Liquid carbon dioxide as an effective solvent for immobilized Candida antarctica lipase B catalyzed transesterification

Article abstract of DOI:10.1016/j.tetlet.2014.12.080

The transesterification of alcohols catalyzed by immobilized Candida antarctica lipase B (Novozym 435) was found to be effectively enhanced using a liquid CO₂ medium when it was compared with that using organic solvents. The large-scale kinetic-resolution of secondary alcohol by the immobilized lipase was also successfully performed with a continuous packed-column reactor that stably afforded corresponding enantiopure products. Herein, liquid CO₂ was proved for the first time to be superior to conventional organic solvents for biotransformation.

Full text of DOI:10.1016/j.tetlet.2014.12.080

Tetrahedron Letters 56 (2015) 639–641
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Liquid carbon dioxide as an effective solvent for immobilized
Candida antarctica lipase B catalyzed transesterification
⇑
Hai Nam Hoang, Tomoko Matsuda
Department of Bioengineering, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama, Kanagawa 226-8501, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The transesterification of alcohols catalyzed by immobilized Candida antarctica lipase B (Novozym
Received 11 November 2014
Revised 5 December 2014
Accepted 14 December 2014
Available online 19 December 2014
Ò
435 ) was found to be effectively enhanced using a liquid CO
2
medium when it was compared with that
using organic solvents. The large-scale kinetic-resolution of secondary alcohol by the immobilized lipase
was also successfully performed with a continuous packed-column reactor that stably afforded corre-
sponding enantiopure products. Herein, liquid CO
tional organic solvents for biotransformation.
2
was proved for the first time to be superior to conven-
Keywords:
Ó 2014 Elsevier Ltd. All rights reserved.
Liquid carbon dioxide
Asymmetric synthesis
Candida antarctica lipase B
Transesterification
Introduction
2
To exploit the potential of liquid CO as a solvent for biotrans-
formation, the transesterification of alcohols catalyzed by immobi-
Ò
The scope of biotransformation, particularly in asymmetric syn-
thesis, has been extended through the use of enzymes in organic
lized Candida antarctica lipase B (Novozym 435 ) was chosen as a
model reaction. We found that the liquid CO
2
medium is superior
to organic solvents. A liquid CO fluid was also successfully
employed for a flow system with a packed-column reactor for
the continuous kinetic resolution of rac-1-phenylethanol and stea-
dily afforded enantiopure products.
1
–3
solvents
as well as in ‘green’ nonaqueous media, such as ionic
2
4
5–7
liquids and supercritical fluids.
Carbon dioxide, which has a
number of positive impacts on green chemistry as a nonflammable,
nontoxic, abundant, and generally chemically inert source, has
been intensively studied in its supercritical phase as an alternative
8
solvent for enzymatic reactions. Some of the advantages of using
Results and discussion
2
supercritical CO as a solvent include low viscosity, low surface
tension, and ease of product recovery. At below critical tempera-
ture (31 °C), carbon dioxide can exist in a liquid phase, which can
also be employed as a solvent for chemical reactions.8 However,
First, the kinetic resolution of rac-1-phenylethanol catalyzed by
Novozym 435Ò was performed in liquid CO
and in organic sol-
vents using vinyl acetate as the acyl donor in a batch system
Scheme 1). As shown in Figure 1, as expected, the activity of the
2
,9
2
to the best of our knowledge, no reports exist on liquid CO as a
(
practical solvent for biocatalysts. Because carbon dioxide in a
liquid phase is intrinsically different from its supercritical phase
in many physical properties as a solvent, the enzymatic behaviors
lipase is closely related to the solvent hydrophobicity (represented
in logP value) where the more hydrophobic solvents generally
showed higher yields. Interestingly, the lipase exhibited the high-
est transesterification activity with excellent enatioselectivity (ee
in liquid and supercritical CO
ent benefit of using liquid CO
2
could be very different. The appar-
is that it can be maintained under
2
>
99%) in liquid CO
, followed by hexane and toluene. Liquid CO
2
2
relatively modest pressure (e.g., 4.5 MPa at 10 °C) that reduces
the cost of specialized equipment for high-pressure (over
7
2
.4 MPa) reaction in supercritical CO . Furthermore, it can be
O
employed at low temperature, which has the potential to enhance
OH
OH
Novozym 435
Vinyl acetate
the enatioselectivity by the low-temperature strategy.1
0–12
O
(
Liquid CO2 or
organic solvents
+
R)
(S)
⇑
Corresponding author. Tel./fax: +81 45 924 5757.
E-mail address: tmatsuda@bio.titech.ac.jp (T. Matsuda).
Scheme 1. Kinetic resolution of rac-1-phenylethanol.
http://dx.doi.org/10.1016/j.tetlet.2014.12.080
040-4039/Ó 2014 Elsevier Ltd. All rights reserved.
0

6
40
H. N. Hoang, T. Matsuda / Tetrahedron Letters 56 (2015) 639–641
Figure 1. Effect of solvents on Novozym 435 catalyzed kinetic resolution of rac-1-phenylethanol. Reaction conditions:15 Substrate 0.83 mmol, vinyl acetate 5.4 mmol,
Novozym 435Ò 5 mg, solvent 10 ml, 20 °C, 2 h, pressure for liquid CO >99%) in all media. aNo reaction
6.5 MPa. Ee of the product (acetate) was found to be excellent (ee
occurred due to insufficient mixing of the enzyme and substrate since the density of the solvent greatly exceeds that of the immobilized enzyme. N.a.: not available. Dielectric
constants of selected solvents: liquid CO (20 °C, 6 MPa) 1.48, hexane 1.88, isooctane 1.94, toluene 2.38, diisopropyl ether 3.88, chloroform 4.81, THF 7.58.
Ò
2
p
2
was reported to have very similar behaviors with a hydrocarbon
2
propanols and 2-octanol, liquid CO promoted the reactions much
13
solvent with very low polarizability. The high activity of the
lipase in liquid CO as well as in solvents with high logP values
more effectively than hexane (entries 4–7 and 11) with statistically
significant differences (p <0.05). With substrates having longer
side-chains, 1-phenylbutanol and 1-phenylpentanol, there was
2
could be partly explained by the inability of hydrophobic solvents
to strip off the essential water present on the outer layers of the
no reaction observed in hexane but some reactions in liquid CO
2
1
4
16
enzyme.
To further investigate the enhancement effect caused by liquid
using different substrates, the transesterification of various
aromatic, aliphatic, and allylic primary and secondary alcohols
was carried out in liquid CO and in hexane (Table 1). The transe-
sterification rates of 1-phenylethanol, 2-phenylethanol, and
-octen-3-ol in liquid CO were comparable with those observed
(entries 8 and 9). Our previous work also found that the activity
of Novozym 435Ò toward selected substrates was also enhanced in
CO
2
supercritical CO
ogeneous distribution of water and CO
regions of the lipase could efficiently facilitate the diffusion of cer-
2
. This agreement supports the idea that the heter-
2
molecules on certain
2
1
7
tain substrates into the catalytic active site. Furthermore, the dif-
ference was even more remarkable at lower temperature with the
reaction with 1-phenyl-2-propanol (entries 5 and 6), but not in the
1
2
in hexane (entries 1–3 and 10). Notably, with 1-, 2,- and 3-phenyl-
Table 1
Ò
using a batch reactora
2
Novozym 435 catalyzed transesterification of various alcohols in hexane or liquid CO
c
Entry
Substrates
Solvent
Time (h)
1
Temp (°C)
Products
Yield (%)
ee
p
(%) (Config.)
E-Value
OH
Hexane
Liquid CO
Hexane
Liquid CO
20
OAc
31
32
28
30
>99 (R)
>99 (R)
>99 (R)
>99 (R)
>200
>200
>200
>200
1
2
2
2
Ph
2
5
Ph
3b
OH
Hexane
Liquid CO
1
20
20
OAc
70
70
À
À
À
À
Ph
OH
Ph
Ph
2
2
Hexane
Liquid CO
18
OAc
Ph
4
25
>99 (R)
>99 (R)
N.d.
>200
4
OH
Hexane
Liquid CO
Hexane
Liquid CO
4
20
5
OAc
10
27
7
97 (R)
96 (R)
97 (R)
96 (R)
62
74
70
80
5
6
2
2
Ph
Ph
Ph
Ph
18
34
7b
OH
Hexane
Liquid CO
0.5
60
20
20
OAc
82
93
À
À
À
À
2
2
OH
OH
Hexane
Liquid CO
OAc
OAc
0
6
N.d.
N.d.
N.d.
N.d.
8
9
Ph
Ph
Ph
Ph
Hexane
Liquid CO
100
20
0
4
2
OH
Hexane
Liquid CO2
2
1
20
20
OAc
21
21
>99 (S)
>99 (S)
>200
>200
1
0
1
OH
Hexane
Liquid CO2
OAc
33
40
>99 (R)
>99 (R)
>200
>200
1
a
Reaction conditions:15 alcohol (0.40 mmol) with vinyl acetate (2.2 mmol) and Novozym 435Ò (10 mg) in 10 ml hexane or liquid CO
(6.5 MPa) at 20 °C. Product yields and
2
ee values are averages of at least three reaction runs determined by GC analysis. Absolute configurations were determined by comparing the optical rotation values with data
from the literature.
b
Novozym 435 (5 mg).
c
ee of the products (acetates) N.d.: not determined due to the low conversion observed.

H. N. Hoang, T. Matsuda / Tetrahedron Letters 56 (2015) 639–641
641
Table 2
Novozym 435Ò catalyzed kinetic resolution of rac-1-phenylethanol with a continuous flow of liquid CO
and substrates into a packed-column reactora
2
Cycle No.b
In-flux ratec
mg/min)
Recoveredd (g)
ee
e
(%)
ee
p
e
(%)
cf (%)
E-Valueg
E-Factorh
s
(
1
2
3
18.72
18.75
18.62
7.08
7.27
6.85
>99
>99
>99
>99
>99
>99
49.9
50.0
49.9
>200
>200
>200
<0.3
<0.3
<0.3
a
Reaction conditions19: immobilized enzyme 1.6 g, CO
À1
À1
2
flow rate 1.0 mL min , substrate flow rate 0.020 ml min
(volume ratio of rac-1-phenylethanol/vinyl
acetate = 2:1), 20 °C, 6.5 MPa.
b
One cycle consists of four steps: pressurization and stabilization for 1 h; sampling for 8 h; washing with liquid CO
2
flow for 1 h; depressurization to 0.1 MPa and storing
for 14 h at ambient conditions.
c
The influx was monitored by a precision balance and measured by the amount of substrate mixture sent over time.
Recovered outflux was collected at the stationary state for 8 h.
d
e
f
ee
s
of the substrate (alcohol) and ee
p
of the product (acetate) were determined by chiral GC analysis.
and ee (and confirmed by independent calculations from the ratio of the integral of the alcohol and the corresponding acetate
Conversion (c) was calculated from ee
s
p
1
signals in the H NMR spectrum).
g
Enantiomer selectivity (E-value) was calculated from ee
s
and ee
The E-factor is equal to the kg waste (acetaldehyde and a small excess of vinyl acetate) divided by kg products (the enantiopure acetate and alcohol). Recyclable factors
such as CO gas and re-used immobilized enzyme are not included in the calculation.
p
.
h
2
case of 1-phenylethanol (entries 1 and 2), suggesting that the
structure of the substrates mainly contributes the facilitation effect
caused by the interaction of the solvent–enzyme–substrate.
Finally, to examine the potential of the industrial applicability
3. Adlercreutz, P. Chem. Soc. Rev. 2013, 42, 6406.
4
.
Itoh, T. Biotransformation in Ionic Liquid. In Future Directions in Biocatalysis;
Matsuda, T., Ed.; Elsevier Science B.V.: Amsterdam, 2007; pp 3–20. Chapter 1.
Hobbs, H. R.; Thomas, N. R. Chem. Rev. 2007, 107, 2786.
5.
6. Matsuda, T.; Watanabe, K.; Harada, T.; Nakamura, K.; Arita, Y.; Misumi, Y.;
Ichikawa, S.; Ikariya, T. Chem. Commun. 2004, 2286.
of liquid CO
-phenylethanol with vinyl acetate by Novozym 435 was per-
formed using a packed-column reactor with a continuous flow of
liquid CO and substrates (Table 2). The flow system was run for
2
fluid for biocatalysis, the kinetic resolution of rac-
7
8
.
.
Matsuda, T. J. Biosci. Bioeng. 2013, 115, 233.
Ò
1
DeSimone, J. M.; Tumas, W. Green Chemistry Using Liquid and Supercritical
Carbon Dioxide; Oxford University Press: New York, 2003.
Haberman, J. X.; Irvin, G. C.; John, V. T.; Li, C. J. Green Chem. 1999, 1, 265.
9
.
2
1
1
0. Phillips, R. S. Trends Biotechnol. 1996, 14, 13.
1. Sakai, T.; Kishimoto, T.; Tanaka, Y.; Ema, T.; Utaka, M. Tetrahedron Lett. 1998,
39, 7881.
three operation cycles (24 h/cycle), to stably afford the correspond-
ing enantiopure acetate and alcohol. For the green chemistry con-
cept with a waste prevention goal, Sheldon’s E-factor18 has widely
12.
Matsuda, T.; Kanamaru, R.; Watanabe, K.; Kamitanaka, T.; Harada, T.;
Nakamura, K. Tetrahedron: Asymmetry 2003, 14, 2087.
been adopted as a green chemistry metric for quickly assessing the
environmental impact of manufacture processes. The ideal E-factor
is zero, which means no waste was generated. Some chemical
industry processes, such as fine chemicals and particularly phar-
maceuticals, easily have E-factor >25.18 Our system minimized
waste with a very low E-factor (<0.3) using no organic solvent
and a sufficient amount of vinyl acetate.
1
3. Hyatt, J. A. J. Org. Chem. 1984, 49, 5097.
14. Laane, C.; Boeren, S.; Vos, K.; Veeger, C. Biotechnol. Bioeng. 2009, 102, 2.
5. Batch reactions with liquid CO were conducted as follows: a mixture of alcohol
1
2
and vinyl acetate, and enzyme was added and sealed in a high-pressure-
resistant stainless-steel vessel (10 ml volume). The temperature was
controlled by
(Eyela, CCA-1111). CO
PU-2080-CO Plus) until the desired pressures were achieved. The vessel was
a
thermostatic bath equipped with
a
recirculating chiller
2
gas was sent into the vessel by a CO
2
pump (Jasco,
2
then vigorously stirred with a magnetic bar. At the end of the reaction, the
mixtures were collected by placing the vessel on ice and depressuring. The
reaction mixture was collected in hexane (10 ml) and analyzed by GC. The
reactants and products were all soluble under the reaction conditions that
were confirmed by visual inspection of the reaction mixture in a reactor
equipped with sapphire windows (2.5 cm in diameter).
Conclusion
2
For the first time, liquid CO was found to be superior to other
The effect of supercritical CO
same reaction conditions but at 40 °C and compared with that of hexane at
0 °C. Supercritical CO (conversion 35.7 ± 1.3%, ee >99%) showed
comparable effect with hexane (conversion 35.5 ± 2.2%, eep >99%).
2
on the reaction was also investigated under the
conventional organic solvents for the lipase catalyzing the transe-
sterification of alcohols. Large-scale asymmetric synthesis also suc-
cessfully achieved waste minimization. On-going studies on liquid
4
2
p
a
CO
laboratory.
Caution: All the reactions in liquid carbon dioxide involve high
2
as a ‘green’ medium for biocatalyzed synthesis continue in our
16. Matsuda, T.; Tsuji, K.; Kamitanaka, T.; Harada, T.; Nakamura, K.; Ikariya, T.
Chem. Lett. 2005, 34, 1102.
1
7. Silveira, R. L.; Martinez, J.; Skaf, M. S.; Martinez, L. J. Phys. Chem. B 2012, 116,
671.
18. Sheldon, R. A. Chem. Soc. Rev. 2012, 41, 1437.
9. The continuous flow apparatus consisted of a CO
Jasco, PU-2080-CO Plus), substrate pump (Jasco, PU-2085 Plus),
5
pressures and should only be conducted with appropriate equip-
ment and proper precautions.
1
2 2
gas cylinder, a CO pump
(
2
a
manometers (Taiatsu Techno, Osaka, 15 or 25 MPa), a packed-column reactor
(3.94 ml, 28.5 cm long), stop valves (Swagelok, SS3NBS4), a turbulent mixer (a
References and notes
0.5 ml tube packed with cotton), a thermostatic bath (Iuchi, TR1), a
recirculation chiller (Eyela, CCA-1111), and a back pressure regulator (Jasco,
BP-2080).
1
2
.
.
Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz, S.; Moore, J. C.;
Robins, K. Nature 2012, 485, 185.
Nestl, B. M.; Hammer, S. C.; Nebel, B. A.; Hauer, B. Angew. Chem., Int. Ed. 2014,
53, 3070.