An efficient activity ionic liquid-enzyme system for biodiesel production

Article abstract of DOI:10.1039/c0gc00230e

The efficient production of biodiesel in hydrophobic ionic liquids using immobilized lipase was demonstrated. The use of ionic liquids containing long alkyl chains on the cation has the important advantage of producing homogeneous systems at the start of the reaction but, when the reaction is complete, a three-phase system is created that allows selective extraction of the products using straightforward separation techniques, while the ionic liquid and the enzyme can be reused. Fifteen ionic liquids based on different alkyl chain length of the methyl imidazolium cation ([C₁₀MIM], [C ₁₂MIM], [C₁₄MIM], [C₁₆MIM] and [C ₁₈MIM]) combined with [BF₄], [PF₆] or [NTf ₂] anions were assayed as reaction media for two immobilized lipases (Candida antarctica lipase B and Pseudomonas fluorescens lipase AK) for biodiesel production. The highest synthetic activity was obtained in [C ₁₆MIM] [NTf₂] using Novozym 435 (immobilized Candida antarctica lipase with 245.13 U g^-1 IME), its activity being more than three times higher than in a solvent-free system. Additionally, in this IL the fatty acid methyl esters production was 90.29% after 3 h, while in the solvent-free system it was 27.3%. The influence of several reaction parameters, such as temperature, methanol-to-oil molar ratio, alkyl-chain length of the alcohols, IL:substrate volume ratios, amount of enzyme, and oils feedstock were studied and optimized.

Full text of DOI:10.1039/c0gc00230e

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 444
www.rsc.org/greenchem
PAPER
An efficient activity ionic liquid-enzyme system for biodiesel production
a
a
a
b
a
Teresa De Diego, Arturo Manj o´ n, Pedro Lozano, Michel Vaultier and Jos e´ L. Iborra*
Received 18th June 2010, Accepted 15th December 2010
DOI: 10.1039/c0gc00230e
The efficient production of biodiesel in hydrophobic ionic liquids using immobilized lipase was
demonstrated. The use of ionic liquids containing long alkyl chains on the cation has the
important advantage of producing homogeneous systems at the start of the reaction but, when the
reaction is complete, a three-phase system is created that allows selective extraction of the
products using straightforward separation techniques, while the ionic liquid and the enzyme can
be reused. Fifteen ionic liquids based on different alkyl chain length of the methyl imidazolium
cation ([C10MIM], [C12MIM], [C14MIM], [C16MIM] and [C18MIM]) combined with [BF
NTf ] anions were assayed as reaction media for two immobilized lipases (Candida antarctica
lipase B and Pseudomonas fluorescens lipase AK) for biodiesel production. The highest synthetic
activity was obtained in [C16MIM] [NTf ] using Novozym 435 (immobilized Candida antarctica
4
], [PF ] or
6
[
2
2
-
1
lipase with 245.13 U g IME), its activity being more than three times higher than in a
solvent-free system. Additionally, in this IL the fatty acid methyl esters production was 90.29%
after 3 h, while in the solvent-free system it was 27.3%. The influence of several reaction
parameters, such as temperature, methanol-to-oil molar ratio, alkyl-chain length of the alcohols,
IL : substrate volume ratios, amount of enzyme, and oils feedstock were studied and optimized.
1
. Introduction
is transesterification (alcoholysis) of oil (triglycerides) with
methanol in the presence of a catalyst, which gives biodiesel
The serious reduction of fossil fuel resources and an increasing
ecological awareness have led to the search for renewable
sources fuel, such as plant biomass. Biodiesel is a low-emissions
diesel substitute fuel made from renewable resources or waste
lipid. The most common method for biodiesel production
(
fatty acid methyl esters, FAMEs) and glycerol (by-product). The
transesterification reaction is normally a sequence of three con-
secutive reversible reactions. In this process, triglyceride (TAG)
is converted stepwise into diglyceride (DAG), monoglyceride
(MAG), and, finally, glycerol, with 1 mol of alkyl esters being
formed in each step (Fig. 1).
The transesterification process can be carried out in different
ways, using an alkali catalyst, an acid catalyst or a biocatalyst.
Chemical processes are used on an industrial scale, although
the reaction has several drawbacks: it is energy intensive,
glycerol recovery is difficult, the acidic or alkaline catalyst
has to be removed from the product, alkaline waste water
requires treatment, and water and free fatty acid cause a
partial saponification reaction which produces soap. Biodiesel
a
Departamento de Bioqu ´ı mica y Biolog ´ı a Molecular B e Inmunolog ´ı a,
Facultad de Qu ´ı mica, Universidad de Murcia, P.O. Box 4021, E-30100,
Murcia, Spain. E-mail: jliborra@um.es, amanjon@um.es,
tdp@um.es, plozanor@um.es; Fax: (+34) 868884148; Tel: (+34)
8
68887398
b
Laboratoire de Chimie et Photonique Mol e´ culaire. UMR CNRS 6510.
Universit e´ de Rennes-1, Institut de Chimie, UMR-CNRS 6510, Campus
de Beaulieu, Av. G e´ n e´ ral Leclerc, F-35042, Rennes, France.
E-mail: Michel.Vaultier@univ-rennes1.fr
Fig. 1 Biocatalytic production of biodiesel.
4
44 | Green Chem., 2011, 13, 444–451
This journal is © The Royal Society of Chemistry 2011

View Article Online
production with a biocatalyst eliminates all these disadvantages,
producing biodiesel with a very high degree of purity. However,
IL : substrate volume ratio, amount of enzyme and oil feedstock
was studied.
1
the process has not been implemented on an industrial scale for
several reasons, including the high cost of enzymes, low lipase
activity and enzyme inhibition by alcohol.
2
. Experimental
2
ꢀ
R
ꢀR
Enzymatic biodiesel production from raw vegetable oils has
Free (Novozym 525 L) and immobilized (Novozym 435)
Candida antarctica lipase B (EC 3.1.1.3) were from Novo
Spain, and free Pseudomonas fluorescens lipase AK was from
Amano Enzyme Japan. Ionic liquids at a purity of 99% were
from IoLiTec (Germany). Accurel MP 1000 was obtained
from Membrane GmbH Accurel Systems, Lewatit VP OC
1600 was obtained from Lanxess Chemical S.L., and Celite
545 was obtained from Merck. Olive and sunflower oils
were purchased from a supermarket, waste cooking oil was
obtained from a restaurant provenience, and palm oil was
purchased from Sigma-Aldrich Chemical Co. Standard fatty
acids (oleic, palmitic, linoleic, and stearic), fatty acid esters
(methylpalmitoleate, methyllinoleate, methylstearate), mono-
glycerides (monolinolein, monoolein, monopalmitin, monos-
tearin), diglycerides (1,3-dilinolein, 1,3-diolein, 1,3-dipalmitin
and 1,3-distearin) and triglycerides (trilinolein and triolein) were
from Sigma–Aldrich Chemical Co. Solvents and other chemicals
were purchased from Merck and were of the highest purity
available.
1–4
been extensively studied by many authors in recent years.
Some authors have even investigated the use of waste cooking
oil to obtain enzymatic biodiesel and to reduce the cost of raw
materials.
2,5
Nevertheless, in lipase-catalyzed biodiesel production,
methanol, which is not soluble in oil, has traditionally caused the
critical problem of enzyme deactivation. To avoid this problem,
a series of conventional organic solvents (petroleum ether,
hexane, cyclohexane, tert-butanol and tert-amyl alcohol) were
assayed. Tert-amyl alcohol was found to be the optimal reaction
medium for enzymatic biodiesel production, presumably due
to its good ability to solubilize the co-substrate, methanol,
and the by-product, glycerol. Other approaches have been
examined, including the stepwise addition of methanol or salt
6
solution. Although these approaches give good results, they
involve laborious procedures, including the periodic addition
of methanol or the removal of organic solvents from the
biodiesel.
In this context, the use of ionic liquids (ILs) as reaction
medium has opened up new opportunities for the enzymatic
production of biodiesel. The interest that ionic liquids hold as
green solvents resides in their negligible vapor pressure, excellent
thermal stability, high ability to dissolve a wide range of organic
and inorganic compounds, and their non-flammable nature,
which can be used to avoid the problem of the emission of volatile
organic solvents to the atmosphere. Moreover, their solvent
properties, such as miscibility or immiscibility with water or
some organic solvents (e.g. hexane), can be tuned by selecting the
appropriate cation and anion, which increases their usefulness
Transesterification reactions
Ninety microlitres of methanol (2.21 mmol) and 310 mL of
triolein (0.32 mmol) were added to screw-capped vials of 1 mL
total volume containing 800 mL of ionic liquid or tert-butanol.
The mixture was vigorously shaken to obtain a homogeneous
system. The reaction was started by adding 28 mg of Novozym
◦
4
35 (10%, w/w, based on oil weight) and run at 60 C in a
glycerol bath with shaking. At regular time intervals, 30 mL
aliquots were withdrawn and suspended in 470 mL of hexane.
The biphasic mixture was vigorously shaken for 3 min to extract
all the substrates and products into the hexane phase. Then
5
for recovering products from the reaction mixture. The most
interesting biotransformations in ILs have been observed at low
water content, or when using nearly anhydrous media, because
400 mL of the hexane extracts were added to 100 mL of 250 mM
ethyl decanoate (internal standard) solution in hexane and 40 mL
of the resultant solution was analyzed by HPLC. The volume of
each reaction medium was increased up to 5 times to distinguish
the phases at the end of the enzymatic reaction. For every
study, the following test conditions were optimized: temper-
ature, methanol-to-oil molar ratio, alkyl-chain length of the
alcohols, IL : substrate volume ratio, amount of enzyme, and oil
feedstock.
7
of the ability of hydrolases to carry out synthetic reactions.
Furthermore, it is possible to design two-phase reaction systems
8
that easily allow product recovery. Recently, four studies have
reported biodiesel enzymatic production by lipases in ionic
9
liquid with promising results. In all these works hydrophobic
ionic liquids were used, and FAME production was higher
than that obtained both in solvent free systems and organic
solvents.
In the present study, we have investigated biodiesel production
in ionic liquids catalyzed by lipase and the straightforward
extraction of the products using the properties of appropriate
ionic liquids. Fifteen ionic liquids based on different alkyl
chain lengths of the methyl imidazolium cation ([C10MIM],
Immobilization method
Lipase AK (30 mg) was dissolved in 2 mL of water and
mixed with 1 g of support. The mixture was shaken for 1 h at
room temperature. The suspension was centrifuged and washed
twice with 3 mL of water. The amounts of lipase adsorbed by
[C12MIM], [C14MIM], [C16MIM] and [C18MIM]) combined with
[
BF ], [PF ] or [NTf ] anions were assayed as reaction media
4
6
2
the different supports were quantified by the Lowry method.
◦
for two immobilized lipases (Candida antarctica lipase B and
Pseudomonas fluorescens lipase AK) for biodiesel production.
The C16MIM NTf
for biodiesel production catalyzed by Novozyme 435. The
influence of several reactions parameters, such as temperature,
methanol-to-oil molar ratio, alkyl-chain length of the alcohols,
The immobilized enzyme was immediately frozen at -20
C
and lyophilized for 48 h. Novozym 525 L was purified by
10a
2
was seen to be an excellent reaction medium
ultrafiltration as reported previously, resulting in a Candida
-
1
antarctica lipase B solution of 18.156 mg mL and a purification
degree of 1.5-fold. The immobilization process was the same as
for lipase AK.
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 444–451 | 445

View Article Online
◦
HPLC analysis
m
Table 1 Melting point (T ) for ILs by DSC ( C)
Reaction products were determined by a Shimadzu HPLC
equipped with a multi-channel pump (mod LC-20AD) and a
DAD detector (mod SPD-M20A), using a LiChroCart Lichro-
spher RP-18 column form Merck (25 cm length, 4.6 mm internal
diameter and 5 mm particle size). Three mobile phases were
employed; phase A, composed of acetronitrile and water (80 : 20,
v/v), phase B of acetronitrile only, and phase C, composed
of isopropanol and hexane (55 : 45, v/v). The flow rate was
Anion
[NTf
Cation
2
]
[PF
6
]
4
[BF ]
a
a
[C MIM]
10
-28.9
16.7
34.6
42.6
51.3
34.8
56.5
73.0
77.0
80.3
-4.2
34.0
38.0
49.6
66.8
[C
[C
[C
[C
12MIM]
14MIM]
16MIM]
18MIM]
a
.2 mL min-1 and the injection volume 40 mL. The protocol
employed for the mobile phase involved a linear gradient of
00% (v/v) A, decreasing to 0% v/v A in 5 min, while phase
Data from Zhang et al., 2006.
1
1
checked in media containing methyl imidazolium ionic liquids
(MIM) of different cation chain lengths (C10, C12, C14, C16 and
18) and in combination with three different anions ([BF
]). Ionic liquids were dried in a vacuum oven at 60 C
at least for 24 h before reaction to remove all water, so that the
final water content was lower than 0.10–0.15%, depending on
B was increased up to 100%, and maintained for 2 min, before
decreasing to 50% in 16 min. Phase C increased up to 50%,
and the final mixture (50 : 50, v/v B/C) was held for 10 min.
Finally, the system was restored to its initial condition for
C
4
], [PF
6
]
◦
and [NTf
2
6
min. Elution profiles were monitored at 210 nm. Peaks in the
chromatograms were identified by comparing retention times
with the appropriate corresponding standards. The conversion
to products was calculated using the following:
the on hygroscopic properties of ILs. The melting point (T )
was determined for all the ionic liquids used by differential
scanning calorimetry (Table 1). It has been pointed out that
m
the T of methyl imidazolium type ionic liquids is governed
m
G
o
= MAG + DAG + TAG
MAG (%) = (MAG /G ) ¥ 100
DAG (%) = (DAG /G ) ¥ 100
TAG (%) = (TAG /G ) ¥ 100
/G ) ¥ 100
by van der Waals forces rather than by the electrostatic forces,
and for such cations, the change of phase transition temperature
t
o
11
depends of the anions more than on cations. Moreover, the
increases with the alkyl chain length of the cation for the
T
m
t
o
same anion. These results are in agreement with those of other
1
1,12
◦
authors.
When using IL with a T
m
higher than 60 C was
t
o
◦
◦
used first liquefied at 80 C and then kept liquid at 60
C
while substrates (methanol and oil) were added to carry out
the enzymatic transesterification process.
FAMEs (%) = (FAMEs
is the glyceride content at time t = 0, MAG
TAGt is the total sum of glycerides (mono, di and triglycerides)
at time t = t and FAMEs is the total sum of all fatty acid methyl
esters, at a time t = t. The areas of substrates and products were
normalized with respect to the internal standard area. One unit
of specific activity was defined as the amount of enzyme that
produced 1 mmol FAME per min.
t
o
where G
o
t
, DAG
t
and
Fig. 2 shows that triglyceride consumption and the production
of methyl esters was almost complete after about 6 h. The
presence of intermediate products (MAG and DAG) was
detected by HPLC, and the total consumption of MAG required
a slightly longer time of about 12 h.
t
Novozym 435 is a commercial lipase and consists of 10%-by-
wt lipase B physically adsorbed within 90% by wt Lewatit VPOC
10b
1
600 support, so, 28 mg of Novozyme 435 contains 2.8 mg of
lipase B.
Differential scanning calorimetry of ILs
A differential scanning calorimeter DSC 2920 (TA Instruments,
New Castle, DE) was used to measure the melting point of
ionic liquids. Samples were hermetically sealed in pressure vessel.
◦
The DSC temperature program was as follows: 25 C of initial
◦
-1
temperature held for 10 min, and then a gradient of 1 C min
up to 100 C.
◦
Fig. 2 Time course for methanolysis of triolein by Novozyme 435
◦
in [C16MIM] [NTf
2
], at 60 C.: (᭹) FAME, (ꢀ) triglycerides (ꢀ)
diglycerides, and (᭜) monoglycerides.
3
. Results and discussion
The enzymatic methanolysis of lipase was tested for the synthesis
of fatty acid methyl esters (FAMEs) in fifteen different ionic
liquids at 60 C. The ionic liquids tested in this work are not
capable of carrying out the transesterification reaction without
the presence of lipase. The catalytic efficiency of the immobilized
enzyme (Novozym 435) on the methanolysis reaction was then
The time course for methanolysis catalyzed by the enzyme was
charted for all ILs, and the initial rate of the reaction for each
case was determined from the slope of the best-fit of the product
concentration versus time curve. For all the assayed ILs, both
the synthetic activity and product yield at 24 h by Novozym 435
are shown in Fig. 3.
◦
4
46 | Green Chem., 2011, 13, 444–451
This journal is © The Royal Society of Chemistry 2011

View Article Online
solubility of methanol in triglycerides. To avoid these effects, the
addition of methanol was conducted in three stages, 1/3 molar
equivalent methanol (30 mL) every eight hours, which increased
-
1
the synthetic activity to 76.55 Ug IME and of the product yield
after 24 h to 87.21%.
On the other hand, for IL containing NTf
maximum synthetic activity was obtained for [C16MIM] cation,
while the activity was higher for IL containing the NTf anion
than for the PF anion (see Fig. 3). The highest synthetic activity
was obtained in [C16MIM] [NTf
more than three times that obtained in the free solvent system).
Additionally, the FAME production in [C16MIM] [NTf ] was
0.29% after 3 h while in the solvent-free system it was only
7.3%. However, for ILs based on the BF anion the observed
behaviour of the enzyme cannot be explained by the properties
of the IL containing this anion. The highest synthetic activity
was obtained for the [C12MIM] cation, although the synthetic
activity decreased as the alkyl chain of the imidazolium cation
increased. This observation is in agreement with others authors
2
and PF
6
anions, the
2
6
-
1
2
] (245.13 Ug IME, a value
2
9
2
4
Fig. 3 Synthetic activity of FAMEs in fifteen ionic liquids based on
different alkyl chain length of methyl imidazolium cation ([C10MIM],
[
(
C
12MIM], [C14MIM], [C16MIM] and [C18MIM]) combined with [BF
dark grey), [PF ] (light grey) or [NTf ] (white). The numbers are the
yield after 24 h of reaction time.
4
]
6
2
In all cases, an increase in alkyl chain length of the corre-
sponding imidazolium cation improved synthetic activity up
to a maximum, after which it decreased, which agrees with
previous results obtained for Novozym 435 when catalyzing
who have described similar behaviours for ILs based on the BF
4
anion. All the immobilized lipases decreased their activity as
the alkyl chain of the imidazolium cation increased in different
16
synthetic reactions. These observations may be explained
by the different enzyme microenvironments of each assayed
7c
the synthesis of citronellyl esters, or in the synthesis of butyl
13
13
propionate. This can be partially explained by the higher
solubility of the oil substrate in the more hydrophobic ILs, since
the hydrophobic character of the IL increases with the alkyl
biocatalyst in their respective interaction with IL.
In most of the ionic liquids tested in this work, the reaction
mixture was homogeneous, but, when the reaction finished,
a triphasic system was created through the appearance of a
FAME phase (upper layer), a glycerol phase (middle layer), and
an ionic liquid phase containing the enzyme (the lower layer).
Additionally, a decrease in the reaction temperature (brought
about with ice) results in the solidification of many ionic liquids
(see Table 1), which facilitates the extraction of the biodiesel
product. These observations demonstrate the suitability of using
the ionic liquids tested for the production of biodiesel.
13,14
chain length of the cation for the same anion.
On the other
hand, viscosity increases with the increasing length of the alkyl
chain substituent on the imidazolium ring increases, although
it also depends on the anion type. For the [CnMIM] cations,
the viscosity increased in this order with the anion type: [PF ]
6
8b
>
[BF
4
] > [NTf ]. Increasing viscosity negatively affects the
2
operating process through increased mass transfer resistance.
Among the ionic liquids tested for biodiesel production, as seen
in the literature, only methyltrioctylamonium trifluoroacetate
Novozym 525-L is a commercially available soluble lipase.
Novozym 435 is a heterogeneous biocatalyst based on lipase
B from Candida antarctica (CALB, Novozym 525-L) immobi-
lized within a macroporous resin of poly-(methylmethacrylate)
(Lewatit VP OC 1600), and containing roughly 10% by weight
9c
(
TOMA TFA) has a long alkyl chain length (C8) However,
this ionic liquid has never been used in its completely pure state,
and free traces of TFA have been demonstrated to act as acid
catalyst or the synthesis of FAMEs by transesterification, so that
the results obtained, in this work may not only be due solely to
17
of CALB physically adsorbed within the support. These two
biocatalysts are among the most widely used enzymes in Biotech-
nology because of their versatility, especially in non-aqueous
media. However, Pseudomonas fluorescens lipase seems to be
a better choice for biodiesel synthesis because it is methanol-
9c
the enzymatic activity.
Tert-butanol was employed as a model system to examine the
effects of ILs on enzyme activity. This organic solvent improves
the enzymatic process because it dissolves both methanol and
triglyceride, and decreases mass transfer resistances. Although
an alcohol, tert-butanol is not a substrate for the lipase, because
this enzyme does not show activity on tertiary alcohols. The
18
resistant. In this sense, the two soluble lipases (Novozym 525-L
and the Pseudomonas fluorescens lipase) were immobilized onto
four supports, Lewatit VP OC 1600, celite, silica and Accurel
-
1
synthetic activity in tert-butanol was 239.04 U g IME and the
MP1000 (a macroporous polypropylene support) to study their
◦
product yield on FAMEs was 98.20% after 24 h, results that
biodiesel synthetic activity in [C16MIM][NTf
2
] at 60 C. The
3
,7a,b
reflect those of authors.
However, it should be noted that
chosen supports have been used by other authors for lipase
immobilization and have also been used for biodiesel production
when a solvent is used, the processing cost is higher due to the
extra processing equipment required for the separation of the
co-solvent and the cost of the co-solvent itself.
18,19
both in organic solvents and in free solvents systems.
In all
cases, the lipases immobilization on the supports was by physical
adsorption.
The synthetic activity for the different immobilization deriva-
tives are depicted in Fig. 4, as can be seen, each immobilized
lipase showed a different synthetic activity level on each one of
the assayed immobilization supports. The synthetic activity of
When methanolysis in a solvent free system without ionic
-
1
liquid was analyzed, a synthetic activity of 11.92 Ug IME
was calculated and a production yield of 45.46% after 24 h was
obtained. Such a decrease in synthetic activity may be due to the
above mentioned enzyme deactivation by methanol or to the low
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 444–451 | 447

View Article Online
Fig. 4 Synthetic activity of FAMEs catalyzed by lipase AK (white) and
lipase B (grey) immobilized on different supports.
Fig.
5
Effect of temperature on the synthetic activity in
[
C
16MIM][NTf
2
], at the fixed conditions of alcohol to oil molar ratio
(
6 : 1) and Novozym 435 (10% w/w, based on oil weight).
CALB lipase (Novozym 525-L) immobilized on Lewatit in the
this value depends on the properties of oil, the type of catalyst,
and the reaction media used.
-
1
2,3,9
laboratory (2.98 U mg enzyme) was similar to that shown by
-
1
their commercially available counterpart (2.45 U mg enzyme).
The lipase AK seemed to be a worse catalyst than CALB for
the proposed reaction, except in the case of Accurel, when the
advantage of using lipase AK was slight. The results shown in
Fig. 4 could be explained by taking into account the active role
of the immobilization support in the microenvironment and its
The effect of the substrate molar ratio (methanol:triolein) on
synthetic activity in [C MIM] [NTf ] is shown in Fig. 6. The
1
6
2
results indicated that the highest activity was achieved at a 6 : 1
ratio, and that methanolysis decreased when the molar ratio
was higher than 6 : 1, because the high methanol concentration
9b
resulted in the inactivation of Novozym 435, and because,
under such conditions, the triolein was not completely dissolved
in methanol. However, in a free solvent system, the immobilized
lipase B was partially deactivated by a methanol : oil ratio of
7d
interaction with ILs. Thus, adsorption onto a macroporous
polyacrylic resin (Lewatit) takes place largely through presumed
hydrophobic interactions. In contrast, the silica support may
interact electrostatically with positive groups on the enzymes
surface, while the macroporous polypropylene support (Accurel)
is hydrophobic, and previous studies have shown that, at low
loading, the hydrophobic interactions between lipases and
polypropylene are strong enough to cause inactivation of a
-
1
6 : 1 (11.92 U g IME). These results led to the conclusion that
the presence of the ionic liquid mitigates the inactivation of the
biocatalyst. These results are in concordance with those of other
9
authors.
18
significant fraction of the enzyme molecules. Furthermore, in
water free systems the hydration level of the enzyme is one of
the fundamental parameters to be considered, since it strongly
affects enzyme activity. In the case of immobilized enzymes,
the partitioning of water among the components of the system
(
enzyme, support and reagent mixture) strongly depends on the
20
hydrophilic/hydrophobic balance of the support. On the basis
of the obtained results, Novozym 435 was used in the subsequent
steps of works.
In order to improve the efficiency of the lipase-catalyzed
synthesis of FAMEs, the influence of temperatures ranging from
◦
5
0 to 90 C was also studied in [C16MIM][NTf
2
] (Fig. 5).
As can be seen, the synthetic activity of Novozyme shows a
-
1
curve with a maximum level of activity (245.13 U g IME) at
Fig. 6 Influence of methanol : triolein molar ratio on the synthetic
◦
6
0 C. The synthetic activity in the range of 60 to 90 degrees is
greater than at 50 C, because IL viscosity decreased when the
activity in [C16MIM][NTf
(60 C) and Novozym 435 (10% w/w, based on oil weight).
2
], at the fixed conditions of temperature
◦
◦
15,21
temperature increased.
In all cases, the FAME production
◦
after 24 h was close to 100%, except at 90 C (93.87%).
The molar ratio of methanol to oil has been reported to
be one of the most important variables affecting the yield
The effect of the ionic liquid-to-substrate volumetric ratio on
synthetic activity in [C16MIM] [NTf
fixed conditions of 60 C, an alcohol to triolein molar ratio of
6 : 1 and 10%w/w Novozym 435 (based on oil weight).
2
] is shown in Fig. 7, in the
◦
4
of fatty acid methyl ester. The stoichiometric ratio for the
transesterification reaction requires only 3 moles of alcohol and
The ionic liquid : substrate volume ratio shown in Fig. 7,
1 : 0.5, 1 : 1, 1 : 5 and 1 : 10, corresponds to ionic liquid : triolein
volume ratio as follows, 1 : 0.38, 1 : 0.77, 1 : 3.87, 1 : 7.75, respec-
tively. In all cases, the initial system for the transterification
1
mole of triglyceride to yield 3 moles of fatty acid methyl ester
and 1 mole of glycerol, although a higher molar ratio resulted
in a larger ester conversion rate in a shorter time. Furthermore,
4
48 | Green Chem., 2011, 13, 444–451
This journal is © The Royal Society of Chemistry 2011

View Article Online
alkyl chain length. Furthermore, alcohols with higher molec-
ular mass showed higher density and greater mass transfer
4
limitations. Among these alcohols, methanol and ethanol are
the cheapest which is one of the reasons why used for industrial
biodiesel production.
To optimize the process biodiesel production, commercial
grade and analytical reagent methanol have been used for
enzymatic methanolysis, it being found that both the synthetic
activity and the yield were the same. As economical consider-
ations are a major issue in biodiesel production, the cheaper
grade methanol is more frequent.
The high cost of lipase is one of the main problems for the
application of enzymatic processes to the industrial production
of biodiesel fuel from vegetable oil. For this reason, the effect of
Fig. 7 The effect of ionic liquid : substrate ratio (based on v/v) on
11
the synthetic activity in [C16MIM] [NTf
temperature (60 C), alcohol to oil molar ratio (6 : 1) and Novozym 435
2
] at the fixed conditions of
the quantity of enzyme was studied on the methanolysis reaction
◦
◦
in [C10MIM][NTf
2
] at 60 C (Fig. 9).
(
10% w/w, based on oil weight).
reaction was a homogenous medium. For the case of [C16MIM]
NTf ], the synthetic activity of FAMEs decreased when the
[
2
IL : substrate ratio increased, because the higher methanol
concentration present in the reaction media the greater the
enzyme deactivation.
None of the recent works by other authors using ionic liquids
as a medium for the enzymatic production of biodiesel has
analyzed this effect, since all used an ionic liquid : oil volume
9
9c
ratio of 1 : 1, except Miyawaki and Tatsuno who demonstrated
that the addition of 80% of methyltrioctylamonium trifluoroac-
etate accelerated the butanolysis of triolein, since traces of free
TFA can act as acid catalyst for the synthesis of FAMEs by
transesterification.
Fig. 9 Influence of immobilized lipase B quantity on synthetic activity
for the synthesis of FAMEs in [C16MIM][NTf
2
] (based on olein weight)
◦
at the fixed conditions of temperature (60 C) and alcohol-to-oil molar
ratio (6 : 1).
In order to analyze the performance of the acyl acceptor on
the synthetic activity of biodiesel in [C16MIM][NTf ] medium,
2
four primary alcohols (methanol, ethanol, n-propanol and n-
The maximal synthetic activity was obtained for 28 mg of
immobilized enzyme, which represents a 10% w/w (based on oil
weight) while higher concentrations led to lows yields. Visual
observations during the assays using enzyme concentration
higher than 10% indicated that the liquid phase was not of
sufficient volume to completely suspend the solid catalyst. In
these circumstances, external mass transfer resistance becomes
the limiting step for oil transesterification. This behaviour is
butanol) was compared in the fixed conditions of temperature
◦
(
60 C), alcohol to triolein molar ratio (6 : 1) and Novozym 435,
1
0% w/w (based on oil weight). (Fig. 8).
2
in accordance with that obtained other researchers. The cost
of any enzyme used in biodiesel production is of primary
important.
The most used vegetable oils in biocatalysed biodiesel produc-
tion are those obtained from soybean seeds, sunflower seeds, and
rape seeds in many European countries. Countries use different
types of oil feedstock depending on abundance or availability;
for example, soybean oil in the USA and coconut and palm oil
in tropical countries.
The effect of triglyceride feedstock on the initial yield of
FAMEs was analyzed in [C16MIM][NTf ]. In the present work
2
four different vegetable oils were used (olive oil, sunflower oil,
palm oil, and waste cooking oil) in addition to triolein. The
fatty acid composition of these oils, as determined by HPLC
analysis, is reported in Table 2 and is in concordance with those
Fig. 8 Effect of different acyl acceptors on the synthetic activity in
◦
[C
16MIM] [NTf
2
] at the fixed conditions of temperature (60 C), alcohol-
to-oil molar ratio (6 : 1) and Novozym 435 (10% w/w, based on oil
weight).
18
reported by other authors. FAME content (mmoles) profiles for
◦
The highest synthetic activity was achieved with methanol,
with enzymatic alcoholysis decreasing with increasing alcohol
all the oils are shown in Fig. 10 by Novozyme 435 at 60 C. The
-
1
reaction rate (mmoles h ) and production yields (%) after 24 h
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 444–451 | 449

View Article Online
Table 2 Fatty acid composition of the oils (%)
and when the reaction is complete a triphasic system is created
through the appearance of a FAMEs phase (upper layer),
a glycerol phase (middle layer) and a lower layer with the
ionic liquid containing the enzyme, which can be solidified
by decreasing the media reaction temperature, thus facilitating
the extraction of the biodiesel product. So, the use of ionic
liquids containing long alkyl chains on the cations as well
Oil Type
Olive oil
Sunflower oil
Palm oil
Waste oil
Palmitic acid
Stearic acid
Oleic acid
12.37
2.98
80.46
4.19
6.59
3.46
23.43
66.52
38.67
3.32
46.14
11.87
6.90
3.89
30.52
58.69
Linoleic acid
as hydrophobic anions ([PF
extraction of the products using straightforward separation
techniques.
6
] or [NTf ]) allows the selective
2
Table 3 Influence of oil type on catalytic efficiency and production
yield after 24 h
-1
Oil Type
Rate (mmoles h )
Yield after 24 h (%)
Acknowledgements
Triolein
Olive
Sunflower
Palm
Cooking waste
478.08
345.71
407.55
428.07
440.35
99.39
93.38
92.78
94.11
96.91
This work was partially supported by the CICYT (Ref.
CTQ2008-00877), SENECA Foundation (Ref.: 11975/PI/09).
We thank Ms. Eulalia G o´ mez L o´ pez, laboratory technician
in the University of Murcia, for her help, and Mr. Ramiro
Mart ´ı nez, from Novo Spain, for the gift of Novozym.
References
1
2
3
4
5
H. Fokuda, A. Kondo and H. Noda, J. Biosour. Bioeng., 2001, 11,
300–1311.
R. Maceiras, M. Vega, C. Costa, P. Ramos and M. C. M a´ rquez, Fuel,
009, 88, 2130–2134.
D. Royon, M. Daz, G. Ellenrieder and S. Locatelli, Bioresour.
Technol., 2007, 98, 648–653.
M. S. Antczak, A. Kubiak, T. Antczak and S. Bielecki, Renew. Energy,
1
2
2
009, 34, 1185–1194.
(a) Z. F. Chen, M. H. Zong and H. Wu, Div. Fuel Chem., 2005, 50,
; (b) M. G. De Paola, E. Ricca, V. Calabr o` , S. Curcio and G. Iorio,
2
Bioresour. Technol., 2009, 100, 5126–5131; (c) N. Li, M.-H. Zong and
H. Wu, Process Biochem., 2009, 44, 685–688.
Fig. 10 Profiles of FAME content (mmoles) for every one of the oils:
6
(a) M. Kaieda, T. Samukawa, T. Matsumoto, K. Ban, A. Kondo and
Y. Shimada, J. Biosci. Bioeng., 1999, 88, 627–631; (b) K. Ban, M.
Kaieda, T. Matsumoto, A. Kondo and H. Fakuda, Biochem. Eng. J.,
(
᭹) triolein, (ꢀ) olive oil, (ꢀ) sunflower oil, (᭺) palm oil and (ꢁ) waste
cooking oil.
2
001, 8, 39–43; (c) K. Ban, S. Hama, K. Nishizuka, M. Kaieda, T.
Matsumnoto, A. Kondo, H. Noda and H. Fakuda, J. Mol. Catal. B:
Enzym., 2002, 17, 157–165; (d) Y. Shimada, Y. Watanabe, A. Sugihara
and Y. Tominaga, J. Mol. Catal. B: Enzym., 2002, 14, 133–142.
are shown in Table 3. There were no significant differences in the
two parameters for the different oils, the yield after 24 h rouging
from 93 to 97%, except in the case of triolein, when it was slightly
higher.
Chen et al. showed that waste cooking oil can be used
as feedstock for biodiesel production and that most of the
properties of the resulting biodiesel are better than those of diesel
in China and of biodiesel from Germany and USA. FAME type
and degree of unsaturation are important factors for the quality
of biodiesel, because they influence cetane number, viscosity,
iodine value, etc.
7 (a) F. van Rantwijk and R. A. Sheldon, Chem. Rev., 2007, 107, 2757–
785; (b) M. Erbeldinger, A. J. Mesiano and A. J. Russell, Biotechnol.
2
5a
Prog., 2000, 16, 1129–1131; (c) P. Lozano, T. De Diego, D. Carri e´ ,
M. Vaultier and J. L. Iborra, J. Mol. Catal. B: Enzym., 2003, 21,
1
31–138; (d) P. Lozano, R. Piamtongkam, K. Kohns, T. De Diego,
M. Vaultier and J. L. Iborra, Green Chem., 2007, 9, 780–784.
(a) T. Itoh, Y. Nishimura, N. Ouchi and S. Hayase, J. Mol. Catal.
B: Enzym., 2003, 26, 41–45; (b) P. Lozano, T. De Diego, S. Gmouh,
M. Vaultier and J. L. Iborra, Biotechnol. Prog., 2004, 20, 661–669;
(c) Y. Abe, K. Kude, S. Hayase, M. Kawatsura, K. Tsunashima and
T. Itoh, J. Mol. Catal. B: Enzym., 2008, 51, 81–85.
(a) S. Sunitha, S. Kanjilal, P. S. Reddy and R. B. N. Prasad, Biotechnol.
Lett., 2007, 29, 1881–1885; (b) S. H. Ha, M. N. Lan, S. H. Lee, S. M.
Hwang and Y.-M. Koo, Enzyme Microb. Technol., 2007, 41, 480–483;
(c) O. Miyawaki and M. Tatsuno, J. Biosci. Bioeng., 2008, 105, 61–64;
8
9
Finally, the triglyceride feedstock had no significant effect on
the enzyme catalytic efficiency. It seems, then that the use of
waste cooking oil could greatly reduce the cost of biodiesel.
(
d) S. Arai, K. Nakashima, T. Tanino, Ch. Ogino, A. Kondo and H.
Fukuda, Enzyme Microb. Technol., 2010, 46, 51–55.
Conclusions
10 (a) P. Lozano, T. De Diego, S. Gmouh, M. Vaultier and J. L. Iborra,
Biotechnol. Lett., 2001, 23, 1529–1533; (b) A. Mahapatro, A. Kumar
and A. Gross, Biomacromolecules, 2044, 5, 62–68.
The suitability of ionic liquids containing long alkyl chains in
cations as reaction media for the enzymatic production biodiesel
has been demonstrated. The highest FAME production after
1
1 S. Zhang, N. Sun, X. He, X. Lu and X. Zhang, J. Phys. Chem., 2006,
5, 1475–1517.
3
1
2 (a) C. M. Gordon, J. D. Holbrey, A. R. Kennedy and K. R. Seddon,
J. Mater. Chem., 1998, 8, 2627–2636; (b) J. D. Holbrey and K. R.
Seddon, J. Chem. Soc., Dalton Trans., 1999, 13, 2133–2139; (c) A. E.
Bradley, C. Hardacre, J. D. Holbrey, S. Johnston, S. E. J. Mcmath
and M. Nieuwenhuyzen, Chem. Mater., 2002, 14, 629–635.
3 T. De Diego, P. Lozano, M. A. Abad, K. Steffensky, M. Vaultier and
J. L. Iborra, J. Biotechnol., 2009, 140, 234–241.
2
4 h was obtained in C16MIM NTf . Several reaction parameters,
2
such as temperature, methanol : oil molar ratio, alkyl-chain
length of the alcohols, IL : substrate volume ratios, quantity of
enzyme, and oils feedstock, were optimized. Furthermore, when
1
C
16MIM NTf
2
is used the reaction mixture is homogeneous,
4
50 | Green Chem., 2011, 13, 444–451
This journal is © The Royal Society of Chemistry 2011

View Article Online
1
1
1
4 L. Ropel, L. S. Belv e` re, S. N. V. K. Aki, M. A. Stadtherr and J. F.
Brennecke, Green Chem., 2005, 7, 83–90.
5 Y. Watanabe, Y. Shimada, A. Sugihara and Y. Tominaga, J. Mol.
Catal. B: Enzym., 2002, 17, 151–155.
6 (a) M. Persson and U. T. Bornscheuer, J. Mol. Catal. B: Enzym., 2003,
18 (a) A. Salis, M. Pinna, M. Monduzzi and V. Solinas, J. Mol. Catal.
B: Enzym., 2008, 54, 19–26; (b) A. Salis, M. S. Bhattacharyya, M.
Monduzzi and V. Solinas, J. Mol. Catal. B: Enzym., 2009, 57, 262–
269.
19 (a) B. Cheirsilp, W. Kaewthong and A. H. Kittikun, Biochem. Eng. J.,
2007, 35, 71–80; (b) B. Cheirsilp, A. H-Kittikun and S. Limkatanyu,
Biochem. Eng. J., 2008, 42, 261–269; (c) L. Freitas, P. C. M. Da
R o´ s, J. C. Santos and H. F. de Castro, Process Biochem., 2009, 44,
1068–1074.
20 P. Adlercreutz, in: A. J. J. Straathof, P. Adlercreutz, ed., (Applied
Biocatalysis, 2nd ed., Harwood Academic Publishers, Amsterdam,
2000, p. 295.
2
5
2
2, 21–27; (b) R. Irimescu and K. Kato, Tetrahedron Lett., 2004, 45,
23–525; (c) O. K o¨ se, M. Tuter and H. A. Aksoy, Bioresour. Technol.,
002, 83, 125–129.
1
7 (a) A. Mahapatro, A. Kumar and R. Gross, Biomacromolecules,
004, 5, 62–68; (b) A. B. R. Moreira, V. H. Perez, G. M.
Zanin GM and H. F. de Castro, Energy & Fuels, 2007, 21,
689–3694; (c) M. Mittelbach and C. Remschmidt, Biodiesel.
The Comprehensive Handbook, Martin Mittelbach Publisher, Graz,
005.
2
3
21 Y. Watanabe, Y. Shimada, A. Sugihara, H. Noda, H. Fukuda and Y.
Tominaga, J. Am. Oil Chem. Soc., 2000, 77, 355–360.
2
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 444–451 | 451