Preparation and properties of xerogels obtained by ionic liquid incorporation during the immobilization of lipase by the sol-gel method

Article abstract of DOI:10.1016/j.molcatb.2010.01.027

Lipase from Pseudomonas fluorescens (Amano AK) has been immobilized by the sol-gel method using tetramethoxysilane and trimethoxysilanes with alkyl or aryl groups as precursors and ionic liquids as immobilization additives. Room temperature ionic liquids

Full text of DOI:10.1016/j.molcatb.2010.01.027

Journal of Molecular Catalysis B: Enzymatic 65 (2010) 79–86
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Preparation and properties of xerogels obtained by ionic liquid incorporation
during the immobilization of lipase by the sol–gel method
Cristina Zarcula, Livia Corîci, Ramona Croitoru, Anca Ursoiu, Francisc Peter∗
University “Politehnica” of Timis¸oara, Faculty of Industrial Chemistry and Environmental Engineering, C. Telbisz 6, 300001 Timis¸oara, Romania
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 1 February 2010
Lipase from Pseudomonas fluorescens (Amano AK) has been immobilized by the sol–gel method using
tetramethoxysilane and trimethoxysilanes with alkyl or aryl groups as precursors and ionic liquids as
immobilization additives. Room temperature ionic liquids (ILs) based on imidazolium cations with dif-
ferent hydrophobicities and inorganic or organic anions were investigated. The biocatalytic activity in the
acylation reaction of secondary aliphatic alcohols by vinyl acetate increased at lower polarity of the IL,
being 58% higher in case of octyl than for ethyl substituent of the alkylimidazolium cation. The optimum
IL/silane molar ratio in the immobilization mixture was around 0.2. Hydrophobic groups from both the
silane precursor and IL converge to the optimal microenvironment for the enzyme action. ILs as reaction
media led to higher activities as common organic solvents in the kinetic resolution of secondary alcohols,
but the highest enantioselectivities (enantiomeric ratio E values > 50) have been observed in acetone and
tetrahydrofuran. Structural investigation of the preparates by SEM–EDX and FT-IR has confirmed the
partial confinement of the IL in the xerogel structure.
Keywords:
Ionic liquid
Sol–gel
Immobilization
Lipase
2-Hexanol
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
lipase exhibited higher activity and enantioselectivity in the ester-
ification of (R,S)-2-chloropropanoic acid in ionic liquids compared
Biocatalytic reactions in non-aqueous solvents have been
attested in the last decades as realistic alternatives to replace con-
ventional chemical procedures, especially when increased regio-
or enantioselectivity was needed [1–3]. However, utilization of
volatile organic solvents could be harmful for the environment,
even if their recovery is usually performed at the end of the reac-
tion. ILs have been attracted a lot of interest among non-aqueous
reaction media, recommended by favorable properties as very low
vapor pressure, high thermal stability, miscibility adjusted by the
appropriate selection of cationic and/or anionic moiety [4]. Biocat-
alytic performance of enzymes in ILs was the subject of extensive
investigations and it was obviously demonstrated that in several
cases was not only maintained, but rather increased compared to
conventional media [5,6]. Gubicza et al. showed that Candida rugosa
to that in hexane [7,8]. As enzymes are not soluble in most ILs
when used as pure solvents, they have been employed either in
immobilized or suspended native form. Based on process efficiency
parameters, immobilized enzymes seem to be the right option,
considering their improved operational and thermal stability [9].
Novozyme 435 has been used with excellent results for the syn-
thesis of isoamyl acetate by direct esterification in ILs, maintaining
its catalytic activity at the same level following 10 reuse cycles [10].
Although ionic liquids are documented primarily as valuable
reaction media to replace volatile organic solvents, they could
have other interesting applications like additives during the sol–gel
immobilization process of enzymes. Sol–gel encapsulation has
proved to be a versatile technique for the immobilization of
a large variety of biomolecules [11]. Hybrid organic–inorganic
sol–gel matrices are formed by the hydrolysis of silane precursors,
tetra-alcoxysilanes mixed with alcoxysilanes carrying hydropho-
bic groups, followed by condensation reactions to yield silica. Silica
particles grow gradually as condensation proceeds, leading to the
formation of colloidal sols and further gels. Whether the formed
gels are allowed to maturate and dry at room temperature, a porous
silica network is formed, with the enzyme entrapped in the sol–gel
matrix and the hydrophobic groups partly covering the surface [12].
Lee et al. showed that ILs can be used to protect Candida rugosa
lipase against inactivation by the released alcohol and shrinking of
gel during the maturation and drying step of the sol–gel immobi-
Abbreviations: [Emim]+, 1-ethyl-3-methyl-imidazolium; [Pmim]+, 1-propyl-
3-methyl-imidazolium; [Bmim]+, 1-butyl-3-methyl-imidazolium; [Hmim]+, 1-
hexyl-3-methyl-imidazolium; [Omim]+, 1-octyl-3-methyl-imidazolium; [Tf2N]−
,
bis(trifluoromethyl-sulfonyl)imide; MeTMOS, methyl-trimethoxysilane; PrTMOS,
propyl-trimethoxysilane; OcTMOS, octyl-trimethoxysilane; PhTMOS, phenyl-
trimethoxysilane; TMOS, tetramethoxysilane; DMF, N,N-dimethylformamide; THF,
tetrahydrofuran; MTBE, methyl t-butyl ether; IL(s), ionic liquid(s); SEM–EDX, scan-
ning electron microscopy coupled with energy dispersive X-ray spectroscopy.
∗
Corresponding author. Tel.: +40 256 404216; fax: +40 256 403060.
E-mail addresses: Francisc52@gmail.com, francisc.peter@chim.upt.ro (F. Peter).
1381-1177/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2010.01.027

80
C. Zarcula et al. / Journal of Molecular Catalysis B: Enzymatic 65 (2010) 79–86
lization process [13]. Karout and Pierre find that ILs as additives
increased the gelation time, influenced the pore size distribution
and some residual ILs remained attached to the sol–gel network
[14]. Our preliminary results demonstrated that the catalytic effi-
ciency of microbial lipases in the acylation reaction of secondary
alcohols could be enhanced by the presence of hydrophobic alkyl
groups in the hybrid sol–gel matrix and utilization of ILs as non-
structural template compounds [15].
In the present work, ionic liquids containing dialkylimidazolium
cations with different hydrophobicities and organic or inorganic
anions have been investigated as structure-directing immobi-
lization additives for sol–gel entrapped Pseudomonas fluorescens
(Amano AK) lipase. The influence of immobilization conditions
and reaction media on catalytic efficiency of the immobilized
preparates was also studied in ILs, compared to common organic
solvents.
2. Experimental
2.1. Materials
Lipase from Pseudomonas fluorescens (Amano Lipase AK) was
purchased from Aldrich. Silane precursors propyl- (PrTMOS),
octyl- (OcTMOS), and phenyl-trimethoxysilane (PhTMOS), as well
as N,N-dimethylformamide (DMF) were purchased from Fluka.
Methyl-trimethoxysilane (MeTMOS), tetramethoxysilane (TMOS),
n-hexane, 2-propanol, acetonitrile, tetrahydrofuran (THF), toluene,
methyl t-butyl ether (MTBE), sodium fluoride, vinyl acetate, 2-
hexanol, and 2-heptanol were from Merck. All the specified
reagents were of analytical grade and have been used as purchased.
Dodecane (99%, Merck) and decane (>99%, Aldrich) were used as
internal standards for the gas-chromatographic analysis.
Fig. 1. Scheme of immobilization methods used.
tion 0.1 M, pH 8.0 (10 mL), isopropyl alcohol again (10 mL) and
hexane (10 mL). The sol–gel encapsulated enzyme was dried in a
vacuum oven at 25 ◦C for 8 h, crushed in a mortar and kept in a
closed vessel in the refrigerator.
Ionic
roborate [Emim]BF4, 1-ethyl-3-methyl-imidazolium acetate
[Emim][COOCH3], 1-ethyl-3-methyl-imidazolium trifluo-
roacetate [Emim][COOCF3], 1-propyl-3-methyl-imidazolium
liquids
1-ethyl-3-methyl-imidazolium
tetrafluo-
2.3. Immobilization Method 2, for preparates from silane
precursor MeTMOS
tetrafluoroborate [Pmim][BF4], 1-butyl-3-methyl-imidazolium
tetrafluoroborate [Bmim][BF4], 1-hexyl-3-methyl-imidazolium
tetrafluoroborate [Hmim][BF4], 1-butyl-3-methyl-imidazolium
This method, based on the methodology proposed by Orc¸ aire et
al. [18] was employed for preparates obtained from MeTMOS and
TMOS silane precursors, as those obtained according to Method 1
exhibited very low activities, probably resulted from a too dense
structure of the sol–gel. In a 4 mL vial 780 ␮L of lipase AK solution
in Tris/HCl buffer 0.1 M, pH 8.0, obtained as indicated in Method 1,
were mixed with 200 ␮L ionic liquid and magnetically stirred for
15 min. In a second 4 mL vial, the silane precursors (total 6 mmoles)
and ethanol (0.5 mL) were mixed for 15 min (magnetic stirring,
200 rot/min, room temperature). Subsequently, the two solutions
were mixed, and magnetically stirred until gelation occurred. The
resulting gel was kept at room temperature for 24 h to complete
polymerization. The bulk gel was washed and dried as described in
Method 1.
hexafluorophosphate
[Bmim][PF6],
1-butyl-3-methyl-
imidazolium bis(trifluoromethyl-sulfonyl)imide [Bmim][Tf2N],
were purchased from Merck at the highest available quality.
1-Octyl-3-methyl-imidazolium tetrafluoroborate [Omim][BF4]
was from Fluka.
2.2. Immobilization Method 1, for preparates from silane
precursors PrTMOS, OcTMOS and PhTMOS
The immobilization method previously described [16] and
based on the Reetz procedure [17] has been used, modified by
replacing polyethyleneglycol additive with ionic liquid, as indi-
cated in the scheme presented in Fig. 1. Lipase AK (100 mg/mL)
was suspended in Tris/HCl buffer solution 0.1 M, pH 8.0, stirred
at room temperature for 30 min, centrifuged 5 min at 25 ◦C and
3000 rot/min (Boeco U-320R centrifuge, Boeckel, Germany), and
780 ␮L of the supernatant (containing 74 mg dissolved enzyme)
have been used for immobilization. The immobilization was accom-
plished in a 4 mL glass vial, by mixing the lipase solution with
200 ␮L ionic liquid, 100 ␮L 1 M NaF aqueous solution and 200 ␮L
isopropyl alcohol (magnetic stirring, 200 rot/min). Silane precur-
sors (total 6 mmoles) were added, and mixing continued at room
temperature until the gelation started. The formed gel was kept
for 24 h at room temperature to complete polymerization. Subse-
quently, the bulk gel was washed to eliminate unreacted monomers
and additives with isopropyl alcohol (10 mL), Tris/HCl buffer solu-
2.4. Acylation of racemic secondary alcohols by vinyl acetate
Acylations have been made in 4 mL capacity glass vials, charged
with a mixture of secondary alcohol (2-hexanol or 2-heptanol,
1 mmole), vinyl acetate (3 mmole), reaction medium (organic sol-
vent or ionic liquid, 2 mL) and free (5 mg) or sol–gel immobilized
(25 mg) AK lipase. The biocatalyst amounts were chosen to ensure
an appropriate conversion during the selected reaction time (24 h).
The solvents, alcohols and ILs used were separately equilibrated
to 0.328 water activity at 25 ◦C over saturated MgCl2 solution for
48 h, as indicated by Bell et al. [19]. The mixture was stirred using
an orbital shaker (MIR-S100, Sanyo, Japan) at 300 strokes/min and
40 ◦C (ILW 115 STD incubator, Pol-Eko-Aparatura, Poland). Samples
taken at different time intervals were analyzed for conversion and

C. Zarcula et al. / Journal of Molecular Catalysis B: Enzymatic 65 (2010) 79–86
81
Fig. 2. Reaction scheme and structures of employed ionic liquids.
enantiomeric excess by a Varian 450-GC chromatograph equipped
with flame ionization detector, using a 30 m × 0.25 mm Elite-
Cyclosil B chiral column, 0.25 mm film thickness (PerkinElmer,
USA). The analysis conditions were set as follows: oven tempera-
ture: 50–120 ◦C with 10 ◦C/min heating rate, injector temperature
240 ◦C, detector temperature 280 ◦C, carrier gas (hydrogen) flow
1.2 mL/min. The conversions were calculated based on the internal
standard method, using the appropriate calibration curve for every
alcohol.
Transesterification activities were calculated based on the alco-
hol conversion at 24 h and expressed as the average amount of
formed 2-acetoxy-alcohol (in micromole) in 1 h interval by 1 mg
of free or immobilized enzyme. The reaction without enzyme did
not give any product in the same conditions. To characterize the
overall efficiency of the immobilization process, total activity yield
was calculated as % of the total enzymatic activity recovered fol-
lowing immobilization, related to the total activity of the lipase
subjected to immobilization. As only the enantiomers of the ester
products have been separated on this column, the enantiomeric
excess of the resulted ester product (eep) was determined based
on the enantiomers peak area. Accordingly, the enantiomeric ratio
(E) values were calculated based on conversion and eep values, as
proposed by Chen et al. [20].
the course of lipase immobilization by the sol–gel method when
ILs are used as additives.
To evaluate the catalytic efficiency of sol–gel immobilized lipase,
we used as model reaction the kinetic resolution of two secondary
alcohols, 2-hexanol and 2-heptanol, by acylation with vinyl acetate
(Fig. 2). Pseudomonas fluorescens lipase is not highly enantioselec-
tive in this process, therefore it was also possible to investigate the
effect of immobilization on the enantioselectivity. The time course
of the mentioned reaction was not influenced by sol–gel immobi-
lization. In case of the 2-heptanol substrate showed in Fig. 3, both
conversion and enantiomeric excess curves were similar for native
and immobilized lipase, in hexane reaction medium. The reaction
progress curve over the studied time period (48 h) was not lin-
ear, as it is usual for the majority of biocatalytic reactions. Our aim
was to characterize the catalytic performance of the immobilized
enzyme, therefore we calculated the transesterification activity
from a single-point experiment, based on the conversion at 24 h.
Calculation of initial rates based on the linear period of the reac-
tion to obtain kinetically valid activity values would have been
difficult for preparates with low activities, as were those obtained
with MeTMOS as precursor. Therefore, the term “activity” used in
the present work and expressed as ␮mole h−1 mg lipase−1, must be
considered only for comparative evaluation of catalytic efficiency,
not for the kinetic behavior of the biocatalyst. The transesterifica-
tion activity (related to 1 mg biocatalyst) was obviously higher for
the native enzyme, as the enzyme usually represents only 8–10%
(w/w) of the sol–gel preparate amount. To characterize the global
efficiency of the immobilization process, it is significant that the
All acylations have been run in duplicate and sampling was also
made in duplicate. Because the differences between the four data
points for the same assay were less than 2% for conversion and 1%
for the enantiomeric excess, average values have been calculated
and presented in the tables and figures.
2.5. Structural analysis
FT-IR spectra were recorded on a Prestige21 spectrophotome-
ter (Shimadzu, Japan) on 400–4000 cm−1 range at 4 cm−1 spectral
resolution, using the attenuated total reflectance (ATR) technique.
Scanning electron microscopy (SEM) coupled with energy disper-
sive X-ray spectroscopy (EDX) was performed with an Inspect
S + EDAX Genesis XM 2i system (FEI Company, The Netherlands).
The analysis parameters were: pressure 1.5 × 10−2 Pa, resolution
<10 nm at 3 kV.
3. Results and discussion
3.1. ILs as immobilization additives
Utilization of ILs as immobilization additives could have mul-
tiple influences on both the entrapment process and catalytic
efficiency of immobilized enzyme. The polarity of IL and its mis-
cibility with the silane precursors and water may influence the
homogeneity of the mixture subjected to the sol–gel process, gela-
tion characteristics, and structure of the formed hybrid sol–gel
matrix. As regards catalytic efficiency, Dang et al. [21] demon-
strated that pretreatment of lipase with ILs could result in enhanced
activity and stability and it is obvious to presume such an effect in
Fig. 3. Time course of 2-heptanol acylation reaction by vinyl acetate, in hexane
at 40 ◦C, catalyzed by Pseudomonas fluorescens lipase, native and immobilized by
the sol–gel method with OcTMOS and TMOS precursors (1:1 molar ratio) and
[Omim][BF4] as additive, at 0.322 IL/silane molar ratio. Conversion (C) curves, in
the lower part, and enantiomeric excess (eep) curves in the upper part of the dia-
gram were plotted only to illustrate the tendency of these parameters evolution,
without kinetic signification.

82
C. Zarcula et al. / Journal of Molecular Catalysis B: Enzymatic 65 (2010) 79–86
Table 1
Influence of IL additive structure and polarity on the catalytic efficiency of sol–gel entrapped lipase in the acylation of racemic 2-heptanol by vinyl acetate in hexane medium.
Immobilization was performed with OcTMOS and TMOS silane precursors at 1:1 molar ratio, IL/silane molar ratio was 0.129.
Ionic liquid
ETN
Conversione (%)
Activity (␮mole h−1 mg−1
)
Total activity yieldf (%)
eep (%)
E
Reference (native lipase)
[Emim][BF4]
[Pmim][BF4]
[Bmim][BF4]
[Hmim][BF4]
[Omim][BF4]
[Emim][COOCH3]
[Emim][COOCF3]
–
39
34
43
40
42
53
41
32
1.383
0.295
0.370
0.352
0.360
0.467
0.344
0.276
100
128
209
199
177
229
167
116
72
77
73
78
75
68
74
79
10
11
11
13
12
12
11
12
0.71a
0.69a
0.68a
0.70b
0.64c
0.60d
0.66d
a
From Ref. [24].
From Ref. [23].
From Ref. [13].
b
c
d
Estimations, based on the values for [Bmim]COOCH3 and [Bmim]COOCF3 from Ref. [23].
Measured at 24-h reaction time.
Total activity recovery yield following immobilization, as defined in Methods.
e
f
total activity recovered after immobilization was usually higher
than the total activity of the enzyme subjected to immobilization.
Particularly, for the reaction presented in Fig. 3, the transesterifica-
tion activities at 24 h were 1.60 and 0.35 ␮mole h−1 mg lipase−1 for
the native and immobilized lipase, respectively, resulting a total
activity yield value of 156%. The enantiomeric excess was influ-
enced by the conversion and decreased slowly during the reaction,
but the enantiomeric ratio E remained in the same range (9–14) for
both native and immobilized lipase.
Polarity is an important parameter influencing the interactions
of ILs with enzymes. There are many possibilities to evaluate the
polarity of ILs and organic solvents, defined as overall solvation
capability. We used the empirical scale of relative solvent polar-
ity ETN determined by the solvatochromic standard Reichardt’s dye.
This scale employs water (ETN 1.00) and tetramethylsilane (ETN 0.00)
as reference solvents for high and low polarity, respectively [22,23].
Using 2-heptanol as model substrate, the catalytic efficiency
of Amano AK lipase immobilized using IL as additive was among
20–34%, related to the native enzyme (Table 1). The total enzy-
matic activity increased next to immobilization for all ILs tested, as
we found total activity up to 2.3-fold higher compared to the native
enzyme.
Correlation of IL polarity with catalytic efficiency of the immo-
bilized lipase is difficult, because ETN values reported by various
groups for the same IL could differ considerably, probably influ-
enced by the remaining water content or traces of other polar
impurities. Nevertheless, it can be observed that increasing polarity
of the IL resulted in lower activities. In the series of ILs with 1-
alkyl-3-methyl-imidazolium cations and tetrafluoroborate anion,
the activity of the preparate increased with the length of the alkyl
group, being 58% higher in case of octyl than for ethyl. Lee et al.
[13] reported results with opposite tendency compared to ours,
as the specific activity of Candida rugosa lipase immobilized with
[Emim][BF4] additive was 8.7-fold higher than with [Omim][BF4],
in the acylation reaction of benzyl alcohol. Such differences could
be explained by the different precursors and immobilization proto-
cols employed. Conformational differences between the microbial
lipases used in the two investigations could also affect the immobi-
lized catalyst properties. However, our results are consistent with
the well-known fact that a more hydrophobic microenvironment
around the enzyme is beneficial for lipase activity. Among ILs with
the same cation [Emim], the highest activity was also measured for
the less polar [Emim][COOCH3].
more important changes of enantioselectivity in relation to the
silane precursor’s nature and relative concentration in the immo-
bilization mixture [25]. It seems that the silane structure has a
more significant function in this respect than the nature of IL
additive.
The influence of increasing amount of IL additive was studied
for the acylation reaction of 2-hexanol at 40 ◦C, in hexane reaction
medium. [Omim][BF4] has been used in molar ratios from 0.064 to
0.322, related to total silanes (Table 2). This IL is not miscible with
water, but miscible with the silanes used as precursors. Increas-
ing the relative amount of IL in the immobilization mixture allows
maximizing its positive effect as structure-directing additive dur-
ing the sol–gel process. The highest transesterification activities
were obtained for an IL/silane molar ratio around 0.2. The enantios-
electivity of the immobilized lipase was not strongly influenced by
the IL amount, but it was slightly improved related to the native
enzyme.
3.2. Catalytic properties of the sol–gel immobilized lipases in
ionic liquids and organic solvents
The biocatalytic properties of the sol–gel immobilized lipases
are strongly influenced by the nature and concentration of the non-
hydrolyzable alkyl or aryl group of the silane precursor (RTMOS)
used in combination with TMOS. Our previous results showed that
a 2:1 RTMOS/TMOS molar ratio was still accurate to ensure both
high activity and physical robustness of immobilized preparates
[25]. This influence has been studied for three different ILs used
as additives. The investigated model reaction was the acylation of
2-hexanol by vinyl acetate in hexane, at 40 ◦C.
Using silane precursors with different nonhydrolyzable groups
and different ILs as additives resulted in immobilized lipases with
activities ranging on a wide scale. Increasing the hydrophobic
groups density in the sol–gel matrix was generally beneficial for
Table 2
Influence of [Omim][BF4] additive amount on catalytic efficiency and enantios-
electivity of 2-hexanol acylation catalyzed by sol–gel immobilized lipase. Silane
precursors for immobilization were OcTMOS and TMOS (1:1 molar ratio), acylation
has been run in hexane.
IL/silane molar ratio
Conversion
(%)
Activity
eep (%)
E
(␮mole h−1 mg−1
)
Reference (native lipase) 29
1.217
0.411
0.494
0.591
0.548
0.419
78
75
63
49
57
75
11
16
13
15
14
14
We also expected possible changes of enantioselectivity induced
by the ionic liquid additive, but it was not the case. Although
the measured eep values of the product at 24-h reaction time
showed some fluctuations ( 5% from the mean value), the enan-
tiomeric ratio E values were almost the same. Previously, we found
0.064
0.129
0.193
0.257
0.322
50
59
67
63
47

C. Zarcula et al. / Journal of Molecular Catalysis B: Enzymatic 65 (2010) 79–86
83
the formation of the porous structure (Fig. 4b). Increasing the
hydrophobic group density in the sol–gel matrix (2:1 RTMOS/TMOS
molar ratio), the maximum immobilization efficiency shifted to
the preparate obtained with the PrTMOS precursor. Considering
the biocatalysts obtained with OcTMOS precursor and different IL
additives, at this molar ratio the highest efficiency was assayed for
[Bmim][BF4], with a shorter alkyl chain. It results that the optimal
content of hydrophobic groups in the silica matrix must be tuned
carefully, considering all the components of this system as well as
the structure of substrate.
The influence of reaction medium on catalytic efficiency of the
lipase immobilized by the sol–gel method with IL additive was
investigated in various ILs, as well as in common organic sol-
vents. The immobilized preparates were obtained from OcTMOS
and TMOS precursors with [Omim][BF4] as IL additive. Selection
of a suitable reaction medium is a difficult issue in biocatalysis
because some organic solvents, commonly used in organic synthe-
ses, can inactivate the enzymes. As regards lipases, the enzymes
with the largest number of synthetic applications, it was demon-
strated an overall trend toward higher activities in hydrophobic
reaction media. The explanation of this phenomenon is related
to nonpolar interactions leading to the displacement of a helical
oligopeptide unit called “lid” which covers the active center, facili-
tating an easier access of the substrate [26]. Lipase immobilization
by the sol–gel method with IL as additive did not change this ten-
dency, although it was not possible to obtain a direct correlation
between the activities and solvent polarities on the Reichardt nor-
malized scale (Table 3). The highest conversion and activity at 24-h
reaction time was achieved in hexane and the lowest in acetoni-
trile. The reduced activity of hydrolases in water-miscible organic
solvents was explained as well by the capability of these com-
pounds to strip the tightly bound water layer from the enzyme
molecule, which is essential to maintain the catalytic activity [27].
In our study, the solvents were equilibrated before the reaction to
a constant water activity, therefore we can presume that the men-
tioned essential water layer was not disturbed and consequently
the enzyme activity remained quite high also in water-miscible
solvents as acetone or THF. Hara et al. [28] associated the lower
activity of Pseudomonas cepacia lipase in more hydrophilic solvents
with the competing hydrolysis reaction of the ester and acylation
agent by the water present in system. We used 3:1 molar excess of
vinyl acetate to overcome negative effects of possible hydrolysis of
the acylation agent. Although slight hydrolysis of the ester product
cannot be excluded, it was not significant since comparable conver-
sions in water-miscible and water-immiscible solvents have been
achieved. The absence of activity in DMF was not caused by exces-
sive polarity, it could be attributed to the well-known denaturating
effect of some solvents as DMF or dimethylsulfoxide on synthetic
activity of lipases and other hydrolases [24,29].
Fig. 4. Influence of the nonhydrolyzable group of silane precursor and ionic liquid
structure on the relative total activity recovered after sol–gel immobilization of
Pseudomonas fluorescens lipase, at (a) 1:1, and (b) 2:1 RTMOS/TMOS molar ratio.
Acylation of 2-hexanol was made by vinyl acetate at 40 ◦C, in hexane.
lipase immobilization efficiency, expressed as total activity recov-
ery yield (Fig. 4).
At 1:1 precursor molar ratio, the tendency was the same for
every ionic liquid tested as additive, namely the activity increased
with the alkyl group chain length from methyl to octyl (Fig. 4a).
Using silane precursor with phenyl instead of alkyl nonhydrolyz-
able group, the activity of the best preparate was comparable,
but the gelation process was usually slowed down and the repro-
ducibility of immobilization results was uncertain. At 1:1 silane
precursor molar ratio a more hydrophobic substituent in both
silane precursor and cationic part of IL additive is needed to obtain
a sol–gel immobilized lipase with increased activity, [Omim][BF4]
being the additive with the highest improvement of catalytic effi-
ciency. However, using alkyltrimethoxysilanes with large alkyl
groups at higher concentration could be detrimental for the immo-
bilization process by decreasing the gelation rate and hindering
The enantioselectivity of kinetic resolution 2-hexanol was
strongly influenced by the nature of reaction medium. Acetone and
THF, water-miscible solvents with intermediate polarity, demon-
strated high efficiency in this respect.
Table 3
Influence of organic solvent nature and polarity on catalytic efficiency of 2-hexanol acylation by sol–gel immobilized lipase. Silane precursors for immobilization were
OcTMOS and TMOS (1:1 molar ratio), [Omim][BF4] was employed as additive at 0.13 IL/silane molar ratio.
Reaction medium (organic solvent)
ETN
Conversion (%)
Activity (␮mole h−1 mg−1
)
eep (%)
E
Hexane
Toluene
MTBE
THF
Acetone
DMF
0.009a
0.099a
0.124a
0.207a
0.355a
0.405a
0.460a
59
49
56
43
47
0
0.494
0.397
0.458
0.339
0.395
0
63
72
77
92
91
0
13
13
34
50
53
0
Acetonitrile
33
0.273
83
16
a
From Ref. [22].

84
C. Zarcula et al. / Journal of Molecular Catalysis B: Enzymatic 65 (2010) 79–86
Table 4
Influence of IL reaction medium structure and polarity on catalytic efficiency of 2-hexanol acylation by sol–gel immobilized lipase. Silane precursors for immobilization were
OcTMOS and TMOS (1:1 molar ratio), [Omim][BF4] was employed as additive at 0.13 IL/silane molar ratio.
Reaction medium (ionic liquid)
ETN
Conversion (%)
Activity (␮mol h−1 mg−1
)
eep (%)
E
[Emim][BF4]
[Pmim][BF4]
[Bmim][BF4]
[Omim][BF4]
[Bmim][Tf2N]
[Bmim][PF6]
[Omim][PF6]
0.71a
0.69a
0.68a
0.64b
0.64c
0.67c
0.63c
72
59
38
36
48
49
61
0.662
0.484
0.316
0.314
0.392
0.412
0.511
38
56
78
84
62
32
42
9
8
13
18
8
4
5
a
From Ref. [21].
From Ref. [13].
From Ref. [23].
b
c
ILs might be an attractive solution as polar reaction media, par-
ticularly to achieve high conversion values for substrates that are
not soluble in solvents that usually allow high lipase activity. As
discussed, the synthetic activity of lipases is usually lower in polar
solvents, but in ILs could be different due to the specific proper-
ties of these compounds. The nature of cation and anion are both
important in this topic. Seven ILs have been tested as reaction media
for the acylation of 2-hexanol by vinyl acetate. The polarity of these
compounds ranked from 0.71 to 0.63, at considerably higher values
compared to acetonitrile, the most polar organic solvent previously
studied. In the series of ILs with [BF4] anion, the transesterification
activity of the sol–gel immobilized lipase increased with increas-
ing polarity of the IL and decreased with the alkyl chain length of
imidazolium cation (Table 4). Such an apparently strange behavior
of lipases in ILs based on 1-alkyl-3-methyl-imidazolium cation and
[BF4] anion was also reported by De Diego et al. [30] for the acy-
lation of 1-butanol by vinyl propionate catalyzed by commercially
available immobilized lipases. At contrary, Lee et al. [31], using the
same enzymes and ILs with [Tf2N] anion, measured the highest ini-
tial rate for the acylation of benzyl alcohol by vinyl acetate for the
more hydrophobic [Omim][Tf2N].
Our results show that enantioselectivity of Pseudomonas fluo-
rescens lipase in the acylation reaction of 2-hexanol was lower
in the best IL than in the best organic solvents, THF and acetone,
but was comparable or higher compared to other organic solvents
(Tables 3 and 4). Probably, by interactions between the electri-
cally charged ions from the reaction medium and specific residues
situated close to the active center may affect the conformational
rigidity of the enzyme molecule, resulting in diminution of the
kinetic discrimination between the two isomers. For this type of
substrates and enzyme, utilization of ILs is not recommended when
the main target is the synthesis of products with high optical purity,
organic solvents still remaining the best choice. We could not set
up a direct correlation between the IL structure and the enantios-
electivity of lipase, but for ILs with [BF4] anion the enantiomeric
ratio values slightly increased at higher hydrophobicity of the alkyl
substituent from the imidazolium cation.
For ILs with the same cation ([Bmim] or [Omim], Table 4), the
activity increased with decreasing polarity. In this case, the anion’s
nucleophilicity seems to have a more important role than the reac-
tion medium polarity. Probably, ILs with increased nucleophilicity
shall coordinate more strongly to positively charged sites at the
enzyme surface, leading to conformational changes and decrease
of enzyme activity, as it was firstly suggested by Sheldon et al. [32].
The [BF4] anion is more nucleophilic than [PF6] and might induce
lower activities, as confirmed by our experimental results.
The observed discrepancies concerning the influence of IL struc-
ture on lipase activity could be also attributed to the different
microenvironments in the vicinity of enzyme active center, deter-
mined by the combined effect of immobilization support, reaction
medium, and substrate structure. The purities of ILs used in the
experiments, especially the water content, and their purification
method are very important, as well. We used commercially avail-
able ILs at the highest possible purity and from the same supplier,
but itisdifficulttocomparetheconditions withthosereported else-
where. For the acylation reaction of 2-hexanol by vinyl acetate, the
best IL reaction medium was [Omim]PF6, yielding higher activities
than the best organic solvent.
The influence of ILs as reaction media on the enantioselectivity
of lipases is also controversial. Enhancement of enantioselectivity
in the resolution of chiral alcohols was reported in some cases. Kim
et al. [33] obtained an important increase of the enantiomeric ratio
E in ILs, principally in [Bmim]PF6, but their study was made for
reactions with very high enantioselectivities in organic solvents,
too. Substantially increased enantioselectivities in ILs compared
to organic solvents have been reported by Schöfer et al. [34] for
lipase from Alcaligenes sp., but they tested only one organic solvent
(MTBE).
Fig. 5. SEM image (a) and EDX spectrum (b) of the sol–gel entrapped AK lipase
obtained from OcTMOS/TMOS precursors at 1:1 molar ratio.

C. Zarcula et al. / Journal of Molecular Catalysis B: Enzymatic 65 (2010) 79–86
85
3.3. Structural characteristics of preparates immobilized with
ionic liquid additive
Detailed structural investigation of the sol–gel immobilized
lipase with ILs as additives was not the aim of this study. We
only attempted to identify how the presence of ILs can modify the
morphology of the sol–gel preparates and to have a possible con-
firmation of the partial incorporation of the IL in the silica matrix.
Encapsulating microbial lipases in hybrid inorganic–organic
matrices with significant content of hydrophobic groups could
induce important differences compared to the porous structure
characteristic for inorganic sol–gel materials. SEM images can pro-
vide useful information regarding the surface morphology, and EDX
spectra allows qualitative and semi-quantitative identification of
the elements present in the sample.
The SEM image of the lipase immobilized using OcTMOS and
TMOS as silane precursors without IL additive (Fig. 5a) shows an
irregular porous structure, with spherical nanoparticles. In the
EDX spectrum (Fig. 5b) the presence of F and Na elements can be
observed, indicating that at least a part of the NaF catalyst was
not eliminated during the gel washing step. The presence of F
might be also a consequence of [BF4]− or [PF6]− anions decompo-
sition in the presence of moisture, even though they not react with
water [35,36]. The C/Si relative ratio (w/w) in the silane precursors
Fig. 7. FT-IR spectra of sol–gel entrapped preparates obtained from MeTMOS and
TMOS precursors at 1:1 molar ratio (a) without, and (b) with [Omim]BF4 as additive.
employed for this immobilization was 2.9. The semi-quantitative
analysis of the EDX spectrum gives a very close value (2.8) of this
ratio, suggesting that in our case the hydrophobic groups were uni-
formly distributed in the sol–gel matrix, not concentrated at the
surface. Adding an IL additive during the immobilization process
resulted in important modifications of the general surface morphol-
ogy of the biocatalyst. The SEM image shows a more amorphous
structure (Fig. 6a), and in the elemental composition resulted from
the EDX spectrum the boron element can be identified (Fig. 6b). It
results that a part of the ionic liquid used as additive [Omim]BF4
was confined in the sol–gel structure. The semi-quantitative ele-
mental analysis gives a much lower C/Si ratio than previously (1.6),
indicating a different distribution of hydrophobic groups in the
surface region of the sol–gel matrix.
The inclusion of a part of the IL additive in the sol–gel mate-
rial during the immobilization process was also confirmed by
FT-IR spectroscopy (Fig. 7). We selected preparates obtained using
MeTMOS as one of the silane precursors, to demonstrate the
incorporation of the additive by the presence of the symmetri-
cal and asymmetrical stretching bands of methylene groups at
2858 cm−1 and 2928 cm−1, that cannot be found in the matrix
obtained without [Omim][BF4]. More specific are the symmetri-
cal and asymmetrical stretching vibrational bands of imidazolium
skeleton, which occur at 1463 cm−1 and 1562 cm−1, respectively.
The attribution of the mentioned IR vibrational bands was made
according to literature data [37].
The presented results confirm that the IL remained partially con-
fined in the sol–gel structure, though the gel was carefully washed
with aqueous buffer, isopropanol, and hexane. The exact nature
of interaction between the IL and the sol–gel matrix needs more
sophisticated techniques. Shi and Deng [38] reported about physi-
cal confinement of ILs in nano-porous silica gel during the sol–gel
process and nano-scale dispersion of the IL within the matrix. It
looks feasible that a similar process takes place in the course of
enzyme immobilization, but it could be strongly influenced by
interactions between the enzyme and IL.
4. Conclusions
Immobilization of lipase in hybrid sol–gel matrices using spe-
cific ILs as additives resulted in biocatalysts with high catalytic
efficiency. The recovery yield of total activity following immobiliza-
tion was usually higher than 100%, the best value being recorded
for [Omim]BF4 as additive. Kinetic resolution of secondary alcohols
is an important aim in biocatalysis. Although native Pseudomonas
Fig. 6. SEM image (a) and EDX spectrum (b) of the sol–gel entrapped AK lipase
obtained from OcTMOS/TMOS precursors and [Omim]BF4 as additive.

86
C. Zarcula et al. / Journal of Molecular Catalysis B: Enzymatic 65 (2010) 79–86
fluorescens lipase is not highly enantioselective for this process, we
demonstrated that using the sol–gel entrapped biocatalyst and an
appropriate selection of the reaction medium E values higher than
50 can be achieved for the acylation of 2-hexanol. Considering this
reaction, ILs are better reaction media than organic solvents, but
with lower enantioselectivity.
[11] V.B. Kandimalla, V.S. Tripathi, H. Ju, Crit. Rev. Anal. Chem. 36 (2006) 73–106.
[12] A.C. Pierre, Biocatal. Biotransform. 22 (2004) 145–170.
[13] S.H. Lee, T.T.N. Doan, S.H. Ha, Y.-M. Koo, J. Mol. Catal. B: Enzym. 45 (2006)
57–61.
[14] A. Karout, A.C. Pierre, J. Non-Cryst. Solids 353 (2007) 2900–2909.
[15] F. Peter, C. Zarcula, S. Kakasi-Zsurka, R. Croitoru, C. Davidescu, C. Csunderlik, J.
Biotechnol. 136S (2008) S374.
[16] F. Peter, L. Poppe, C. Kiss, E. Szo˝ cs-Bíro, G. Preda, C. Zarcula, A. Olteanu, Biocatal.
Biotransform. 23 (2005) 251–260.
[17] M.T. Reetz, P. Tielmann, W. Wiesenhöfer, W. Könen, A. Zonta, Adv. Synth. Catal.
345 (2003) 717–728.
Acknowledgements
[18] O. Orc¸ aire, P. Buisson, A.C. Pierre, J. Mol. Catal. B: Enzym. 42 (2006) 106–113.
[19] G. Bell, P.J. Halling, L. May, B.D. Moore, D.A. Robb, R. Ulijn, R.H. Valivety, in: E.N.
Vulfson, P.J. Halling, H.L. Holland (Eds.), Methods in Biotechnology: Enzymes
in Nonaqueous Solvents, Humana Press, Totowa, NJ, 2001, pp. 105–126.
[20] C.-S. Chen, Y. Fujimoto, G. Girdauskas, C.J. Sih, J. Am. Chem. Soc. 104 (1982)
7294–7299.
[21] D.T. Dang, S.H. Ha, S.-M. Lee, W.-J. Chang, Y.-M. Koo, J. Mol. Catal. B: Enzym. 45
(2006) 57–61.
[22] C. Reichardt, Chem. Rev. 94 (1994) 2319–2358.
Financial support of research was provided by CNCSIS through
PNII-IDEI grant 368/2007. This work was partially supported by the
strategic grants POSDRU 6/1.5/S/13, 2008 (fellowships of L. Corici
and R. Croitoru) and POSDRU 2009, project ID 50783 (fellowship
of A. Ursoiu), of the Ministry of Labour, Family and Social Protec-
tion, Romania, co-financed by the European Social Fund-Investing
in People. The authors acknowledge Ms. Paula Sfârloaga˘ (INCEMC,
Timis¸ oara) for the SEM analysis.
[23] C. Reichardt, Green Chem. 7 (2005) 339–351.
[24] S. Park, R.J. Kazlauskas, J. Org. Chem. 66 (2001) 8395–8401.
[25] C. Zarcula, R. Croitoru, L. Corîci, C. Csunderlik, F. Peter, Proceedings of the
World Academy of Science, Engineering and Technology, vol. 52, 2009, pp. 179–
184.
References
[26] M.T. Reetz, Curr. Opin. Biol. Chem. 6 (2002) 145–150.
[27] A.M. Klibanov, Nature 409 (2001) 241–246.
[28] P. Hara, U. Hanefeld, L.T. Kanerva, J. Mol. Catal. B: Enzym. 50 (2008) 80–86.
[29] A. Zaks, A.M. Klibanov, J. Biol. Chem. 263 (1988) 3194–3201.
[30] T. De Diego, P. Lozano, M.A. Abad, K. Steffensky, M. Vaultier, J.L. Iborra, J. Biotech-
nol. 140 (2009) 234–241.
[31] S.H. Lee, Y.-M. Koo, S.H. Ha, Korean J. Chem. Eng. 25 (2008) 1456–1462.
[32] R.A. Sheldon, R.M. Lau, M.J. Sorgedrager, F. van Rantwijk, K.R. Seddon, Green
Chem. 4 (2002) 147–151.
[33] K.-W. Kim, B. Song, M.-Y. Choi, M.-J. Kim, Org. Lett. 3 (2001) 1507–1509.
[34] S.H. Schöfer, N. Kaftzik, U. Kragl, P. Wasserscheid, Chem. Commun. (2001)
425–426.
[1] A. Illanes (Ed.), Enzyme Biocatalysis. Principles and Applications, Springer, Hei-
delberg, 2008.
[2] W.-D. Fessner, T. Anthonsen (Eds.), Modern Biocatalysis. Stereoselective and
Environmentally Friendly Reactions, Wiley–VCH, Weinheim, 2009.
[3] K.M. Koeller, C.-H. Wong, Nature 409 (2001) 232–240.
[4] M. Sureshkumar, C.-K. Lee, J. Mol. Catal. B: Enzym. 60 (2009) 1–12.
[5] F. van Rantwijk, R.M. Lau, R.M. Sheldon, Trends Biotechnol. 21 (2003) 131–138.
[6] N. Jain, A. Kumar, S. Chauhan, S.M.S. Chauhan, Tetrahedron 61 (2005)
1015–1060.
[7] L. Gubicza, N. Nemestóthy, T. Fráter, K. Bélafi-Bakó, Green Chem. 5 (2003)
236–239.
[35] H. Wiengärtner, Angew. Chem. Int. Ed. 47 (2008) 654–670.
[36] N.V. Plechkova, K.R. Seddon, Chem. Soc. Rev. 38 (2008) 123–150.
[37] N.E. Heimer, R.E. Del Sesto, Z. Meng, J.S. Wilkes, W.R. Carper, J. Mol. Liq. 124
(2006) 84–95.
[8] O. Ulbert, T. Fráter, K. Bélafi-Bakó, L. Gubicza, J. Mol. Catal. B: Enzym. 31 (2004)
39–45.
[9] U. Kragl, M. Eckstein, N. Kaftzik, Curr. Opin. Biotechnol. 13 (2002) 565–571.
[10] E. Fehér, V. Illeová, I. Kelemen-Horváth, K. Bélafi-Bakó, M. Polakovicˇ, L. Gubicza,
J. Mol. Catal. B: Enzym. 50 (2008) 28–32.
[38] F. Shi, Y. Deng, Spectrochim. Acta A 62 (2005) 239–244.