CO2-expanded bio-based liquids as novel solvents for enantioselective biocatalysis

Article abstract of DOI:10.1016/j.tet.2017.04.024

For the first time, CO₂-expanded bio-based liquids were reported as novel and sustainable solvents for biocatalysis. Herein, it was found that by expansion with CO₂, 2-methyltetrahydrofuran (MeTHF), and other bio-based liquids, which were not favorable solvents for immobilized Candida antarctica lipase B (Novozym 435) catalyzed transesterification, were tuned into excellent reaction media. Especially, for the kinetic resolution of challenging bulky secondary substrates such as rac-1-adamantylethanol, the lipase displayed very high activity with excellent enantioselectivity (E value > 200) in CO₂-expanded MeTHF (MeTHF concentration 10% v/v, 6 MPa), whereas there was almost no activity observed in conventional organic solvents.

Full text of DOI:10.1016/j.tet.2017.04.024

Tetrahedron 73 (2017) 2984e2989
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
CO2-expanded bio-based liquids as novel solvents for enantioselective
biocatalysis
Hai Nam Hoang a, Yoshihiro Nagashima a, Shuichi Mori b, Hiroyuki Kagechika b,
,
*
Tomoko Matsuda a
a School of Life Science and Technology, Tokyo Institute of Technology, 4259 Nagatsuta-cho, Midori-ku, Yokohama, Kanagawa, 226-8501, Japan
b Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University, 2-3-10 Kanda-Surugadai, Chiyoda-ku, Tokyo, 101-0062, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
For the first time, CO2-expanded bio-based liquids were reported as novel and sustainable solvents for
biocatalysis. Herein, it was found that by expansion with CO2, 2-methyltetrahydrofuran (MeTHF), and
other bio-based liquids, which were not favorable solvents for immobilized Candida antarctica lipase B
(Novozym 435) catalyzed transesterification, were tuned into excellent reaction media. Especially, for the
kinetic resolution of challenging bulky secondary substrates such as rac-1-adamantylethanol, the lipase
displayed very high activity with excellent enantioselectivity (E value > 200) in CO2-expanded MeTHF
(MeTHF concentration 10% v/v, 6 MPa), whereas there was almost no activity observed in conventional
organic solvents.
Received 22 March 2017
Received in revised form
6 April 2017
Accepted 11 April 2017
Available online 12 April 2017
Keywords:
CO2-expanded bio-based liquids
Alternative solvents
© 2017 Elsevier Ltd. All rights reserved.
Bulky secondary alcohols
Lipase-catalyzed transesterification
Kinetic resolution
1. Introduction
given its low polarity, pressurized carbon dioxide scarcely dissolves
polar or high molecular-weight organic compounds.7
Solvents play a key role in the chemical industry. They are often
the major component which constitutes up to 80% of the total
volume of chemicals used in a process.1 As a consequence, most of
the waste generation in organic synthesis, especially in pharma-
ceutical industries, comes from the use of solvents.2 The pressing
eco-environmental urge to develop sustainable processes have
raised the interest in the search for new alternatives for traditional
solvents. The concept of a green solvent started with the well-
known 12 principles of green chemistry proposed by P. Anastas
Adding an organic co-solvent (1e2% mass) can assist the solu-
bility of pressurized carbon dioxide media.8 For enhanced the sol-
ubility, the co-solvent can be introduced in greater amount until
the saturated point, and the co-solvent may separate into another
phase and the system becomes biphasic.8 If CO2 dissolves in this
phase and the phase expands as a result of CO2 dissolution, it is
called a CO2-expanded liquid (CXL).7a,9 CXLs inherit the advantages
of pressurized CO2 and of traditional solvents in an optimal fashion
that pressurized CO2 enhances the transport properties by
increasing diffusivity and by decreasing viscosity of the component
solvent, while this component solvent improves the solubility po-
wer of the pressurized CO2. Moreover, fire suppression capability of
CO2 is another great merit of CXLs. However, another liquid is
needed when CXLs are used and this is usually a petroleum-sourced
volatile organic compound, which is depletable, and often harmful
to the environment.
and J. Warner3; and the term was later clarified into 12 criteria by Y.
4
ꢀ ^
Gu and F. Jerome. A guideline of solvent selection from common to
less classical solvents has been elaborated and ranked based on a
simple combination of regulations.5
As an alternative solvent, pressurized carbon dioxide as liquid and
supercritical phases has attracted great attention because of its
greenness, lack of reactivity, low viscosity, high diffusivity, and ease
of recovery of the reaction products.6 However, despite of all these
Recently some attention has been drawn to the use of bio-derived
solvents, which are renewable sources to replace petroleum-sourced
solvents.4,10 Bio-derived solvents provide clear environmentally ad-
vantages, such as sustainable production, better biodegradability and
lower toxicity. For example, 2-methyltetrahydrofuran (MeTHF), which
can be derived from renewable resources (e.g., furfural or levulinic
interesting advantages, pressurized carbon dioxide presents
a
drawback that hinders its utilization in many industrial applications;
* Corresponding author.
E-mail address: tmatsuda@bio.titech.ac.jp (T. Matsuda).
http://dx.doi.org/10.1016/j.tet.2017.04.024
0040-4020/© 2017 Elsevier Ltd. All rights reserved.

H.N. Hoang et al. / Tetrahedron 73 (2017) 2984e2989
2985
dissolves in the solution. For example, the diffusion of ben-
zene was reported to increase by over 300% on replacing
acid), has been employed as an emerging alternative solvent for
several organic synthesis11 including some biocatalysis processes.12
The concept of combination of scCO2 and a non-volatile bio-based
liquid as a biphasic system was raised by Medina-Gonzalez et al.8 and
recently highlighted by Abou-Shehada,9a however, to the best of our
knowledge, no report on CXL using a bio-based liquid as homogenous
medium for any reaction has yet been communicated. This study is
the first attempt to utilize a CO2-expanded bio-based liquid as a new
class of sustainable, efficient solvent for enzymatic reaction.
14
ꢀ
pure methanol with 90% CO2 at 40 C and 15 MPa. On the
other hand, the viscosity decreases about 75% from pure
15
ꢀ
methanol to CXL methanol at 25 C and 5.7 MPa.
(2) Another important factor should be considered is that the
low hydrophobicity of MeTHF (log P ¼ 1.0), which is not very
preferable for the lipase-catalyzed transesterification,6c,16
was increased by the dissociation of the non-polar CO2
molecules. The lipase was suspected to be less active in hy-
drophilic solvents, and higher esterification rates were ob-
tained with more hydrophobic solvents because hydrophilic
solvents have tendency to strip off the essential water from
enzymes.17 We cited here again the case of CO2-expanded
MeOH to show tunability of this system. By measuring the
dielectric constant, Roskar et al. has shown the polarity of
CO2-expanded methanol dropped with the increase of CO2
pressure.18 On the other hand, Wyatt et al. have observed
changes of Kamlet-Tarf parameters19 of the mixed CO2/
2. Results and discussion
We started our study with measuring the volumetric expansion
of MeTHF by varying CO2 pressure at 20 ꢀC. The visual measure-
ment was performed with 1.0 mL MeTHF magnetically stirred in an
11 mL autoclave equipped with sapphire-glass windows
(Supplementary Fig. S1). The volume increased rapidly as soon as
the pressure reached 5.3 MPa (Supplementary Fig. S2). This
expansion is very much similar to that of THF, where the solubility
of CO2 is explained by the intermolecular interactions of CO2 in
solution.13
The behavior of Candida antarctica lipase B (CAL-B, Novozym
435®) catalyzed kinetic resolution of rac-1-phenylethanol 1a was
performed in CO2-expanded MeTHF under different pressure of CO2
(Fig. 1). There was a clear increased activity towards higher system
pressure. With a same amount of MeTHF, the higher pressures up to
4 MPa, which cause higher composition of CO2, resulted in higher
conversions. Moreover, it should be noticed that even though there
was a ten-time expansion of system volume from 5 MPa to 6 MPa
(Supplementary Fig. S2), which accordingly decreased substrate
concentration, only a slightly lower conversion was witnessed
(17.5% conversion at 5 MPa, 16.8% conversion at 6 MPa). The lipase
was retained its excellent enantioselectivity in all media (E-
value > 200). The higher conversion at higher composition of CO2
could be explained by (1) improved transport properties, (2) a more
favorable hydrophobicity of the solvent systems, and (3) reduced
solvent inhibition caused by polar MeTHF molecules.
methanol solvent,20 and found that the basicity (
b) and the
dipolarity/polarizability (p*) dropped dramatically in the
near critical area.20 Sih et al. reported decreases in molar
volume (Vm) with the increases of CO2 pressure.15 It should
be notice that log P was defined by Kamlet et al. as21
:
log P ¼ 0.24e3.38
b
þ 0.0266Vm ꢁ 0.96
p*
Put all these together, we calculated that log P of methanol/CO2
solution at 35 ꢀC increases from ꢁ0.54 at 0.1 MPa to 0.29 at 7.5 MPa.
(3) A possible reason contributed to higher activity observed in
solvent systems with higher CO2 compositions (0 MPae
4 MPa) is that polar solvent molecules (MeTHF) can inten-
sively interact with the enzyme22 and compete with the
substrate to bind to the active site.23 However, it is noticed
that this solvent-inhibition effect of MeTHF is only significant
when the MeTHF is used at bulk volume. When MeTHF
was used at low concentrations (up to 5 times equivalent
to substrate concentration), the effect was negligible
(Supplementary Table S1).
(1) The transport properties of CO2-expanded MeOH, in term of
diffusivity and viscosity, which are improved as more CO2
We further performed the reaction in other CO2-expanded bio-
based liquids (Fig. 2). Organic carbonates, which can be considered
as non-VOC-producing solvents,24 were used as a partner solvent
with pressurized CO2. The volumetric expansion of diethylene
carbonate was very alike to that of MeTHF and MeOH,
(Supplementary Fig. S2). According to Jessop and Subramaniam,
Fig. 1. Effect of CO2 pressure on CAL-B-catalyzed transesterification of rac-1-
phenylethanol 1a in CO2-expanded MeTHF. Reaction conditions: 0.20 mmol 1a,
0.53 mmol vinyl acetate, 5.0 mg Novozym 435®, 1.0 mL MeTHF, 0e6 MPa of CO2, 20 ꢀC,
1 h. Enantiomeric excess of the corresponding acetate 1b was excellent (ee > 99%) in all
media.
Fig. 2. CAL-B-catalyzed transesterification of rac-1-phenylethanol 1a in different CO2-
expanded bio-based systems and in sole bio-based liquids. Reaction conditions:
0.20 mmol 1a, 0.53 mmol vinyl acetate, 5 mg Novozym 435, 10 mL sole bio-based
liquid (0 MPa of CO2) or 10 mL CO2-expanded bio-based systems (with initial 1 mL
bio-based liquid), 20 ꢀC, 1 h. *Reaction volume 2 mL.

2986
H.N. Hoang et al. / Tetrahedron 73 (2017) 2984e2989
these solvents can be classified as class II liquids, which dissolve
CO2 so effective that volumetric expansion is immense.9b Inter-
estingly, the volumetric expansion of propylene carbonate was at a
maximum of 200% only at 5.3 MPa, and then separated into another
phase (Supplementary Fig. S2). Consequently, propylene carbonate
can be classified as class I liquids, which dissolves CO2 so poorly
that volumetric expansion is small. There has been lack of data of
class I liquids; only water, glycerol and other polyols have been
found as this class.6a,9b Therefore, propylene carbonate can be
nominated for other studies as a new model for class I liquids. CXLs
once again were superior to sole bio-based liquids for CAL-B cata-
lyzed transesterification of rac-1-phenylethanol (Fig. 2). MeTHF (log
P ¼ 1.0) and diethylene carbonate (log P ¼ 1.1) systems resulted in
better performance than propylene carbonate (log P ¼ ꢁ0.41)
system.
The substrate specificity of CAL-B towards several aliphatic and
aromatic secondary alcohols was studied in CO2-expanded MeTHF
system and compared with that in sole MeTHF (Table 1). For sub-
strate 1-6a, reaction in CO2-expanded MeTHF resulted in 2e7 times
better conversion than that in sole MeTHF. Surprisingly, for rac-1-
adamantylethanol 7a bearing a challenging bulky quaternary car-
bon substituent, CAL-B displayed very high activity (29% conver-
sion) in CO2-expanded MeTHF with an excellent enantioselectivity
(E-value > 200), while almost no activity was observed in MeTHF
(Table 1).25
sec-Alcohols having bulky moieties at the stereo-genic center
has gained much interest for many studies.26 However, due to the
steric hindrance of the substrate structures, most of enzyme-
mediated reactions resulted in very low activity27 or unusual/
engineered enzymes had to be used.28 With the impressively high
reactivity of the bulky substrate 7a in CO2-expanded MeTHF, we
further conducted the reaction in various concentration of MeTHF
(Fig. 3). As can be seen, for CO2-expanded MeTHF systems, higher
compositions of CO2 gave the better performance. However, pure
liquid CO2 gave a lower conversion compared with CO2-expanded
MeTHF systems (5% and 10% v/v MeTHF). It is likely due to lower
substrate solubility in liquid CO2 compared with CO2-expanded
Fig. 3. Effect of MeTHF concentration on CAL-B-catalyzed kinetic resolution of rac-1-
adamantylethanol 7a. Reaction conditions: 0.10 mmol rac-1-adamatylethanol 7a,
0.53 mmol vinyl acetate, 10 mg Novozym 435, 10 mL MeTHF at 0 MPa of CO2, or 3 mL,
2 mL, 1 mL, 0.5 mL, 0 mL MeTHF at 6 MPa of CO2, at 20 ꢀC, 24 h. Enantiomeric excess of
the corresponding acetate 7b was excellent (ee > 99%) in all media.
MeTHF system. It was also found that the lipase displayed almost
no activity towards rac-1-adamantylethanol 7a in other conven-
tional organic solvents with a wide range of hydrophobicity
(Supplementary Table S2). Only a trace activity was detected in
hexane (3.1% conversion).
To explain the very difference between reactivity of CAL-B to-
wards bulky substrate rac-1-adamantylethanol 7a in CO2-
expanded MeTHF (and liquid CO2) and conventional organic sol-
vents, apart from the plausible causes as discussed above with the
small substrate rac-1-phenylethanol 1a, the CO2-induced acceler-
ation phenomenon is likely related to the more flexibility of the
enzyme caused by pressurized CO2, which makes the enzyme more
tolerant to bulky substrates. It has been reported that the lipase had
high structural fluctuations in supercritical CO2, particularly the
V139-S150 and T267-A276 residues of active site funnel entrance.29
A similar phenomenon was observed with liquid CO2 as a solvent,
in which the lipase exhibited high activity towards bulky 1-
phenylalkanols.6c,6e Moreover, it is worth to emphasize that the
solvent inhibition effect caused by high polarity of MeTHF is not the
Table 1
Comparison of CAL-B catalyzed kinetic resolution of sec-alcohols in MeTHF and in CO2-expanded MeTHF using various substrates.
Substrate 1-7a
Solvent
Timea (h)
1
Product 1e7b
Conv. (%)
eeproduct (%)
E value
1a
MeTHF
CO2-expanded MeTHF
1b
5.7
23
>99
>99
>200
>200
2a
3a
MeTHF
CO2-expanded MeTHF
5
1
2b
3b
5.1
22
N.d.
95
N.d.
51
MeTHF
CO2-expanded MeTHF
19
40
>99
99
>200
>200
4a
5a
MeTHF
CO2-expanded MeTHF
1
1
4b
5b
2.6
16
N.d.
>99
N.d.
>200
MeTHF
CO2-expanded MeTHF
2.6
19
N.d.
>99
N.d.
>200
6a
7a
MeTHF
CO2-expanded MeTHF
1
6b
7b
17
31
>99
>99
>200
>200
MeTHF
24
<1
N.d.
N.d.
CO2-expanded MeTHF
29
>99
>200
Reaction conditions: alcohol (1-6a: 0.40 mmol; 7a: 0.10 mmol), vinyl acetate (0.53 mmol), Novozym 435 (10 mg) in 10 mL MeTHF or 10 mL CO2-expanded MeTHF (MeTHF
concentration 10% v/v, 6.0 MPa) at 20 ꢀC. N.d.: not determined due to low conversions observed.*
a
The reactions were stopped before reaching 50% conversion to show differences between the reactions in two media and differences amongst substrates. Conversion can
be improved by increasing the reaction time.

H.N. Hoang et al. / Tetrahedron 73 (2017) 2984e2989
2987
4.2. Analytical methods
Conversion (c) was calculated as c ¼ ees/(eep þ ees) and confirm
by GC molar response factor of acetate/alcohol using isoamyl
alcohol as an internal standard where needed. Enantiomeric ratio
(E) value was calculated as E ¼ ln[1 e c(1 þ eep)]/ln[1 - c(1 e eep)] or
E ¼ ln[(1 e c)(1 e ees)]/ln[(1 e c)(1 þ ees)].30
The GC and HPLC profiles for enantiomeric excess measurement
were as followed:
Scheme 1. Synthetic application: scale-up experiment.
GC data for 1a and 1b. tr (CP-Chiralsil DEX CB, 110 ꢀC, 15 min),
min: 6.9, (R)-1a; 7.6, (S)-1a; 4.8, (R)-1b; 4.0, (S)-1b.
crucial factor for the low activity of the lipase towards 7a, as the
lipase also showed very low activity even in hydrophobic organic
solvents such as hexanes or i-octane (Supplementary Table S2).
Encouraged by this feasible and eco-environmentally friendly
process, the lipase-catalyzed kinetic resolution of rac-1-
adamantylethanol 7a was subsequently performed in a gram
scale (Scheme 1). The corresponding chiral products (S)-7a (0.94 g,
yield 44%, ee > 99%) and (R)-7b (1.21 g, 46% yield, ee > 99%) were
isolated. It should be awared that, due to the poor solubility of rac-
1-adamantylethanol in liquid CO2, the reaction at such high con-
centration is not practical in only sole liquid CO2.
GC data for 2a and 2b. tr (CP-Chiralsil DEX CB, 90 ꢀC, 25min),
min: 16.9, (R)-2a; 17.7, (S)-2a; 18.6, (R)-2b; 15.2, (S)-2b.
GC data for 3a and 3b. tr (CP-Chiralsil DEX CB, 110 ꢀC, 25 min),
min: 16.1, (R)-3a; 15.1, (S)-3a; 17.2, (R)-3b; 13.6, (S)-3b.
HPLC data for 4a and 4b. Chiralcel OD-H, hexane: 2-
propanol ¼ 9: 1, 0.5 mL/min, UV 220 nm, min: 21.1, (R)-2a;
20.0, (S)-2a; 10.0, (R)-2b; 11.0, (S)-2b.
GC data for 5a and 5b. tr (CP-Chiralsil DEX CB, 40 ꢀC,
ꢀ
ꢀ
2 min, þ10 C/min, 130 C, 20 min), min: 21.6, (R)-5a; 21.0, (S)-
5a; 17.8, (R)-5b; 18.3, (S)-5b.
GC data for 6a and 6b. tr (CP-Chiralsil DEX CB, 70 ꢀC, 35 min),
min: 20.0, (R)-6a; 20.4, (S)-6a, unseparated; 29.1, (R)-6b; 17.8,
(S)-6b; GC molar response factor of acetate/alcohol (rac-6b/rac-
6a): 0.606.
3. Conclusion
A novel approach to utilize CO2 and bio-based liquids as reaction
media has been demonstrated in a favorable manner for biotrans-
formation. The lipase exhibited very high activity and enantiose-
lectivity in this new solvent, especially towards challenging bulky
substrates. Further studies are on-going in our laboratory to explore
the scope and limits of using CO2-expanded bio-based liquids as
new reaction media.
GC data for 7a and 7b. tr (CP-Chiralsil DEX CB, 130 ꢀC, 30 min),
min: 22.4, (R)-7a; 22.8, (S)-7a, unseparated; 20.4, (R)-7b; 18.6,
(S)-7b; GC molar response factor of acetate/alcohol (rac-7b/rac-
7a): 0.829.
4.3. General procedure for preparation of racemic alcohols (4a, 7a)
4. Experimental section
NaBH4 (0.60 g, 15.8 mmol) was added to a stirred solution of 1-
acetyladamantane (1.62 g, 9.09 mmol) in dry ethanol (20 mL) for
5 h. The exotherm was controlled by an ice bath. The suspension
was stirred at room temperature for 24 h, monitored by TLC. After
the reaction was quenched by an addition of water (20 mL), ethanol
was removed under vacuum and the residue was extracted with
ethyl acetate (3 ꢂ 30 mL). The combined organic phases were
washed with brine (20 mL), dried over MgSO4 and then filtered. The
organic solvent was evaporated under reduced pressure; and the
residue was purified by silica gel column chromatography (hexane/
ethyl acetate, 3:1) to give 7a (1.56 g, 8.7 mmol, 95% yield). The 1H
NMR spectra of the obtained rac-alcohol 4a31 and 7a32 was in
agreement with those reported value.
4.1. General
Novozym 435 (CAL-B) was purchased from MIK Pharm Co., Ltd.
(Japan). Chemicals were purchased from Nacalai Tesque, Inc., Wako
Pure Chemical Industries, Ltd., Tokyo Chemical Industry Co. Ltd., or
Aldrich Chemical Co. and used without further purification unless
otherwise mentioned. Vinyl acetate and MeTHF were freshly dis-
tillated and dried over molsieve 4 Å before used.
GC analysis was performed on a Shimadzu GC-2010 Plus
equipped with FID detector and a CP-Chirasil DEX CB column
(Varian, 25 m ꢂ 0.32 mm, 0.25
mm film thickness) using He carrier
gas (5 mL/min, head pressure: 274 kPa, injector: 180 ꢀC, detector:
180 ꢀC). HPLC analysis was performed on a Shimadzu LC-20AD
instrument, equipped with a Shimazu SPD-20A variable wave-
length UV monitor and a Chiralcel OD-H column. TLC was carried
out on Kiesegel 60F254 (Merck) sheets; spots were visualized by UV
light (254 nm) or developed by treatment with 5% ethanolic
phosphomolybdic acid solution and heating of the dried plates. 1H
NMR spectra were recorded at 400 MHz on a Bruker biospin
AVANCE III 400 spectrometer in CDCl3. The optical rotations were
measured with 10 cm path-length cells on a JASCO P-2200 Polar-
imeter. The visual inspection was observed in an autoclave equip-
ped with sapphire-glass windows (Taiatsu Techno, Osaka, 11 mL).
The experimental apparatus for CXL reactions in batch reactor
consisted of a CO2 gas cylinder, liquid CO2 pump (Jasco, PU-2080-
CO2), manometer (Taiatsu Techno, Osaka, 25 MPa or 35 MPa),
stop valve (Swagelok, SS2NBS4), magnetic stir (Koike, HE-16GA),
thermostatic bath (Iuchi, TR1), recirculation chiller (Eyela, CCA-
1111), and 10 mL stainless steel pressure-resistant vessel (Taiatsu
Techno, Osaka, TVS-N2 type).
4.3.1. 1-(2-Naphthyl)ethyl acetate, rac-4a
Clear crystal (1.31 g, 90%). 1H NMR (400 MHz, CDCl3) 1.57 (d,
J ¼ 6.4 Hz, 3H), 1.97 (s, 3H), 5.02e5.08 (m, 1H), 7.43e7.50 (m, 3H),
7.79e7.83 (m, 4H).
4.3.2. 1-Adamantylethanol, rac-7a
Clear crystal (1.56 g, 95%). 1H NMR (400 MHz, CDCl3)
d 1.10 (d,
J ¼ 6.4 Hz, 3H, CHCH3), 1.28e1.29 (m, 1H), 1.46e1.51 (m, 3H),
1.57e1.66 (m, 6H), 1.71e1.74 (m, 3H), 1.99e2.00 (m, 3H), 3.25e3.31
(m, 1H).
4.4. General procedure for preparation of racemic acetates (1e7b)
as authentic sample for GC/HPLC analysis
In a typical reaction, acetic anhydride (400
idine (200 L) and triethylamine (200 L) were added subsequently
to a stirred solution of rac-1-adamantylethanol 7a (213 mg,
mL, 4.2 mmol), pyr-
m
m

2988
H.N. Hoang et al. / Tetrahedron 73 (2017) 2984e2989
1.18 mmol) in dichloromethane (2 mL) at room temperature. The
reaction was stirred for 48 h at room temperature, monitored by
TLC with assistant of KMnO4 stain. Dichloromethane (10 mL) was
added and the mixture was quenched with an addition of 1 M aq.
HCl (2 mL). The aqueous layer was extracted with dichloromethane
(2 ꢂ 10 mL). The combined organic extracts were neutralized and
washed with saturated aq. NaHCO3 (5 mL) and brine (10 mL), then
dried over MgSO4 and filtered. The organic solvent was evaporated
under reduced pressure and the residue was purified by silica gel
column chromatography (hexane/ethyl acetate, 3:1) to give corre-
sponding rac-acetate 7b (238 mg, 1.07 mmol, 90% yield). The 1H
NMR spectra of acetate 1b,33 2b,33 3b,33 4b,33 5b,33 6b,33 7b34 were
all in agreement with those reported in the literature.
4.6. General procedure for enzymatic reaction in sole bio-based
solvents
Novozym 435 and vinyl acetate were added to a solution of
alcohol in bio-based solvent at 20 ꢀC. The reaction mixture was
vigorously stirred with a magnetic bar. Samples (20
mL) were
withdrawn, filtered through EXtrelut® and diluted with diethyl
ether before being analyzed by GC or HPLC.
4.7. General procedure for preparation of authentic chiral
compound for absolute configuration determination
The kinetic resolution was run on a preparative scale of sec-
alcohol (1.5 mmol) with vinyl acetate (1.0 mmol) by Novozym 435
(100 mg) in CO2-expanded MeTHF or in hexane at 20 ꢀC. After the
reaction, the mixture was filtered off the enzyme, which was
washed with hexane and then concentrated. Silica gel chroma-
tography (hexane/ethyl acetate, 3:1) of the residue gave a fraction
of unchanged alcohol and another fraction of product acetate,
which were found to be pure by 1H NMR spectroscopy. The enan-
tiomeric excess of 1-5a and 1e7b was determined by chiral GC or
HPLC according to experimental section 4.2. The enantiomeric
excess of 6a and 7a was determined by chiral GC analysis of the
resultant acetate after acylating the corresponding alcohol ac-
cording to experimental section 4.4. The absolute configuration of
the alcohols and acetated were determined by comparing the sign
of optical rotation with those of literature.
4.4.1. 1-Phenylethyl acetate, rac-1b
Colorless oil (0.51 g, 88%). 1H NMR (400 MHz, CDCl3)
d 1.53 (d,
J ¼ 6.4 Hz, 3H), 2.07 (s, 3H), 5.85e5.91 (dd, J ¼ 6.4, 13.2 Hz, 1H),
7.26e7.29 (m, 5H).
4.4.2. 1-Phenylpropan-2-yl acetate, rac-2b
Colorless oil (0.41 g, 80%). 1H NMR (400 MHz, CDCl3)
d 1.21 (d,
J ¼ 6.4 Hz, 3H), 1.99 (s, 3H), 2.75 (dd, J ¼ 6.4, 13.6 Hz, 1H), 2.92 (dd,
J ¼ 6.8, 13.6 Hz, 1H), 5.11 (sextet, J ¼ 6.4 Hz, 1H), 7.18e7.24 (m, 5H).
4.4.3. 4-Phenylbutan-2-yl acetate, rac-3b
Colorless oil (0.31 g, 86%). 1H NMR (400 MHz, CDCl3)
d 1.25 (d,
3H, J ¼ 6.4 Hz), 1.79e1.83 (m, 1H), 1.90e1.94 (m, 1H), 2.02 (s, 3H),
2.61e2.68 (m, 2H), 4.91e4.95 (m, 1H), 7.16e7.20 (m, 3H), 7.26e7.30
(m, 2H).
(S)-1a ½a 2D3
ꢃ
¼ ꢁ56.5 (c 1.0, CHCl3, ee ¼ 99%); lit.35 (S)-1a:
½
a 2D5
ꢃ
¼ ꢁ55.1 (c 1.63, CHCl3, ee > 99%).
(R)-1b: ½a 2D3
ꢃ
¼ þ99.2 (c 1.0, CHCl3, ee ¼ 99%); lit.33 (R)-1b:
½
a 2D2
ꢃ
¼ þ106 (c 1.0, CHCl3, ee ¼ 99%).
4.4.4. 1-(2-Naphthyl)ethyl acetate, rac-4b
(S)-2a: ½a 2D4
ꢃ
¼ þ41.6 (c 1.1, CHCl3, ee > 99%); lit.36 (S)-2a:
Colorless oil (0.34 g, 89%). 1H NMR (400 MHz, CDCl3)
d
1.62 (d,
½ ꢃ ¼ þ42.6 (c 1.01, CHCl3, ee > 99%).
a D
3H, J ¼ 6.8 Hz), 2.10 (s, 3H), 6.05 (q, J ¼ 6.6 Hz, 1H), 7.44e7.50 (m,
(R)-2b: ½a 2D4
ꢃ
¼ ꢁ6.9 (c 1.8, CHCl3, ee ¼ 95%); lit.33 (R)-2b:
3H), 7.80e7.84 (m, 4H).
½
a 2D2
ꢃ
¼ ꢁ6.7 (c 5, CHCl3, ee ¼ 93%).
(S)-3a: ½a 2D1
ꢃ
¼ þ17.5 (c 1.2, CHCl3, ee ¼ 99%); lit.36 (S)-3a:
½ ꢃ ¼ þ17.4 (c 1.80, CHCl3, ee ¼ 98%).
a D
4.4.5. 1-(20-Naphthyl)ethyl acetate, rac-5b
(R)-3b: ½a 2D1
ꢃ
¼ þ10.6 (c 1.2, CHCl3, ee ¼ 99%); lit.33 (R)-3b:
Yellowish oil (0.34 g, 89%). 1H NMR (400 MHz, CDCl3)
½
a 2D2
ꢃ
¼ þ13 (c 1, CHCl3, ee ¼ 94%).
d
1.78e1.86 (m, 1H), 1.91e2.02 (m, 3H), 2.08 (s, 3H), 2.70e2.78 (m,
(S)-4a: ½a 1D9
ꢃ
¼ ꢁ54.3 (c 1.8, CHCl3, ee>99%); lit.28b (S)-4a:
1H), 2.84e2.90 (m, 1H), 6.00 (t, J ¼ 4.4 Hz, 1H), 7.11e7.28 (m, 4H).
½
a 2D5
ꢃ
¼ ꢁ24.1 (c 1.02, CHCl3, ee ¼ 54.2%).
(R)-4b: ½a 1D9
ꢃ
¼ þ105.0 (c 1.4, CHCl3, ee ¼ 96%); lit.33 (R)-4b:
4.4.6. 2-Octyl acetate, rac-6b
½
a 2D2
ꢃ
¼ þ122 (c 1, CHCl3, ee>99%).
Colorless oil (0.40 g, 78%). 1H NMR (400 MHz, CDCl3)
d
0.88 (t,
(S)-5a:
½
a 1D9
ꢃ
¼
þ32.5(c 1.1, CHCl3, ee>99%); lit.37 (S)-5a:
J ¼ 8.0 Hz, 3H), 1.20 (d, 3H, J ¼ 8.0 Hz), 1.22e1.35 (m, 8H), 1.42e1.61
½
a 2D3
ꢃ
¼ þ32.1 (c 1.05, CHCl3, ee ¼ 97%).
(m, 2H), 2.02 (s, 1H, COCH3), 4.85e4.93 (sextet, J ¼ 6.4 Hz, 1H).
(R)-5b: ½a 1D9
ꢃ
¼ þ101.2 (c 1.3, CHCl3, ee ¼ 99%); lit.33 (R)-5b:
½
a 2D2
ꢃ
¼ þ110 (c 1.0, CHCl3, ee ¼ 99%).
(S)-6a: ½a 2D3
ꢃ
¼ þ7.6 (c 1.0, CHCl3, ee ¼ 99%); lit.35 (S)-6a:
4.4.7. 1-(1-Adamantyl)ethyl acetate, rac-7b
½
a 2D0
ꢃ
¼ þ9.0 (c 1.23, CHCl3, ee ¼ 97%).
Colorless oil (0.24 g, 90%). 1H NMR (400 MHz, CDCl3)
d 1.10 (d,
(R)-6b: ½a 2D3
ꢃ
¼ ꢁ2.8 (c 1.0, CHCl3, ee ¼ 96.%); lit.33 (R)-6b:
J ¼ 6.4 Hz, 3H), 1.49e1.58 (m, 6H), 1.62e1.73 (m, 6H), 1.98 (br, 3H),
½
a 2D7
ꢃ
¼ ꢁ2.4 (c 3, CHCl3, ee ¼ 97%).
2.04 (s, 3H), 4.53 (q, J ¼ 6.6 Hz, 1H).
(S)-7a: ½a 2D0
ꢃ
¼ ꢁ3.6 (c 1.1, CHCl3, ee > 99%); lit.38 (R)-7a:
½
a 2D0
ꢃ
¼ þ1.26 (c 0.95, CHCl3, ee ¼ 68%).
4.5. General procedure for enzymatic reaction in CO2-expanded
liquids
(R)-7b: ½a 2D3
ꢃ
¼ þ18.5 (c 2.4, CCl4, ee > 99%); lit.39 (S)-7b:
½
a 2D1
ꢃ
¼ ꢁ16 (c 0.46, CCl4, ee ¼ 88%).
Novozym 435, bio-based liquid, alcohol, vinyl acetate, and a
magnetic bar were added and sealed in a high-pressure-resistant
stainless-steel vessel (10 mL volume) at 20 ꢀC. CO2 gas was sent
into the vessel by a CO2 pump until desired pressure was achieved.
The reaction mixture was magnetically stirred. At the end of the
reaction, the vessel was gradually depressurized on ice. An aliquot
4.8. Kinetic resolution of rac-1-adamantylethanol 7a on a gram
scale
Novozym 435 (0.50 g), MeTHF (1.0 mL), rac-1-adamantylethanol
7a (2.13 g, 11.8 mmol), vinyl acetate (0.70 g, 8.14 mmol), and a
magnetic bar were added and sealed in a high-pressure-resistant
stainless-steel vessel (10 mL) at 20 ꢀC. CO2 gas was sent into the
(20
ether before being analyzed by GC or HPLC.
m
L) was filtered through EXtrelut® and diluted with diethyl

H.N. Hoang et al. / Tetrahedron 73 (2017) 2984e2989
2013;17:1188e1199;
2989
vessel by a CO2 pump until 6 MPa. The reaction mixture was
magnetically stirred for 5 days. At the end of the reaction, the vessel
was gradually depressurized on ice. The reaction mixture was
filtered off the enzyme, which was washed with hexane and then
concentrated. Silica gel chromatography (hexane/ethyl acetate, 3:1)
of the residue gave a fraction of (S)-7a (0.94 g, 44%) and (R)-7b
(1.21 g, 46%). The enantiomeric excess of (R)-7b was determined to
be >99%. The enantiomeric excess of (S)-7a was determined to be
>99% by chiral GC analysis of the resultant acetate after acylating
according to experimental section 4.4.
b) Alternative Solvents for Green Chemistry (2), The Royal Society of Chemis-
try, 2013, pp. 149e174.
11. Pace V, Hoyos P, Castoldi L, de Maria PD, Alcantara AR. ChemSusChem. 2012;5:
1369e1379.
12. a) Simeo Y, Sinisterraa JV, Alcantara AR. Green Chem. 2009;11:855e862;
b) Chen ZG, Zhang DN, Cao L, Han YB. Bioresour Technol. 2013;133:82e86;
c) Gao WL, Liu H, Li N, Zong MH. Bioresour Technol. 2012;118:82e88;
d) Gao WL, Li N, Zong MH. J Biotechnol. 2013;164:91e96;
e) Paggiola G, Hunt AJ, McElroy CR, Sherwood J, Clark JH. Green Chem. 2014;16:
2107e2110;
f) Lanctot AG, Attard TM, Sherwood J, McElroy CR, Hunt AJ. RSC Adv. 2016;6:
48753e48756;
g) Nakamura K, Kinoshita M, Ohno A. Tetrahedron. 1994;50:4681e4690.
13. Lazzaroni MJ, Bush D, Brown JS, Eckert CA. J Chem Eng Data. 2005;50:60e65.
14. Sassiat PR, Mourier P, Caude MH, Rosset RH. Anal Chem. 1987;59:1164e1170.
15. Sih R, Dehghani F, Foster NR. J Supercrit Fluid. 2007;41:148e157.
16. Laane C, Boeren S, Vos K, Veeger C. Biotechnol Bioeng. 1987;30:81e87.
17. Gorman LAS, Dordick JS. Biotechnol Bioeng. 1992;39:392e397.
18. Roskar V, Dombro RA, Prentice GA, Westgate CR, Mchugh MA. Fluid Phase
Equilibr. 1992;77:241e259.
Acknowledgments
The research leading to these results has receiving funding with
Sasakawa Scientific Research Grant from the Japan Science Society
(no. 27-343 to H. N. H.), and Novozymes Japan Research Fund 2016.
19. a) Kamlet MJ, Abboud JL, Taft RW. J Am Chem Soc. 1977;99:6027e6038;
b) Taft RW, Kamlet MJ. J Am Chem Soc. 1976;98:2886e2894;
c) Kamlet MJ, Taft RW. J Am Chem Soc. 1976;98:377e383.
20. Wyatt VT, Bush D, Lu J, Hallett JP, Liotta CL, Eckert CA. J Supercrit Fluid. 2005;36:
16e22.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.tet.2017.04.024.
21. Kamlet MJ, Abraham MH, Doherty RM, Taft RW. J Am Chem Soc. 1984;106:
464e466.
22. Li C, Tan TW, Zhang HY, Feng W. J Biol Chem. 2010;285:28434e28441.
23. a) Banik SD, Nordblad M, Woodley JM, Peters GH. ACS Catal. 2016;6:
6350e6361;
References
b) Graber M, Irague R, Rosenfeld E, Lamare S, Franson L, Hult K. BBA-Proteins
Proteom. 2007;1774:1052e1057.
24. Schaffner B, Schaffner F, Verevkin SP, Borner A. Chem Rev. 2010;110:
4554e4581.
25. All the reactions were conducted using immobilized CAL-B (Novozym 435).
The same effect was also observed with free CAL-B.
26. a) Greschner W, Neumann B, Stammler HG, Groger H, Kuck D. Angew Chem Int
Ed. 2015;54:13764e13768;
b) Mori S, Takagaki R, Fujii S, Matsumura M, Tanatani A, Kagechika H. Tetra-
hedron Asym. 2014;25:1505e1512.
27. Janssen AJM, Klunder AJH, Zwanenburg B. Tetrahedron. 1991;47:7645e7662.
28. a) Hatzakis NS, Smonou I. Bioorg Chem. 2005;33:325e337;
b) Ema T, Inoue H. Chem Lett. 2015;44:1374e1376.
29. Monhemi H, Housaindokht MR. J Supercrit Fluid. 2012;72:161e167.
30. Chen CS, Fujimoto Y, Girdaukas G, Sih CJ. J Am Chem Soc. 1982;104:7294e7299.
31. Ema T, Miyazaki Y, Kozuki I, Sakai T, Hashimoto H, Takada J. Green Chem.
2011;13:3187e3195.
32. Evans DA, Michael FE, Tedrow JS, Campos KR. J Am Chem Soc. 2003;125:
3534e3543.
1. a) Henderson RK, Jimenez-Gonzalez C, Constable DJC, et al. Green Chem.
2011;13:854e862;
b) Anastas PT, Kirchhoff MM. Acc Chem Res. 2002;35:686e694.
2. Jimenez-Gonzalez C, Ponder CS, Broxterman QB, Manley JB. Org Process Res Dev.
2011;15:912e917.
3. Horvath IT, Anastas PT. Chem Rev. 2007;107:2169e2173.
4. Gu YL, Jerome F. Chem Soc Rev. 2013;42:9550e9570.
5. Prat D, Wells A, Hayler J, et al. Green Chem. 2016;18:288e296.
6. a) Marriott R, Jessop P, Barnes M. In: Quadrelli EA, Armstrong K, eds. Carbon
Dioxide Utilisation. Amsterdam: Elsevier; 2015:73e96;
b) Alternative Solvents for Green Chemistry (2), The Royal Society of Chemistry,
2013, pp. 115e148;
c) Hoang HN, Matsuda T. Tetrahedron Lett. 2015;56:639e641;
d) Cao C, Matsuda T. In: Stewart JD, ed. Organic Synthesis Using Biocatalysis.
Academic Press; 2016:67e97;
e) Hoang HN, Matsuda T. Tetrahedron. 2016;72:7229e7234.
7. a) Subramaniam B. Coord Chem Rev. 2010;254:1843e1853;
b) Rezaei K, Temelli F, Jenab E. Biotechnol Adv. 2007;25:272e280.
8. Medina-Gonzalez Y, Camy S, Condoret JS. ACS Sustain Chem Eng. 2014;2:
2623e2636.
9. a) Abou-Shehada S, Clark JH, Paggiola G, Sherwood J. Chem Eng Process. 2016;99:
88e96;
b) Jessop PG, Subramaniam B. Chem Rev. 2007;107:2666e2694;
c) Akien GR, Poliakoff M. Green Chem. 2009;11:1083e1100;
d) Alternative Solvents for Green Chemistry (2), The Royal Society of Chemistry,
2013, pp. 262e284;
33. Paivio M, Mavrynsky D, Leino R, Kanerva LT. Eur J Org Chem. 2011:1452e1457.
34. Kuwano R, Sawamura M, Shirai J, Takahashi M, Ito Y. B Chem Soc Jpn. 2000;73:
485e496.
35. Nakamura K, Matsuda T. J Org Chem. 1998;63:8957e8964.
36. Nakamura K, Inoue Y, Matsuda T, Misawa I.
J Chem Soc Perk T. 1999;1:
2397e2402.
37. Matsumura K, Arai N, Hori K, Saito T, Sayo N, Ohkuma T. J Am Chem Soc.
2011;133:10696e10699.
38. Boukachabia M, Zeror S, Collin J, Fiaud JC, Zouioueche LA. Tetrahedron Lett.
2011;52:1485e1489.
e) Hutchenson KW, Scurto AM, Subramaniam B, American Chemical Society.
Meeting, American Chemical Society. Division of Industrial and Engineering
Chemistry. Gas-expanded Liquids and Near-critical Media : Green Chemistry and
Engineering. Washington, D.C: American Chemical Society; 2009.
10. a) Perez-Sanchez M, Sandoval M, Hernaiz MJ, de Maria PD. Curr Org Chem.
39. Zhu DM, Yang Y, Hua L. J Org Chem. 2006;71:4202e4205.