacity and
stability of the biocatalyst. The enzyme was utilized ten
times for each alcohol (methanol, ethanol, n-propanol and
n-butanol) and the time of witch reaction was 24 h. The
enzyme was filtered, washed with n-hexane and used again.
After each reaction, the enzyme was filtered in qualitative
2
.3.2 Reaction Time
The reactions were conducted in Erlenmeyer flasks on
orbital shaker (130 rpm) at 32 °C at different reaction
times (1, 4, 8, 15 and 24 h). The amount of enzyme in these
ácido oléico padrão.esp
metil 25%.esp
metil 50%.esp
metil 75%.esp
metil oleato padrão.esp
2
.54
0.14
2.25
2.39
2.36
0.32
2.27
2.16
0.40
2.27
2.03
0.66
2.27
1.68
0.74
2.27
2
.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.35
2.34Che 2m . 3i c 3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.26
2.25
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.26
2.25
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.26
2.25
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.26
2.25
2.24
2.23
2.22
2.21
ácido oléico padrão.esp
etil 25%.esp
etil 50%.esp
etil 75%.esp
etil oleato padrão.esp
2
.54
0.14
2.25
2.29
2.34Che m2 . i3
0.25
2.25
2.08
2.34Che m2 . i3
0.34
2.25
1.87
2.34Che m2 . i3
0.52
2.25
1.60
2.34Che 2m . 3i c
0.76
2.25
2
.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che m2 . i3
cal Sh
3
2
if.t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
cal Sh
3
2
if.t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
cal Sh
3
2
if.t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
cal Sh
3
2
if.t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
3
al S h2 i f. 3t (2 ppm2
)
.31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
ácido oléico padrão.esp
propil 25%.esp
propil 50%.esp
propil 75%.esp
propil oleato padrão.esp
2
.54
0.14
2.25
2.19
2.34Che m2 . i3
0.16
2.26
2.12
0.37
2.26
2.05
0.44
2.26
1.74
0.81
2.26
2
.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che m2 . i3
cal Sh
3
2
if.t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
cal Sh
3
2
if.t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.25
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che m2 . i3
cal Sh
3
2
if
.
t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.25
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3
al Sh
2
if
.
t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.25
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3
al Sh
2
if.t
3
(
2
ppm)2.31
2.30
2.29
2.28
2.27
2.25
2.24
2.23
2.22
2.21
ácido oléico padrão.esp
butil 25%.esp
butil 50%.esp
butil 75%.esp
butil oleato padrão.esp
2
.54
0.14
2.25
2.19
0.14
2.26 2.25
2.18
0.35
2.26 2.25
1.93
0.50
2.25
1.54
0.65
2.25
2
.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che m2 . 3i c
3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.27
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che m2 . 3i c
3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.27
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3a l S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
2.44
2.43
2.42
2.41
2.40
2.39
2.38
2.37
2.36
2.35
2.34Che 2m . 3i c
3
al S h2 i f. t3 (2 ppm 2) .31
2.30
2.29
2.28
2.27
2.26
2.24
2.23
2.22
2.21
Fig. 5 Overlap for all esters produced and oleic acid in different concentrations (method B)
1
23

Enzymatic Esterification of Oleic Acid
a
b
7
Y = A + BX
A = -0,84329
B = 0,71582
R = 0,99917
Y = A + BX
A = 0,35481
B = 1,40674
R = 0,99978
1
0
8
6
4
2
0
6
5
4
3
2
1
0
-1
0
2
4
6
8
10
0
1
2
3
4
5
6
7
analyte concentration / P.I. concentration
analyte concentration / P.I. concentration
c 10
d 12
Y = A + BX
A = 0,3589
B = 0,81044
R = 0,99844
Y = A + BX
A = -0,91014
B = 1,16686
R = 0,99828
1
0
8
6
4
2
0
8
6
4
2
0
0
2
4
6
8
10
12
0
2
4
6
8
10
analyte concentration / P.I. concentration
analyte concentration / P.I. concentration
Fig. 6 Analytical curves obtained by GC-FID for the quantification of enzymatic reactions. a MeOH; b EtOH; c n-PrOH; d n-BuOH
1
2.5 H NMR Quantification
filter paper, washed with 20 mL of n-hexane P.A., dried at
2
5 °C for 1 h and reutilized. After the reactions, the
products were extracted and analyzed by GC-FID and
1
The quantification by H NMR was based on areas of
1
H NMR.
a-CH
respective esters produced (methyl: 2.30 ppm; ethyl: 2.28
2
protons of oleic acid (2.34 ppm) and a-CH of the
2
ppm; n-propyl: 2.29 ppm and n-butyl: 2.29 ppm). Due to the
2
.4 GC-FID Quantification
ester formation, the signal of a-CH protons (triplet) is more
2
shielded when compared with the acid (2.34 ppm). However,
this difference is small, resulting in a partial overlap these
signals (a-CH of acid with the a-CH of esters) when there
Four analytical curves were prepared for the quantification
of the reaction catalyzed with lipase CALB. Mixtures of
analyte (ester) and internal standard (tricaprin) were pre-
pared and analyzed by GC-FID. For quantification, the
areas of the analyte (respective ester) and internal standard
were obtained by GC-FID and plotted with the concen-
tration of these compounds. The linear regressions of the
curves were used for the quantification of the enzymatic
reactions. The GC-FID is the most employed method to
quantify these reactions, therefore it was used in this study
2
2
exists a mixture of these compounds (Fig. 2).
Through the phenomenon called spin–spin deployment,
the NMR signals undergo deployment, which can be
empirically explained by the n ? 1 rule. According to this
rule, which proton ‘‘feel’’ the equivalents number of pro-
tons (n) of the neighbor carbon atom that the proton is
connected. Therefore, the resonance signal ‘‘unfold’’ on
n?1 components. These deployments respect the Pascal
triangle. The quantification by NMR of the signals of the
1
to compare the results with the H NMR quantification.
123

I. G. Rosset et al.
a 100
b
Y = A + BX
A = 132.0323
Y = A + BX
A = 134.9083
100
B = 46.8817
R = 0.9926
B = 47.4422
R = 0.9974
80
80
6
0
60
40
40
20
20
0
0
-3,0
-2,5
-2,0
-1,5
-1,0
-0,5
-3,0
-2,5
-2,0
-1,5
-1,0
-0,5
ln (A /A )
ln(A /A )
1
2
1
2
c
d
Y = A + BX
Y = A + BX
A = 140.9043
B = 48.5452
R = 0.9993
1
00
100
80
60
40
20
0
A = 137.2453
B = 46.3542
R = 0.9979
8
0
6
0
4
0
2
0
0
-3,0
-2,5
-2,0
-1,5
-1,0
-0,5
-3,0
-2,5
-2,0
-1,5
-1,0
-0,5
ln(A /A )
ln(A /A )
1 2
1
2
1
Fig. 7 Analytical curves obtained by H NMR analysis (method B) for the quantification of enzymatic reactions. a MeOH; b EtOH; c n-PrOH;
d n-BuOH
a-CH atoms of esters was based on the triplet formed,
This external signal was chosen because it does not suffer
any overlap from the signals of oleic acid. In method B,
mixtures of oleic acid and respective ester were prepared in
different proportions (0, 25, 50, 70 and 100 %) and ana-
2
whose total area is the sum of external components of the
0
triplet (smaller intensities, a and a ), which have an area
equal to 2/4 (1/4 for each external signal) and of total area
of internal component (bigger intensity, b) that have a total
area of 2/4 in relation the total area of triplet (Fig. 3).
1
lyzed by H NMR. The relation between the areas (A /A ,
2
1
0
a /a?b) was conducted and it was possible to obtain the
analytical curves and use them to calculate the yield of the
enzymatic reactions. The overlap of the signals of all esters
produced and the oleic acid are showed in Fig. 5.
1
Based on the deployment of the signals, the H NMR
quantification was conducted in two forms; method A:
through a simple equation it was possible to calculate
directly the amount of ester formed through of the external
signal on the right side of the ester triplet; method B: an
analytical curve obtained by ration of areas of triplet
components (Fig. 4).
1
The H NMR quantification employing two methods
was conducted only for the first experiment (amount of
catalyst), to compare the analytical methods, because in
these experiments, the yields varied of 1.0–97 %. To verify
these methodologies by NMR, the GC-FID analysis was
realized.
In method A, the numerator of the equation is multiplied
0
by four as it represents 1/4 of the total area of the signal (a ).
1
23

Enzymatic Esterification of Oleic Acid
a 100
b
100
9
8
7
0
0
0
80
60
40
20
0
60
50
40
30
MeOH
EtOH
PropOH
ButOH
MeOH
EtOH
PropOH
ButOH
20
10
0
0
1
2
3
4
5
0
5
10
15
20
25
Amount of catalyst (%)
Time (h)
c
100
d
100
80
60
40
20
0
80
60
40
20
0
MeOH
EtOH
PropOH
ButOH
MeOH
EtOH
PropOH
ButOH
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
0
2
4
6
8
10
Amount of water (%)
numbers of cycles
Fig. 8 Results of esterification reactions of oleic acid (quantified by GC-FID): a amount of catalyst, b reaction time c hydration level d turnover
of the enzyme
3
Results and Discussion
the amount of catalyst employing the esterification reaction
of oleic acid, it was possible to observe that the larger the
amount of biocatalyst used, the higher the yield of reaction
and was used until 5.0 % of catalyst (Fig. 8a). When
methanol was used in the reaction, the behavior of the
curve was very different from that of other alcohols. The
results are showed in Table 1.
3
.1 Analytical Curves
The GC-FID analysis provided four analytical curves by
the Microcal Origin 6.0 program, with the respectives
linear regressions, showing a good correlation among the
points (Fig. 6). The curves were used for all quantifications
of enzymatic reactions and the results were compared with
It is possible to observe that was used ethanol and
n-propanol on the esterification, 1.0 % of enzyme was
already enough to increase of yield (93.1 and 91.1 %, entry
4). A recent study demonstrated that 2.24 % of C. ant-
1
those of the H NMR analysis.
Four analytical curves for four alcohols were obtained
1
employing method B of quantification by H NMR (Fig. 7)
Ò
arctica lipase (Novozyme 435 ) under similar reaction
with a good correlation among the points.
conditions provided an yield of 95 % with ethanol before
24 h [24]. The n-butanol also showed a good yield
3
.2 Effect of the Amount of Catalyst
(91.0 %, entry 6) with 5.0 % of enzyme. A different
behavior in relation to that of other alcohols when metha-
nol was used and were necessary 5.0 % of enzyme
(90.8 %, entry 6).
The discussion below was based on GC-FID quantification,
as it is the standard quantification method. In the study of
123

I. G. Rosset et al.
3
.3 Effect of Time Reaction
Table 1 Yields of the reactions of oleic acid esterification with
various alcohols in the studied parameters
a
The transesterification reaction time also varies greatly
from a low of 8 h for immobilized lipase from Pseudo-
monas cepacia with Jatropha oil in ethanol to a high of
Parameter
Yield (% of oleic ester)
Amount of catalyst (%) MeOH EtOH n-PrOH n-BuOH
9
0 h for the same free enzyme transesterifying soybean oil
with methanol [25]. A space-time of between 127 and
1 min using Candida sp. immobilized on textile with
Entry
1
2
3
4
5
6
0.1
0.25
0.5
1.0
2.5
5.0
1.0
2.1
46.3
67.0
82.4
93.1
95.1
96.5
40.6
62.8
79.6
91.1
94.0
95.0
17.8
34.0
54.4
74.1
89.7
91.0
5
petroleum ether and 10 % water as co-solvents and con-
stant enzyme activity during 500 h of operation. A con-
version of 90 % with a space–time of 2.7 h using
2.5
6.3
21.9
90.8
Ò
Novozyme 435 in a solvent-free system producing fatty
acid methyl ester (FAME) from waste oil. The enzymes
kept their activity for 100 days when three PBRs in series
where used with intermediate glycerol removal and meth-
anol addition [26].
Reaction time (h)
7
8
9
1
1
1
4
31.5
46.7
72.1
80.2
90.8
35.2
52.4
78.9
92.4
96.5
30.1
79.5
91.7
92.0
95.1
24.2
60.1
78.3
89.9
91.0
8
The results showed that the reaction time ranging of
0
1
15
24
1
–24 h, just the n-propanol provided a yield of 91.7 %
after 8 h of reaction (entry 9), what show that the enzyme
has good tolerance to n-propanol. This is the time that can
be equivalent to acid catalysis [27]. For other alcohols
Water (%)
1
1
1
1
1
2
3
4
5
6
0.5
1.0
1.5
2.0
4.0
98.4
98.5
98.1
98.0
97.1
95.4
95.7
95.8
95.2
95.1
94.3
94.0
93.5
93.2
91.0
92.1
91.6
91.2
90.4
88.3
(
ethanol and n-butanol), 15 h were necessary for the yield
to achieve 90 % (92.4 and 89.9 %, entry 10). Just to
methanol needed 24 h to obtained a satisfactory yield
(
90.8 %, entry 11) (Fig. 8b).
Turnover (cycles)
1
1
1
2
2
2
2
2
2
2
7
8
9
0
1
2
3
4
5
6
1
2
90.8
73.2
45.4
34.6
14.7
6.4
96.5
96.3
96.9
96.8
95.6
95.7
95.3
95.7
95.0
95.8
95.0
95.2
94.8
94.9
94.6
94.4
94.7
94.3
94.6
94.2
91.0
91.5
91.2
91.3
90.9
90.8
91.0
90.7
90.4
90.1
3
.4 Hydration Level
3
Another fundamental factor in this type of reaction is the
amount of water in the alcohol, i.e., the hydration level.
The biocatalyst frequently requires a certain amount of
water in the reaction to maintain activity [28]. This is an
important factor in transesterification and esterification
when enzymes are used. Water is essential to maintain the
specify tridimensional structure of the enzyme and the total
removal may lead to irreversible changes in the structure
4
5
6
7
2.3
8
1.3
9
0.2
10
0.09
a
[
29]. The enzyme activation involves the unmasking and
All yields were calculated by GC-FID
restructuration of its active site through conformational
changes of lipase, which require the presence of an oil–
water interface. The lipase activity usually depends on the
interfacial area available. An increase in the water addition
increase the amount of water available to form oil–water
droplets, increasing the interfacial area available. The good
water contends is the accord between the minimum
hydrolysis and the maximum activity of enzyme on the
reaction [30].
of enzyme and dehydrating the enzyme, this way doing
with lose the activity by modification of structure. In the
presence of water, this effect is filled, increasing its effi-
ciency. The reaction yield was followed according to the
chain length alcohols (methanol until 98.4–92.1 % for
n-butanol, entry 12), but with the increase of water amount,
the yield was decreasing, showing that a little amount of
water is needed to offer a good yield (Fig. 8c).
The results of our study suggest that the presence of
water directly affects the yield of reaction with methanol.
When anhydrous methanol was used, the yield was 90.8 %
with 5.0 % of enzyme (entry 6). However, the 0.5 %
increase in water did the meant that the income increase in
3.5 Reuse of the Enzyme
7
.6 % (98.4 %, entry 12). This may be due to the absence
Excellent results were obtained for all alcohols except
methanol. For this alcohol, at the end of the 10 reaction
of water, the methanol can be removing the structural water
1
23

Enzymatic Esterification of Oleic Acid
1
Table 2 Comparison of the quantification methods employed (GC-FID and H NMR)
a
Catalyst (%)
Yields
MeOH
CG
EtOH
GC
n-PrOH
n-BuOH
NMR A
NMR B
NMR A
NMR B
GC
NMRA
NMR B
GC
NMR A
NMR B
0
0
0
1
2
5
.1
1.0
2.1
–
–
–
–
46.3
67.0
82.4
93.1
95.1
96.5
38
61
75
88
89
90
45
68
84
94
94
96
40.6
62.8
79.6
91.1
94.0
95.0
29
50
70
83
87
87
39
61
81
92
96
96
1
17.8
34.0
54.4
74.1
89.7
91.0
6
21
43
62
82
85
16
32
53
75
91
92
.25
.5
2.5
–
–
.0
6.3
0.5
9
4
.5
21.9
90.8
23
90
.0
84
1
A method by H NMR using an simple equation (% ester = 4A
1
/A
3
) for direct quantification, B method by H using analytical curve
a
1
The yields were calculated by GC-FID and by methods A and B by H NMR
cycles, the income was 90.8 % (cycle 1, entry 17) fell to
.09 % (cycle 10, entry 26). This huge drop comes is due to
the ratio of the integrals is placed on a curve, which min-
imizes this effect, resulting in a better correlation between
the points and minor errors. For both methods of quanti-
0
the fact that methanol can inhibit the enzyme. The enzyme
activity was drastically reduced for methanol, which was
used dried (Fig. 8d). For all other alcohols during the 10
reaction cycles, the yields did not change significantly 95.8,
1
fication by H NMR, is not possible when the conversion
yields are low (less than 5.0 % when analyzed by GC-FID)
as an external signal such as NMR measurement used in
the triplet is very small, making it difficult to obtain the
integral.
9
4.2 and 90.1 for ethanol, n-propanol and n-butanol,
respectively), showing that the enzyme can be reused for
the preparation of esters for at least 10 cycles. The results
are shown in Table 1.
The all spectrums and chromatograms are in supple-
mentary data.
Studies have shown that the degree of deactivation is
proportional to the number of carbon atoms present in the
alcohol, which, methanol has a greater influence [31, 32]
and the use of ethanol in methanol increases the rate of
speed of esterification and transesterification reactions [33].
It has been reported that lipase is inactivated by the use of
methanol (insoluble) in the form of droplets into oil and fat
4
Conclusions
The enzymatic esterification of fatty acid for the biodiesel
production with aliphatic alcohols showed excellent yields
including was possible the reuse the enzyme to ten cycles
and keep the yield to reactions. The esterification with
methanol showed a little dependence of water on the
reaction environment, due to the dehydration of enzyme
[
34, 35]. In transesterification, the immobilized lipase from
Ò
C. antarctica (Novozyme 435 ) was inactivated by using
.5 equivalents of methanol in relation to the oil. However,
this effect can be minimized if methanol is added in steps
36].
The enzymatic reactions presented in this work with the
1
1
and its inactivation. The quantification by H NMR showed
very efficient even when there was a partial overlap of the
signals, allowing the quantification of reaction with resid-
ual fatty acid quickly and non-destructive.
[
lipase CALB performance can be improved by performing
a study of different types of media that are most appro-
priate for this enzyme in the production of biodiesel.
Acknowledgments IGR thanks to CNPq for scholarship. The
authors would like to acknowledge FAPESP, CAPES and CNPq for
the financial support provided to this research and Novo Nordisk for
the donation of the enzyme.
1
.6 Comparison Between GC-FID and H NMR
3
In the comparison of the quantification methods employed,
was evaluated for the experiment 1 (amount of catalyst)
References
(
Table 2).
The quantification by H NMR using method B showed
1
. Robles AM, Gonz a´ lez PAM, Esteban LC, Molina EG (2009)
Biotech Adv 27:398
1
better results in comparison to method A. This could be
because, as is done in method A direct integration of the
signal to obtain the integral value and used as a low-power
device (200 MHz), the integration of the external signal
does not correctly represents the actual value. In method B,
2. Meher LC, Vidya DS, Naik SN (2006) Renew Sust Energy Rev
0:248
3
1
. Akoh CC, Chang SW, Lee GC, Shaw JF (2007) J Agric Food
Chem 55:8995
. Freedman B, Pryde EH, Mounts TL (1984) J Am Oil Chem Soc
62:1638
4
123

I. G. Rosset et al.
5
. S e´ verac E, Galy O, Turon F, Pantel CA, Condoret JS, Monsan P,
Marty A (2011) Enz Microb Technol 48:61
. Hwang ET, Gu MB (2013) Eng Life Sci 13:49
21. Gelbard G, Bres O, Vargas RM, Vielfaure F, Schuchardt UF
(1995) J Am Oil Chem Soc 10:1239
22. Neto PRC, Caro MSB, Mazzuco LM, Nascimento MG (2004)
J Am Oil Chem Soc 81:1111
23. Rosset IG, Tavares MCH, Assaf EM, Porto ALM (2011) Appl
Catal A: General 392:136
6
7
´
. Garcia-Galan C, Berenguer-Murcia A, Fernandez-Lafuente R,
Rodrigues RC (2011) Advan Synth Catal 353:2885
. Hernandez K, Fernandez-Lafuente R (2011) Enz Microb Technol
8
4
. Brady D, Jordaan J (2009) Biotechnol Lett 31:1639
8:107
24. Marchetti JM, Errazu AF (2008) Fuel 87:3477
25. Fjerbaek L, Christensen KV, Norddahl B (2009) Biotechnol
Bioeng 102:1298
26. Watanabe Y, Shimada Y, Sugihara A, Tominaga Y (2001) J Am
Oil Chem Soc 78:703
27. Pinto AC, Guarieiro LLN, Rezende MJC, Ribeiro NM, Torres
EA, Lopes WA, Pereira PAP, Andrade JB (2005) J Braz Chem
Soc 16:1313
28. Jegannathan KR, Abang S, Poncelet D, Chan ES, Ravindra P
(2008) Critical Rev Biotech 28:253
29. Lu J, Chen Y, Wang F, Tan T (2009) J Mol Cat B 56:122
30. Noureddini H, Gao X, Philkana R (2005) Biores Technol 96:769
31. Chen JW, Wu WT (2003) J Biosci Bioeng 95:466
32. Ranganathan SV, Narasimhan SL, Muthukumar K (2008) Biores
Technol 99:3975
33. Antczak MS, Kubiak A, Antczak T, Bielecki S (2009) Renew
Energy 34:1185
34. Al-zuhair S, Ling FW, Jun LS (2007) Process Biochem 42:951
35. Salis A, Pinna M, Moduzzi M, Solinas V (2005) J Biotech 2119:291
36. Shimada Y, Watanabe Y, Samukawa T, Sugihara A, Noda H,
Fukuda H, Tominaga Y (1999) J Am Oil Chem Soc 76:789
9
1
1
0. Iyer PV, Ananthanarayan L (2008) Process Biochem 43:1019
1. Mateo C, Palomo JM, Fernandez-Lorente G, Guisan JM,
Fernandez-Lafuente R (2007) Enzyme Microb Technol 40:1451
2. Maceiras R, Vega M, Costa C, Ramos P, M a´ rquez MC (2011)
Chem Eng J 166:358
3. Az o´ car L, Ciudad G, Heipieper HJ, Mu n˜ oz R, Navia R (2011)
J Biosci Bioeng 112:583
4. Zhang B, Weng Y, Xu H, Mao Z (2012) Appl Microbiol Bio-
technol 93:61
1
1
1
1
1
5. Rosset IG, Assaf EM, Porto ALM (2013) Curr Catal 2:53
6. Lai JQ, Hu ZL, Sheldon RA, Yang Z (2012) Process Biochem
4
7:2058
1
1
1
2
7. Lukovic N, Kneevic Z, Bezbradic D. (2011) In alternative fuel:
InTech
8. Dorado MP, Ballesteros E, de Almeida JA, Schellert C, L o¨ hrlein
HP, Krause R (2002) Trans ASAE 45:525
9. Knothe G, Gerpen JV (2005) The biodiesel handbook. AOCS
Publishing, Champaign, pp 25–89
0. Knothe G (2006) J Am Oil Chem Soc 83:823
1
23