Enzymatic kinetic resolution of sec-alcohols using an ionic liquid anhydride as acylating agent

Article abstract of DOI:10.1016/j.tetasy.2014.05.001

A task-specific ionic liquid bearing an anhydride moiety was synthesized for the first time in good yield (83%) through a carbodiimide-mediated coupling reaction. The enantiomeric separation of a series of sec-alcohols was performed via enzymatic kinetic resolution, employing an ionic anhydride as acylating agent and Candida antarctica Lipase B as a biocatalyst. A fast and efficient recovery of both enantiomers was achieved separately due to the ionic nature of the acyl donor, combined with the possibility of carrying out the enzymatic step in an organic solvent.

Full text of DOI:10.1016/j.tetasy.2014.05.001

Tetrahedron: Asymmetry 25 (2014) 944–948
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Enzymatic kinetic resolution of sec-alcohols using an ionic liquid
anhydride as acylating agent
⇑
Raquel Teixeira, Nuno M. T. Lourenço
IBB—Institute for Biotechnology and Bioengineering, Centre for Biological and Chemical Engineering, Instituto Superior Técnico, Av. Rovisco Pais, 1, 1049-001 Lisboa, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
A task-specific ionic liquid bearing an anhydride moiety was synthesized for the first time in good yield
(83%) through a carbodiimide-mediated coupling reaction. The enantiomeric separation of a series of sec-
alcohols was performed via enzymatic kinetic resolution, employing an ionic anhydride as acylating
agent and Candida antarctica Lipase B as a biocatalyst. A fast and efficient recovery of both enantiomers
was achieved separately due to the ionic nature of the acyl donor, combined with the possibility of car-
rying out the enzymatic step in an organic solvent.
Received 27 March 2014
Accepted 9 May 2014
Available online 16 June 2014
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
obtained via enzymatic kinetic resolution.14 The lipase-catalyzed
acylation of sec-alcohols is widely employed to obtain enantiome-
In addition to ionic liquids being recognized as good solvents or
co-solvents in biocatalysis,1 they are also used for other purposes
such as enzyme coatings in ionic-liquid-coated enzymes endowed
with improved catalytic performances, or as phase transfer agents
in supported-ionic-liquid membranes.2 Moreover, the structure of
an ionic liquid can be modified in order to include a group of atoms
that confer an additional and specific function to the ionic liquid
itself. These are known as task-specific ionic liquids.3,4 Two exam-
ples of its application in biocatalysis include the work of Salunkhe
et al.5 who proposed that an ibuprofen-anchored ionic liquid can
be used as a substrate in enzymatic kinetic resolution to isolate
(S)-ibuprofen, and our own work on the use of ester ionic liquids
as promising acylating agents with which to perform lipase-cata-
lyzed kinetic resolutions in ionic liquid solvents.6,7
Lipases are ubiquitous in nature. They are versatile biocatalysts
used extensively in organic synthesis and in many industries (e.g.,
for foods, cosmetics and biofuel production).8–13 The enzymatic
kinetic resolution of racemic compounds is among the principal
transformations promoted by lipases. The success of the method
relies on the fact that enantiomers react at different rates with
the chiral active site of the biocatalyst. Furthermore, it allows us
to recover both enantiomers of a racemic mixture. The interest in
obtaining enantiomerically pure compounds is clear since they
are key intermediates in the production of fine chemicals. Many
chiral intermediates used in the pharmaceutical industry are
rically pure or enriched forms of different compounds.13,15
Some lipase-catalyzed reactions have to be carried out in non-
aqueous media, either to prevent reverse hydrolysis of the
products or to bypass solubility issues. Lipases have shown good
activity and stability in organic solvents and in ionic liquids. Even
so, thermodynamic aspects can present a problem to enzymatic
reactions carried out in non-aqueous solvents (e.g. if nucleophiles
are released as reaction by-products).
Herein we report the synthesis of a new task-specific ionic
liquid bearing an anhydride moiety for enzyme recognition and
its successful application as an acyl donor in the enzymatic resolu-
tion of sec-alcohols. Acyl-transfer reactions become irreversible
when anhydrides are employed as acylating agents because no
nucleophiles are released throughout the reaction. It also avoids
the use of another ionic liquid as a solvent to undergo vacuum
extraction of the by-products, which is mandatory with acylating
ionic esters.6
2. Results and discussion
Ionic anhydride 3 was obtained from ester 1 (Scheme 1). Ester 1
was first hydrolyzed to yield the ionic acid 2 (Step a), which was
further converted into one-half equivalent of 3 by means of a
carbodiimide-mediated coupling (Step b). The possibility of under-
taking the synthesis of the anhydride under mild conditions, such
as those afforded by the carbodiimide-coupling, is fundamental
because it allows us to introduce the anhydride functionality in
the last step of the synthesis. That, in turn, can assure a straightfor-
ward regeneration of the acylating agent.
⇑
Corresponding author. Tel.: +351 21 841 91 37 (065); fax: +351 21 841 90 62.
E-mail address: nmtl@tecnico.ulisboa.pt (N.M.T. Lourenço).
http://dx.doi.org/10.1016/j.tetasy.2014.05.001
0957-4166/Ó 2014 Elsevier Ltd. All rights reserved.

R. Teixeira, N. M. T. Lourenço / Tetrahedron: Asymmetry 25 (2014) 944–948
945
OH
PF6
PF6
PF6
a)
OH
N
N
O
N
N
R
N
N
rac
9
9
O
9
O
DIPU
DIC
O
1
O
a) CALB, acetone
2
, 93%
2
3
b) ether extraction
OH
a) 2 eq. KOH/MeOH, r.t., 3h
b)
b) 0.5 eq. DIC, dry CH2Cl2, r.t., 16h
(
)
S
R
PF6
PF6
PF6
N
O
N
N
N
O
N
N
OH
9
+ 2 (1 eq.)
9
9
O
O
R
O
2
(2 eq.)
2
4
3
, 83%
Scheme 1. Synthesis of the ionic liquid anhydride 3, used as acyl donor in the
enzymatic kinetic resolution of sec-alcohols (DIC = N,N0-diisopropylcarbodiimide).
OH
c) KOH/MeOH
d) ether extraction
R
(R)
Previous to the synthesis of anhydride 3 according to Scheme 1,
we failed in several attempts to prepare an ionic anhydride from
the bromide analogue of 2, which was likely due to the very low
solubility of that compound in suitable organic solvents. Thus,
we decided to replace the bromide with another counter-anion,
in order to reduce the polarity of the ionic acid. Among the anions
that reportedly do not affect lipase activity (BFꢀ4 , PF6ꢀ, and
NTFꢀ2 ),16,17 we chose PF6ꢀ based on our previous experience in the
enzymatic kinetic resolution of sec-alcohols with CALB and
imidazolium cation-based ionic liquids.6 By replacing the bromide
with PFꢀ6 , we solved our solubility issues and managed to obtain
the anhydride under mild conditions and in good yield.
Scheme 2. Methodology of the enzymatic kinetic resolution and separation of the
enantiomers of sec-alcohols. CALB was used as the biocatalyst and ionic anhydride 3
as the acylating agent (DIPU = N,N0-diisopropylurea).
whereas hydrophobic solvents are favorable to enzymatic reac-
tions.19 Acetone, a mildly polar solvent, has previously been shown
to provide good results in CALB-catalyzed reactions.20,21
The main resolution step in our enzymatic kinetic resolutions is
the enzymatic reaction (Step a, Scheme 2), wherein the biocatalyst
selectively transforms one of the enantiomers into a product,
which is a new chemical entity with distinct physico-chemical
properties from the initial substrate, leaving the other enantiomer
unchanged. In this step, we used one equivalent of acylating agent
to react with the racemic alcohol, which represents a 2:1 ratio of
acylating agent to the fast-reacting enantiomer. Ionic anhydride
3 was then employed in the enzymatic kinetic resolution of the
aromatic, aliphatic, and allylic sec-alcohols presented in Figure 1.
The length of the alkyl chain linking the anhydride group to
both ionic moieties (C11) was also chosen based on the previously
observed effects of the ionic group on the active center of the
enzyme.7 Therein we showed that long alkyl-chain acyl donors
give better results in the resolution of 1-phenylethanol, our model
molecule. We also attempted the enzymatic kinetic resolution of
substrate 5 with a shorter-chain anhydride (C7, data not shown),
but we observed no catalytic activity. Ionic liquids are favorable
solvents with regard to both medium recycling possibility and
the catalytic activity of lipases. Nonetheless, enzymatic kinetic res-
olution reactions are often very slow when ionic liquids are used as
reaction media (taking from hours to days), which might be due to
mass transfer limitations caused by their high viscosity.18 By using
an anhydride-functionalized ionic acylating agent to carry out the
enzymatic kinetic resolution of sec-alcohols, it is possible to obtain
both enantiomers separately by means of two simple ether extrac-
tions and by using an ‘enzyme-friendly’ organic solvent as the reac-
tion medium (Scheme 2). One extraction is performed after the
enantioselective lipase-catalyzed esterification to recover the
(S)-enriched form of the sec-alcohol while the other one affords
the remaining (R)-enriched sec-alcohol; the latter is performed
after chemical hydrolysis of the chiral ester produced in the
enzymatic reaction. The recovery of enantiomerically enriched
mixtures or pure enantiomers via solvent extraction is only possi-
ble due to the different physico-chemical properties of the reaction
components; again, we are able to take advantage of the unique
properties of task-specific ionic liquids, providing an ionic phase
where the fast-reacting enantiomers can be trapped as ionic esters.
Separation of the enantiomers via extraction has the advantage of
being more expeditious and energy effective than distillation, as
well as requiring lesser amounts of organic solvents than chro-
matographic separations.
OH
OH
OH
5
6
7
1-cyclohexylethanol
1-phenylethanol
4-phenyl-2-butanol
OH
OH
OH
9
8
10
2-pentanol
2-octanol
6-methyl-5-hepten-2-ol
Figure 1. Secondary alcohols employed in the enzymatic kinetic resolutions
catalyzed by CALB, using 3 as the acyl donor.
After Step a, the immobilized enzyme was removed by filtration
and the solvent was evaporated under reduced pressure, to give in
solution a mixture of ionic compounds (carboxylic acid, ester, and
non-reacted anhydride) and the unreacted alcohol. CALB displayed
high selectivity toward the (R)-enantiomers of the sec-alcohols.
Therefore, the non-ionic phase was mainly composed of the (S)-
enantiomer, which could be extracted with diethyl ether (Step b,
Scheme 2). The product of the enzymatic esterification was the
(R)-enriched ionic ester 4, which had to be chemically hydrolyzed
(Step c, Scheme 2) to give the (R)-sec-alcohol, after ether extraction
(Step d, Scheme 2). Finally, the two equivalents of ionic acid
All of the lipase-catalyzed acylations were performed in distilled
acetone. It is known that solvent effects play a determinant role in
CALB catalytic performance and that, in general, polar solvents
may displace the essential water causing enzyme inactivation,

946
R. Teixeira, N. M. T. Lourenço / Tetrahedron: Asymmetry 25 (2014) 944–948
recovered in the ionic phase were coupled together with N,N0-di-
isopropylcarbodiimide to regenerate 3. As discussed previously,
this step is very simple. It can be carried out in acetone and the
only by-product generated is the corresponding urea, which pre-
cipitates from this solvent.
Increasing the length of the alkyl chain that links the asymmet-
ric carbon to the aromatic ring, in 4-phenyl-2-butanol 6, resulted
in a decrease in the enantiomeric ratio but, at the same time led
to a slight increase in the conversion (entry 2, Table 1). Compound
(S)-6 was isolated in 53% yield and with 88% ee, while (R)-6 was
isolated in 43% yield and with 96% ee.
In the first set of experiments performed, the enantiomerically
enriched alcohols were obtained in low yields, which was ascribed
to the non-specific adsorption of alcohol enantiomers and/or ionic
ester to the acrylic support of the enzyme. It is known from the lit-
erature that non-specific adsorption of substrates to CALB support
might occur in the Novozym 435Ò formulation.21,22 Moreover, it is
known that ionic liquids, such as [bmim][PF6], are likely to interact
with PMMA-based resins, resulting in blends23 or ion gels24 that
display high ionic conductivity. With Novozym 435Ò one might
not expect that dissolution of the support would occur because
the resin is cross-linked and the ionic liquid in solution will not
dissolve it. Nonetheless, these findings shed some light into the
loss of substrates and products during enzymatic reactions with
Novozym 435Ò. To circumvent this putative problem, we sought
a molecule capable of blocking the sites available for non-specific
interactions. n-Hexane was the first molecule tested as a solvent
where Novozym 435Ò-catalyzed enantiomeric resolutions were
carried out with success.25 For the same reason, toluene was also
tested. The adsorption of apolar hexane or aromatic toluene on
the CALB support is likely to cause repulsion of polar and/or ali-
phatic molecules, thus preventing retention and enhancing mass
transfer. In fact, we observed an improvement both in product
yields and enantiomeric excess (ee) obtained after the enzyme
pre-treatment with n-hexane (entries 1, 2, and 4, results between
brackets, Table 1). After this set of experiments, all the presented
results were obtained upon the enzyme pre-treatment with n-hex-
ane (Table 1).
The enzymatic kinetic resolutions performed with CALB and
ionic anhydride 3 resulted in very high enantiomeric ratios (E)
for all of substrates investigated. The reaction times required to
carry out the efficient resolutions of these sec-alcohols were
remarkably short, especially for the linear aliphatic and allylic ones
(entries 4–6, Table 1). With the exception of alcohol 7 (entry 3,
Table 1), whose enzymatic acylation took almost 2 days, all of
the enzymatic reactions were carried out within 2.5–14 h.
Among the secondary benzylic alcohols, rac-1-phenylethanol 5
is commonly used as a model substrate for enzymatic resolutions.
The enzymatic kinetic resolution of this alcohol, following
Scheme 1, resulted in a good separation of both enantiomers. Its
enzymatic acylation took 13 h, after which (S)-5 was obtained in
58% yield and with an ee of 80%. After chemical hydrolysis of the
corresponding ionic ester, (R)-5 was isolated in 40% yield, and with
an ee higher than 99%.
The enzymatic kinetic resolution of 1-cyclohexylethanol 7 was
much slower than the other enzymatic acylations, but it was also
very efficient, providing both enantiomers with an enantiomeric
purity of higher than 90% (entry 3, Table 1). Likewise, the conver-
sion and enantiomeric ratios were very good.
Enzymatic resolutions of alcohols 5 and 7 with CALB and other
lipases have been reported by Csajági et al.26 These were per-
formed both in continuous-flow and batch reactions, using vinyl
acetate as the acylating agent. The authors reported that CALB pre-
sents the highest enantioselectivity for these substrates, in both
processes. Comparing their results with our batch process, our
strategy provides a lower ee for (S)-5 (80% vs 99%26), whereas for
enantiomers (R)-5 and (R)-7 we obtained comparable ee values.
On the other hand, our process gave a higher conversion in the res-
olution of 7 (48% vs 44%26), which indicates that we were able to
isolate (S)-7 with higher enantiomeric purity. However, it should
be noted that besides the different acyl donor, their experimental
conditions (e.g. ratio of subtrates/enzyme, temperature and reac-
tion times) were also different from ours. Another important
aspect to be considered is how readily the enantiomers can be
recovered after enzymatic resolution. In our method, both enanti-
omers can be obtained in good yields after one chemical hydrolysis
and two simple ether extractions, whereas in the aforementioned
process, the enantiomers need to be separated via chromatography
and, in some cases, are isolated in very low yields.
Enzymatic acylation reactions with the linear aliphatic com-
pounds 8 and 9 are much faster (5 h and 3 h, respectively), and
all of the enantiomers are obtained with very high enantiomeric
purity (P93%, entries 4 and 5 in Table 1). Accordingly, conversions
are close to the maximum attainable value (P49%), and the enanti-
oselectivity of the immobilized CALB toward these substrates is
also excellent (E >200).
Allylic sec-alcohol 10, also known as sulcatol, is an insect pher-
omone and an important intermediate in the synthesis of other
natural products.27 It can be resolved enzymatically from its race-
mic mixture even faster than 8 and 9, as can be seen in Table 1
(entry 6). Compound (S)-10 was obtained with an enantiomeric
excess of 84% in 56% yield, whereas (R)-10 was produced with a
higher purity (97%), but lower yield (40%).
In comparison with other resolution methodologies described in
the literature, the main advantage of our approach is the use of two
simple ionic liquid-ether extractions to obtain both enantiomers
Table 1
Enzymatic kinetic resolution and separation of the sec-alcohol enantiomers using CALB as the biocatalyst and ionic anhydride 3 as the acylating agent
Entry
Alcohol
(S)-Enantiomera
(R)-Enantiomerd
ce (%)
Ef
Time (h)
eeb (%)
Yieldc (%)
eeb (%)
Yieldc (%)
1
2
3
4
5
6
5
6
7
8
9
10
13
14
43
5
3
2.5
80(74)
88(75)
91
95(92)
93
58(56)
53(40)
44
42(45)
51
>99(99)
96(>99)
99
98(93)
97
40(28)
43(13)
26
36(32)
31
45
48
48
50
49
46
>200
144
>200
>200
179
84
56
97
40
157
a
All enzymatic reactions were carried out at 35 °C in acetone (0.8 mL) with 0.41 mmol of sec-alcohol, 0.41 mmol of 3, and 20 mg of Novozym 435Ò pretreated with
n-hexane, between bracket results obtained with no pretreatment of Novozym 435Ò
.
b
Enantiomeric excess determined by chiral GC.
Isolated yields.
c
d
All chemical hydrolysis were carried out at 50 °C in 1 mL of a 4 M solution of KOH in methanol for 30 min.
e
ees
ees þeeP
c ¼
.
In½1ꢀcð1þeeP Þꢁ
In½1ꢀcð1ꢀeeP Þꢁ
f
E ¼
.

R. Teixeira, N. M. T. Lourenço / Tetrahedron: Asymmetry 25 (2014) 944–948
947
Table 2
anhydride synthesis was dried at reflux over CaH2. The acetone
for the enzymatic reactions was distilled prior to use. Immobilized
Candida antarctica lipase B, CAL B (Novozym 435Ò with 1–2% water
w/w and 7000 PLU/g) was a gift from Novo Nordisk Bioindustrial,
Spain. (1(11-Undecanoic acid)-3-methyl)imidazolium hexafluoro-
phosphate was prepared according to the literature.6 1H and 13C
NMR spectra were recorded on a Bruker Avance III 400 spectrom-
eter. NMR chemical shifts (d) are reported in parts per million
(ppm) relative to a residual peak of the solvent, d = 2.50 (1H) and
d = 39.52 (13C) for DMSO-d6. Coupling constants (J) are reported
in Hz. FT-IR spectra were recorded as KBr pellets using a JASCO
FT/IR 4100 spectrometer. Elemental analyses and DSC measure-
ments were conducted by REQUIMTE-DQ-FCT laboratory (Portu-
gal) and carried out with a Thermo Finnigan Flash EA 111 and
Setaram DSC 131 calorimeter at a scanning rate of 10 °C minꢀ1
respectively.
Enzymatic acylations with succinic anhydride
Entry
Alcohol
(S)-Enantiomera
(R)-Enantiomer
cc (%)
Ed
Time (h)
eeb (%)
eeb (%)
1
2
3
5
9
10
40
20
20
33
86
78
98
95
99
25
47
44
172
117
>200
a
All enzymatic reactions were carried out at 35 °C in acetone (0.8 mL) with
0.41 mmol of sec-alcohol, 0.41 mmol of succinic anhydride, and 20 mg of Novozym
435Ò
.
b
Enantiomeric excess determined by chiral GC.
c ¼
c
ees
.
eesþeeP
In½1ꢀcð1þeeP Þꢁ
In½1ꢀcð1ꢀeeP Þꢁ
d
E ¼
.
individually. In fact, this is similar to the strategy employed by other
authors in the enzymatic resolution of sec-alcohols with succinic
anhydride.28,29 Therein, the slow-reacting enantiomer is extracted
with ether from a basic aqueous solution, whereas the other enan-
tiomer is obtained after saponification of the remaining enantiopure
ester, through another ether extraction. In order to compare the effi-
ciency of both ionic and organic acylating anhydrides, we performed
the resolution of aromatic 5, linear aliphatic 9, and allylic 10 with
succinic anhydride (Table 2), under exactly the same conditions as
with the ionic acyl donor.
As can be observed in Table 2, ionic anhydride 3 appeared to be
much better than succinic anhydride in the resolution of the sec-
alcohol 5. It took 13 h to perform the enzymatic acylation to yield
the (S)-enantiomer with an ee of 80% and the (R)-enantiomer with
an ee higher than 99%, using the ionic anhydride. On the other hand,
during the enzymatic kinetic resolution with succinic anhydride, the
enzymatic reaction took 40 h to separate both enantiomers with a
much lower ee and conversion (25% versus 45% with the ionic anhy-
dride). Regarding the second alcohol investigated, 2-octanol 9, both
anhydrides gave good results, even though the enantiomeric purity
of both enantiomers was slightly higher in the resolutions with 3. In
this case, the competitive advantage of this ionic anhydride is that
the enzymatic reaction, which is the rate-determining step of the
resolution and separation process, is much faster (3 h vs 20 h with
succinic anhydride). Likewise, the ionic anhydride gave better
results than succinic anhydride in the resolution of sulcatol 10,
together with a much faster enzymatic acylation reaction.
The enantiomeric excess was calculated by GC analysis and the
enantiomers were identified by comparison to the standards. GC
analysis was performed in a GC-2010-Plus Shimadzu with FID detec-
tion and a Varian CP-CHIRASIL-DEX-CB (25 m ꢂ 0.25 mm ꢂ 0.25
lm)
column. Column flow (He): 1.0 mL/min; Injector: 250 °C; detector:
250 °C; split ratio:100. 1-phenylethanol—oven: 120 °C for 20 min,
ramp 15 °C/min to 180 °C, and 180 °C for 5 min, tR(dode-
cane) = 6.26 min, tR(S) = 10.84 min, tR(R) = 10.10 min; 4-phenyl-2-
butanol—oven: 120 °C for 20 min, ramp 5 °C/min to 180 °C, and
180 °C for 5 min, tR(dodecane) = 6.26 min, tR(S) = 20.98 min,
tR(R) = 21.65 min; 1-cyclohexylethanol—oven: 60 °C for 2 min,
ramp 5 °C/min to 120 °C, ramp 10 °C/min to 180 °C, and 180 °C for
5 min, tR(dodecane) = 15.66 min, tR(S) = 18.07 min, tR(R) = 18.00 min;
1-cyclohexylethylpropionate—oven: 60 °C for 2 min, ramp 5 °C/min
to 120 °C, ramp 10 °C/min to 180 °C, and 180 °C for 5 min, tR(dode-
cane) = 15.66 min, tR(S) = 20.14 min, tR(R) = 20.65 min; 2-pentanol—
oven: 50 °C for 1 min, ramp 5 °C/min to 120 °C, 120 °C for 1 min, ramp
1 °C/min to 180 °C, and 180 °C for 5 min, tR(dodecane) = 18.36 min,
tR(S) = 9.95 min, tR(R) = 9.76 min; 2-octanol—oven: 40 °C for 10 min,
ramp 1 °C/min to 100 °C, ramp 10 °C/min to 180 °C, and 180 °C for
5 min, tR(dodecane) = 64.67 min, tR(S) = 58.79 min, tR(R) = 58.48 min;
octan-2-yl propionate—oven: 40 °C for 10 min, ramp 1 °C/min to
100 °C, ramp 10 °C/min to 180 °C, and 180 °C for 5 min, tR(dode-
cane) = 64.67 min, tR(S) = 66.81 min, tR(R) = 70.17 min; sulcatol—
oven: 80 °C for 40 min, ramp 15 °C/min to 180 °C, 180 °C
for
20 min,
tR(dodecane) = 38.31 min,
tR(S) = 24.46 min,
tR(R) = 26.38 min.
3. Conclusion
4.2. Preparation of 1(11-undecanoic acid)-3-methyl)imidazolium
hexafluorophosfate 2
In conclusion, we have reported for the first time the synthesis
of an anhydride-functionalized task-specific ionic liquid and its
efficient application as an acylating agent in lipase-catalyzed
kinetic resolutions of sec-alcohols. It was observed that CALB
immobilized in a hydrophobic acrylic support (Novozym 435Ò) is
highly enantioselective toward either aromatic, aliphatic, or allylic
sec-alcohols when enzymatic kinetic resolutions are performed in
acetone using the ionic liquid anhydride presented herein as the
acyl donor. Mild reaction conditions and high stereoselectivity,
combined with the possibility to recycle the enzyme and regener-
ate the anhydride for use in further cycles, make our scheme
potentially useful for the resolution of sec-alcohols. By taking
advantage of the medium reuse option, we are currently studying
the regeneration of the ionic anhydride in situ.
To a solution of 1-methyl-3(11-ethoxycarbonyl-undecyl) imid-
azole hexafluorophosfate 1 (2.07 g, 4.68 mmol) in 20 mL of metha-
nol was added 2.3 mL of a methanolic solution of KOH (4 M,
9.2 mmol). The mixture was stirred at room temperature for 3 h.
After evaporation of the solvent, the crude was acidified with HClaq
(2 M) and then the product was extracted with CH2Cl2 (3 ꢂ 10 mL).
The organic layers were collected, dried under NaSO4, concentrated
under reduced pressure, and passed through a charcoal/SiO2/celite
column with CH2Cl2. The solvent was evaporated under reduced
pressure to give 1.80 g (93.1%) of 2. 1H NMR (DMSO-d6) d 9.09
(1H, s, –NCHN–), 7.75 (1H, s, –NCHCHN–), 7.69 (1H, s,
–NCHCHN–), 4.14 (t, 2H, J = 7.2, –NCH2CH2–), 3.84 (3H, s, CH3N–),
2.18 (t, 2H, J = 7.3, –CH2CO2H), 1.77 (m, 2H, –NCH2CH2–), 1.47
(m, 2H, –CH2CH2CO2–), 1.24 (s, 12H, –NCH2CH2(CH2)6CH2CH2CO2H);
13C NMR (DMSO-d6) d 175.0 (CO acid), 136.9 (–NCHN–), 124.1
(–NCHCHN–), 122.7 (–NCHCHN–), 49.2 (–NCH2CH2–), 36.2
(CH3N–), 34.2 (–CH2CO2–), 29.8 (–NCH2CH2–), 29.2 (CH2), 29.2
(CH2), 29.2 (CH2), 29.0 (CH2), 28.8 (CH2), 25.9 (CH2), 25.0 (CH2).
In agreement with the literature.7
4. Experimental
4.1. Instruments and general methods
All reagents were obtained commercially and used as received,
unless otherwise noted. The dichloromethane used in the

948
R. Teixeira, N. M. T. Lourenço / Tetrahedron: Asymmetry 25 (2014) 944–948
4.3. Preparation of bis((1(11-undecanoic acid)-3-methyl)imida-
zolium hexafluorophosfate) anhydride 3
4.5. General procedure for esterification of alcohols for GC
analysis
After dissolving (1(11-undecanoic acid)-3-methyl)imidazolium
hexafluorophosfate 2 (1.70 g; 4.1 mmol) in dry CH2Cl2 (3 mL), 0.5
equivalents of di-isopropylcarbodiimide (321.4 lL; 2.05 mmol)
After separation of the enantiomers, 1-cyclohexylethanol and
2-octanol were esterified with propionic anhydride in order to
improve the separation of their peaks in GC. For this, the alcohols
were added into a single-neck round-bottomed flask equipped with
a magnetic stirrer. The reaction mixture was stirred overnight at
room temperature. After that, the solvent was evaporated under
reduced pressure and the crude was extracted with CHCl3
(3 ꢂ 25 mL) in order to remove the di-isopropylurea by-product.
PVPyr resin (50 mg) was added to the reaction product resuspended
in acetonitrile. This was left for 1 h with stirring at rt, in order to
remove the unreacted acid. Finally, the ionic liquid was concen-
trated under reduced pressure and extracted once with H2O
(20 mL). After solvent evaporation under reduced pressure, com-
pound 3 was obtained as a colorless oil (1.38 g, 83.0%). DSC:
Tg = ꢀ39.83 °C, 1H NMR (DMSO-d6) d 9.08 (1H, s, –NCHN–), 7.75
(1H, s, –NCHCHN–), 7.68 (1H, s, –NCHCHN–), 4.14 (t, 2H, J = 7.2,
ꢀNCH2CH2–), 3.84 (3H, s, CH3N–), 2.49 (t, 2H, J = 7.3, –CH2CO2H),
1.77 (m, 2H, –NCH2CH2–), 1.53 (m, 2H, –CH2CH2CO2–), 1.25 (s,
12H, –NCH2CH2(CH2)6CH2CH2CO2H); 13C NMR (DMSO-d6) d 169.8
(CO anhydride), 136.5 (–NCHN–), 123.6 (–NCHCHN–), 122.2
(–NCHCHN–), 48.8 (–NCH2CH2–), 35.7 (CH3N–), 34.4 (–CH2CO2–),
30.7 (–NCH2CH2–), 29.4 (CH2), 28.7 (CH2), 28.6 (CH2), 28.3 (CH2),
28.1 (CH2), 25.5 (CH2), 23.7 (CH2). IR mmax (KBr) 3341, 3169, 3124,
2929, 2857, 1815 (anhydride C@O), 1739 (anhydride C@O), 1699,
1618, 1575, 1522, 1465, 1170, 1044, 842, 741, 624, 558 cmꢀ1. Anal.
Calcd for C30H52N4O3F12P2ꢃ3/2H2O: C, 43.22; H, 6.65; N, 6.72. Found:
C, 42.93; H, 6.29; N, 6.87.
were reacted with 1.2 equiv of propionic acid (0.25 mmol, 32
lL)
and 1.2 equiv of pyridine (0.25 mmol, 20 L) in 2 mL of i-Pr2O, at
l
reflux for 4 h. The crude was then washed with HCl 2 M
(2 ꢂ 10 mL) and water (10 mL), dried over anhydrous Na2SO4,
and dried under reduced pressure.
Acknowledgments
Fundação para a Ciência e Tecnologia (POCI 2010) and FEDER
(SFRH/BPD/41175/2007 and PTDC/QUI-QUI/119210/2010, PTDC/
EQU-EQU/122106/2010) are gratefully acknowledged for their
financial support. IBB is funded by grant PEst—OE/EQB/LA0023/
2011 from FCT/MCTES.
References
1. Madeira Lau, R.; Van Rantwijk, F.; Seddon, K. R.; Sheldon, R. A. Org. Lett. 2000, 2,
4189–4191.
2. Domínguez de María, P. Angew. Chem. Int. Ed. 2008, 47, 6960–6968.
3. Lee, S. G. Chem. Commun. 2006, 1049–1063.
4. Sawant, A. D.; Raut, D. G.; Darvatkar, N. B.; Salunkhe, M. M. Green Chem. Lett.
Rev. 2011, 4, 41–54.
5. Naik, P. U.; Nara, S. J.; Harjani, J. R.; Salunkhe, M. M. J. Mol. Catal. B Enzym. 2007,
44, 93–98.
6. Lourenço, N. M. T.; Afonso, C. A. M. Angew. Chem., Int. Ed. 2007, 46, 8178–8181.
7. Lourenço, N. M. T.; Monteiro, C. M.; Afonso, C. A. M. Eur. J. Org. Chem. 2010,
2010, 6938–6943.
8. Reetz, M. T. J. Am. Chem. Soc. 2013, 135, 12480–12496.
9. DiCosimo, R.; McAuliffe, J.; Poulose, A. J.; Bohlmann, G. Chem. Soc. Rev. 2013, 42,
6437–6474.
4.4. General procedure for the enzymatic kinetic resolutions
10. Ansorge-Schumacher, M. B.; Thum, O. Chem. Soc. Rev. 2013, 42, 6475–6490.
11. Patel, R. N. ACS Catal. 2011, 1, 1056–1074.
12. Hasan, F.; Shah, A. A.; Hameed, A. Enzyme Microb. Technol. 2006, 39, 235–251.
13. Ghanem, A.; Aboul-Enein, H. Y. Tetrahedron: Asymmetry 2004, 15, 3331–3351.
14. Muñoz Solano, D.; Hoyos, P.; Hernáiz, M. J.; Alcántara, A. R.; Sánchez-Montero,
J. M. Bioresour. Technol. 2012, 115, 196–207.
15. Ghanem, A.; Aboul-Enein, H. Y. Chirality 2005, 17, 1–15.
16. Park, S.; Kazlauskas, R. J. Curr. Opin. Biotechnol. 2003, 14, 432–437.
17. Kaar, J. L.; Jesionowski, A. M.; Berberich, J. A.; Moulton, R.; Russell, A. J. J. Am.
Chem. Soc. 2003, 125, 4125–4131.
18. Taylor, A. W.; Licence, P.; Abbott, A. P. Phys. Chem. Chem. Phys. 2011, 13, 10147–
10154.
19. Idris, A.; Bukhari, A. Biotechnol. Adv. 2012, 30, 550–563.
20. Chen, Z. G.; Tan, R. X.; Cao, L. Green Chem. 2009, 11, 1743.
21. Heinsman, N. W. J. T.; Schroën, C. G. P. H.; van der Padt, A.; Franssen, M. C. R.;
Boom, R. M.; Van‘t Riet, K. Tetrahedron: Asymmetry 2003, 14, 2699–2704.
22. José, C.; Bonetto, R. D.; Gambaro, L. A.; Torres, M. d. P. G.; Foresti, M. L.; Ferreira,
M. L.; Briand, L. E. J. Mol. Catal. B Enzym. 2011, 71, 95–107.
23. Jeddi, K.; Qazvini, N. T.; Cangialosi, D.; Chen, P. J. Macromol. Sci., B Phys. 2013,
52, 590–603.
The amount of CALB beads necessary to perform each enzy-
matic kinetic resolution (20 mg) was first embedded in n-hexane
(1 mL) for 1 hour, then filtered and washed three times with ace-
tone (3 ꢂ 0.5 mL). After filtration, the beads were used without
drying to carry out the enzymatic reaction.
The enzymatic reactions began with the addition of a stirred
solution of acylating agent (0.41 mmol) and racemic sec-alcohol
(0.41 mmol) in 0.8 mL of distilled acetone to 20 mg of CALB (Nov-
ozym 435Ò). Reactions were carried out in 4 mL glass screw-cap
vials placed in an incubator at 35 °C and 170 rpm, for the time
required to reach ꢄ50% of conversion. The optimal reaction time
for each enzymatic kinetic resolution was previously determined
by GC analysis of several aliquots (100 lL) withdrawn from the
enzymatic reaction at different time intervals. Both the aliquots
and the complete reactions were treated as follows: (i) the enzyme
was filtered off and acetone was evaporated under reduced pres-
sure; (ii) the unreacted (S)-enriched-alcohol was extracted with
Et2O, passed through a pipette sized column with silica, and dried
under reduced pressure; (iii) the remainder of the enzymatic reso-
lution, containing the attached (R)-enantiomer and the ionic acid,
was resuspended in 1 mL of a 4 M solution of KOH in methanol
and stirred for 30 min at 50 °C; (iv) the (R)-enriched-alcohol was
extracted with Et2O, passed through a pipette size column with sil-
ica, and dried under reduced pressure.
24. Susan, M. A. B. H.; Kaneko, T.; Noda, A.; Watanabe, M. J. Am. Chem. Soc. 2005,
127, 4976–4983.
25. Rocha, L. C.; Rosset, I. G.; Luiz, R. F.; Raminelli, C.; Porto, A. L. M. Tetrahedron:
Asymmetry 2010, 21, 926–929.
}
26. Csajági, C.; Szatzker, G.; Rita Toke, E.; Ürge, L.; Darvas, F.; Poppe, L. Tetrahedron:
Asymmetry 2008, 19, 237–246.
27. Davies, S. G.; Smyth, G. D. Tetrahedron: Asymmetry 1996, 7, 1005–1006.
28. Wang, Y.; Wang, R.; Li, Q.; Zhang, Z.; Feng, Y. J. Mol. Catal. B Enzym. 2009, 56,
142–145.
29. Bouzemi, N.; Debbeche, H.; Aribi-Zouioueche, L.; Fiaud, J.-C. Tetrahedron Lett.
2004, 45, 627–630.