Chemoenzymatic deracemization of chiral secondary alcohols: Process optimization for production of (R)-1-indanol and (R)-1-phenylethanol

Article abstract of DOI:10.1021/op700253t

A process for preparing chiral secondary alcohols by chemoenzymatic deraeemization was optimized. First, the transesterification process of 1-indanol and 1-phenylethanol with vinyl acetate as the acyl donor was optimized using lipase Novozym 435 as biocatalyst The effects of acyl donors, substrate concentration, solvent type, and enzyme amount on activity and enantioselectivity of the said transesteriflcation were investigated. Second,on the basis of the optimized conditions, an efficient biocatalytic resolution system was established with high selectivity, where volumetric productivity of the reaction against (R)-l-indanol and (R)-lphenylethanol reached 529 and 198 g L^-1 d^-1, respectively, during reuse of the enzyme in repeated-batch transesteriflcation reactions. After 10 batches of the reaction, the enzyme still remained stable. Finally, (R)- 1-indanol and (R)-l-phenylethanol were obtained in 95% and 97% ee and in 67% and 71% isolated yields from the corresponding resolution mixture, through an in situ Mitsunobu inversion of the unreaeted alcohols followed by chemical hydrolysis of (R)-acetates.

Full text of DOI:10.1021/op700253t

Organic Process Research & Development 2008, 12, 192–195
Chemoenzymatic Deracemization of Chiral Secondary Alcohols: Process Optimization
for Production of (R)-1-Indanol and (R)-1-Phenylethanol
Ling Ou,† Yi Xu,† Daniel Ludwig,‡ Jiang Pan,† and Jian He Xu*,†
Laboratory of Biocatalysis and Bioprocessing, State Key Laboratory of Bioreactor Engineering, East China UniVersity of
Science & Technology, 130 Meilong Road, Shanghai 200237, China, and Department of Chemical and Bioengineering,
Erlangen-Nuremberg UniVersity, Erlangen 91054, Germany
Abstract:
solubility of the hydrophobic substrate, and the toxicity of
ketones.2 To solve such problems, lipase-catalyzed kinetic
resolution has already been studied for the synthesis of chiral
alcohols for over 2 decades; its disadvantages and bottlenecks
lie in a theoretical yield of 50% and difficulty in separation of
the reactant and the product. Dynamic kinetic resolution (DKR),
especially the combined use of enzymes and transition metal
catalysis, provides a convenient route to resolution of secondary
alcohols.4,5 Using metal catalysts for the in situ racemization
of enzymatically unreactive enantiomers in the enzymatic
resolution of racemic substrates overcomes the limitation of the
maximum 50% yield in the traditional kinetic resolution. As
an example, enzymatic resolution of secondary alcohols, e.g.,
1-phenylethanol and 1-indanol, coupled with ruthenium-
catalyzed racemization employing catalyst 1 resulted in full
transformation of the racemic alcohols to enantiomerically pure
acetate in yields of 80% and 77%.6 However, the reactions
needed a large amount of enzyme (i.e., 30 mg enzyme/1 mmol
scale) as a result of the activity loss/inhibition in the presence
of chemical reagents. In addition, at least three column
extractions are required to purify complex 1.7 Using catalyst 2,
dynamic kinetic resolution of 1-phenylethanol also obtained
enantiomerically pure acetate in 83% selectivity at 91%
conversion,7 but the reactions needed TEMPO as a co-catalyst,
which was expensive and led to formation of a side product.
As an alternative approach, a method combining chemical
and enzymatic approaches has been developed, by which on
the basis of classic enzymatic resolution, chiral alcohols are
directly converted into carboxylic esters of opposite configu-
ration by addition of some chemical reagents, without separation
of the product from the substrate, and as a result optically pure
product may be obtained in a theoretical yield of 100%.8–10 In
addition, the activity of enzyme can be fully kept by this two-
step one-pot method.
A process for preparing chiral secondary alcohols by chemo-
enzymatic deracemization was optimized. First, the transesterifi-
cation process of 1-indanol and 1-phenylethanol with vinyl acetate
as the acyl donor was optimized using lipase Novozym 435 as
biocatalyst. The effects of acyl donors, substrate concentration,
solvent type, and enzyme amount on activity and enantioselectivity
of the said transesterification were investigated. Second,on the basis
of the optimized conditions, an efficient biocatalytic resolution
system was established with high selectivity, where volumetric
productivity of the reaction against (R)-1-indanol and (R)-1-
phenylethanol reached 529 and 198 g L-1 d-1, respectively, during
reuse of the enzyme in repeated-batch transesterification reactions.
After 10 batches of the reaction, the enzyme still remained stable.
Finally, (R)-1-indanol and (R)-1-phenylethanol were obtained in
95% and 97% ee and in 67% and 71% isolated yields from the
corresponding resolution mixture, through an in situ Mitsunobu
inversion of the unreacted alcohols followed by chemical hydrolysis
of (R)-acetates.
1. Introduction
Bearing specific functional groups, chiral alcohols are
important intermediates for the synthesis of many chiral
medicines and also are widely used in the preparation of
hormones, flavors, fragrances, liquid crystals and chiral auxil-
iaries. Chiral alcohols can be synthesized by chemical or
biological approaches. The technology of biocatalysis has drawn
broad attention for its high efficiency, mild reaction condition,
outstanding stereospecificity, and so on.1–3 This method nor-
mally refers to kinetic resolution of racemic alcohols or
asymmetric reduction of prochiral ketones. Generally, optically
pure chiral alcohols can be obtained by biocatalyzed asymmetric
reduction in 100% yield in theory. However, only very few
successful cases of biocatalytic asymmetric reductions have been
reported in industrialized applications, because the reaction
catalyzed by reductase demands cofactors, which are often too
expensive. Other reasons also exist, such as low substrate
concentration and low catalysis efficiency resulting from low
In the technology transfer process of biocatalysis from
laboratory scale to industrialization level, activity, selectivity
(4) Dinh, P. M.; Howarth, J. A.; Hudnott, A. R.; Williams, J. M. J.; Harris,
W. Tetrahedron Lett. 1996, 37, 7623–7626.
(5) Pamies, O.; Bäckvall, J.-E. Chem. ReV. 2003, 103, 3247–3261.
(6) Persson, B. A.; Larsson, A. L. E.; Ray, M. L.; Bäckvall, J.-E. J. Am.
Chem. Soc. 1999, 121, 1645–1650.
* To whom correspondence should be addressed. Tel: +86-21-6425-2498.
Fax: +86-21-6425-2250. E-mail: jianhexu@ecust.edu.cn
† East China University of Science & Technology.
‡ Erlangen-Nuremberg University.
(7) Dijksman, A.; Elzinga, J. M.; Li, Y. X.; Arends, I.W.C.E.; Sheldon,
R. A. Tetrahedron: Asymmetry 2002, 13, 879–884.
(8) Vanttinen, E.; Kanerva, L. T. Tetrahedron: Asymmetry 1995, 6, 1779–
1786.
(1) Reetz, M. T. Curr. Opin. Chem. Biol. 2002, 6, 145–150.
(2) Kroutil, W.; Mang, H.; Edegger, K.; Faber, K. Curr. Opin. Chem.
Biol. 2004, 8, 120–126.
(9) Bouzemi, N.; Aribi-Zouiouechea, L.; Fiaud, J. C. Tetrahedron:
Asymmetry 2006, 17, 797–800.
(3) Wu, H. Y.; Xu, J. H.; Tsang, S. F. Enzyme Microb. Technol. 2004,
34, 523–528.
(10) Boyd, D. R.; Sharma, N. D. J. Mol. Catal. B: Enzym. 2002, 19–20,
31–42.
192
•
Vol. 12, No. 2, 2008 / Organic Process Research & Development
10.1021/op700253t CCC: $40.75
2008 American Chemical Society
Published on Web 02/07/2008

Scheme 1. Preparation of chiral secondary alcohols by
chemoenzymatic deracemization
Table 1. Effect of acyl donor on enzyme activity
initial rate (mM · min-1 ·g enzyme-1
)
substrate
vinyl acetate
vinyl propionate
1-phenylethanol
1-indanol
80.0
482.5
125.0
432.5
and stability of an enzyme all have to be taken into consider-
ation, and thus selection of a biocatalyst with high activity and
establishment of a biocatalytic system with high efficiency and
in adaptation to the former are of special importance. In addition,
for evaluation of the biocatalyst used in a specific process, not
only performance specifications of the biocatalyst but those of
the process are essential, including turnover number, stability
of the biocatalyst, and volumetric productivity of the reactor.
Although the resolution-inversion routes shown in Scheme 1
have been reported previously in the literature,8–10 they were
investigated only from the aspect of synthetic chemistry and
were not optimized systematically in respect to reaction
engineering principles. In this study, the enzymatic resolution
process via transesterification of two typical aryl secondary
alcohols, i.e., 1-indanol (substrate 1) and 1-phenylethanol
(substrate 2), was optimized, through which the effects of some
key parameters such as acyl donors, solvents, reaction temper-
ature, substrate concentration, and enzyme amount on activity
and enantioselectivity of the catalyst were investigated in detail
in order to achieve higher efficiency. Stability of the biocatalyst
in repeated use was also examined. As a result, chiral alcohols
of single configuration were obtained by a Mitsunobu inversion
of the reaction mixture followed by a saponification reaction
(see Scheme 1). The purpose of this study is to develop a highly
efficient chemoenzymatic process for producing valuable chiral
alcohols. The results will provide some beneficial references
for industrial preparation of chiral alcohols by biocatalysis.
Figure 1. Effect of substrate concentration on the initial
reaction rate: ([) 1-indanol and (0) 1-phenylethanol.
that in the case of 1-indanol, vinyl acetate gave a higher ee
value than vinyl propionate, whereas a minor effect was shown
in the case of 1-phenylethanol. Because a relatively higher ee
value is preferred in the Mitsunobu inversion, we chose vinyl
acetate as the acyl donor in following experiments.
Effect of substrate concentration on the enzyme activity was
investigated. As shown in Figure 1, the initial rate increases
linearly with the substrate concentration until the latter reaches
1 mol/L, and further increase of substrate concentration brings
a slow increase in the initial rate, which keeps continuously
increasing even when substrate concentration tops 2 mol/L. This
effect of substrate concentration is far beyond that of using
carboxylic acid as the acyl donor.11
The effect of temperature on the enzymatic activity was also
investigated, indicating that the initial rates for substrates 1 and
2 at 50 °C were 2.3- and 1.6-fold, respectively, that at 30 °C,
while the effect on enantioselectivity of substrates 1 and 2 is
minor. Therefore we chose 50 °C as the temperature optimum
in the following experiments.
Effect of Organic Solvents. Six kinds of organic solvents
were used to investigate their effect on the transesterification
reaction. The results are shown in Table 2.
Transesterification does not occur in a strongly polar solvent
like DMSO, whereas it occurs quickly in hydrophobic solvents
(e.g. n-hexane, diisopropyl ether, n-heptane, and isooctane) with
similar enzymatic activity, which is significantly higher than
that in the middle-polarity solvent vinyl acetate. The enantio-
selectivity (E value) of the enzyme differs largely among
different solvents and substrates. In the case of 1-phenylethanol,
the E value is significantly higher than that with 1-indanol at a
substrate concentration of 100 mM. Among various solvents
employed, diisopropyl ether (DIPE) gave relatively high activity
and selectivity towards 1-indanol. By comparison of enzyme
activity and selectivity against substrates 1 and 2 in different
2. Results and Discussion
Effect of Acyl Donor and Substrate Concentration on
Resolution Reaction. The application of carboxylic acid as an
acyl donor for an ester formation reaction in an organic solvent
system is limited by its acidity and water formation in the
reaction.11 In contrast, in the case of an enol ester as an acyl
donor, the lipase-catalyzed transesterification with a secondary
alcohol can take place more easily, affording the corresponding
ester of the chiral alcohol and vinyl alcohol. The latter can
escape from the reaction system in the form of an aldehyde
after isomerization, therefore making the reaction irreversible.
For this reason, the effect of two enol esters as acyl donors on
the activity of the enzyme in the resolution of substrates 1 and
2 was investigated in this study. Table 1 indicates that in the
case of 1-indanol the effect of the two acyl donors on the
enzyme activity is minor, whereas in the case of 1-phenyl-
ethanol, the effect is more significant: using vinyl propionate
as the acyl donor brings higher enzyme activity than using vinyl
acetate. Investigation of the enzyme enantioselectivity revealed
(11) Suan, C. L.; Sarmidi, M. R. J. Mol. Catal. B: Enzym. 2004, 28, 111–
(12) Straathof, A. J. J.; Jongejan, J. A. Enzyme Microb. Technol. 1997,
21, 559–571.
119.
Vol. 12, No. 2, 2008 / Organic Process Research & Development
•
193

Table 2. Effect of solvents on transesterification of 1-indanol
and 1-phenylethanol with vinyl acetate catalyzed by lipase
Novozym 435
1-indanol
initial ratea
1-phenylethanol
solvent
Eb
initial ratea
Eb
n-hexane
diisopropyl ether
n-heptane
isooctane
vinyl acetate
DMSO
1738
1445
1558
1363
483
3
618
338
633
510
80
300
55
28
32
182
/
>1000
300
500
>1000
0
0
/
a Calculated as mM ·min-1 ·g enzyme-1 b The reactions were carried out in a
.
2-mL volume containing 0.1 M of (()-1 or (()-2 and 4 mg of Novozym 435 at
50 °C. The enantioselectivity (E value) was calculated according to ref 12 by ees
and eep values measured at 25 h of reaction.
Figure 3. Repeated batch reactions of enzymatic transesteri-
fication of 1-indanol (A) and 1-phenylethanol (B): (solid bar)
conversion, ([) volumetric productivity; (O) eep.
efficiency and catalyst cost, an enzyme load of 6 g/L was
adopted in the following repeated-batch reactions.
Operation Stability of the Enzyme. It is shown in Figure
3 that the enzyme remains stable with high activity, enantio-
selectivity, and stability after 10 batches of transesterification
of substrates 1 and 2 at a high substrate concentration of 1 mol/
L. After 10 batches of reaction, the conversions for both of
these two alcohols were kept at ca. 49.5-50.2%, with ee values
above 99% for the unreacted (S)-alcohols. Generally, volumetric
productivity for biocatalytic reaction on industrial scale13 ranges
from 100 to 1000 g ·L-1 ·d-1. Straathof14 mentioned that among
134 bioconversions at the industrial level, the average of
volumetric productivity for fine chemicals reached 372
g·L-1 ·d-1 (or 15.5 g·L-1 ·h-1). In this work, the volumetric
productivity for enzymatic transesterification of 1-indanol and
1-phenylethanol using merely 6 g of enzyme per liter were 529
and 198 g ·L-1 ·d-1 ,respectively, which are approaching or
beyond the average value for industrial practice.
After filtering off the enzyme and evaporating the solvent
contained in the transesterification mixture collected and
combined from 10 batch reactions, the residue was treated by
chemical inversion and saponification, followed by purification
with silica gel column. Consequently, (R)-1-indanol was
obtained in 67% isolated yield and 95% ee and (R)-1-
phenylethanol was obtained in 71% isolated yield and 97% ee.
Figure 2. Initial rate (A) and productivity (B) of the enzymatic
transesterification of 1-indanol (9) and 1-phenylethanol (4) with
vinyl acetate with different enzyme loads.
solvents, DIPE was chosen as an appropriate solvent in
following experiments.
Effect of Enzyme Load. Using DIPE as the solvent, the
effects of the enzyme load on the initial rate of the trans-
esterification and on the enzyme productivity were then studied.
It is shown in Figure 2 that initial rates of enzymatic trans-
esterification for substrates 1 and 2 increased along with the
enzyme amount, and the augment decreased gradually. Pro-
ductivity of the enzyme catalyzing transesterification of substrate
1 decreased along with the enzyme amount, whereas in the case
of substrate 2 it changed little within the range of enzyme
amount examined. Taking into consideration both the catalysis
(13) Bommarius, A. S.; Riebel B. R. Biocatalysis: Fundamentals and
Applications; Wiley-VCH: New York, 2004; pp16–36..
(14) Straathof, A. J. J.; Panke, S.; Schmid, A. Curr. Opin. Biotechnol. 2002,
13, 548–556.
194
•
Vol. 12, No. 2, 2008 / Organic Process Research & Development

In conclusion, the biocatalytic reaction system was carefully
optimized for optical resolution of 1-indanol and 1-phenyl-
ethanol, resulting in the following optimal conditions: vinyl
acetate as acyl donor, DIPE as solvent, initial concentration of
alcohol 1 mol/L, enzyme load 6 g/L, and reaction temperature
50 °C. Under these optimized conditions, operational stability
of the enzyme for substrates 1 and 2 was examined with 10
batch reactions. After combining the reaction mixture of 10
batches and evaporation of the organic solvent, followed by in
situ Mitsunobu inversion of the unreacted alcohols and saponi-
fication of the esters, a single enantiomer of each alcohol was
obtained. The volumetric productivities of enzymatic trans-
esterification for 1-indanol and 1-phenylethanol were as high
as 529 and 198 g ·L-1 ·d-1 respectively. This study suggests
that chemoenzymatic deracemization has industrial level ef-
ficiency and is potentially applicable for the efficient preparation
of chiral secondary alcohols in industry. It will also be a
reference for the production of other valuable chiral alcohols
with high efficiency by biocatalysis.
transferred to a 250-mL flask to which were added AcOH (0.78
g, 13 mmol), PPh3 (3.40 g, 13 mmol), and 50 mL of diethyl
ether. The reaction mixture was immediately cooled to 0 °C,
and a solution of diisopropyl azodicarboxylate (DIAD) (2.63
g, 13 mmol) with 50 mL of diethyl ether was added dropwise,
under vigorous magnetic stirring during 1 h. The mixture was
allowed to warm to room temperature and stirred for 24 h.
To the same flask was added 15 mL of a solution of KOH
in methanol (10% w/v), and the reaction mixture was stirred at
50 °C for 2 h. After the saponification was completed (as
determined by GC), to this mixture was added 150 mL of
demineralized water. The pH was adjusted with hydrochloric
acid to 7.0, followed by extraction with dichloromethane (150
mL × 3) and washing of the organic phase with saturated
sodium chloride solution (20 mL × 2), and the collected organic
phase was then dehydrated overnight with anhydrous sodium
sulfate. Concentration of the organic phase followed by silica
gel column chromatographic purification of the residue using
hexane and ethyl acetate (5:1, v/v) gave only (R)-1-indanol and
1
(R)-1-phenylethanol. (R)-1-Indanol: H NMR (CD3OD, 500
3. Experimental Section
MHZ) 1.89–1.91 (m, 1H), 2.38–2.40 (m, 1H), 2.74–2.78 (m,
1H), 2.98–3.02 (m, 1H), 5.14 (t, J ) 6.21, 1H), 7.17–7.38 (m,
4H); MS (EI+) m/z 135 (MH+). (R)-1-Phenylethanol: 1H NMR
(CD3OD, 500 MHZ) 1.43 (d, J ) 6.51 HZ, 1H), 4.81 (dd, J1 )
12.95 HZ, J2 ) 6.47 HZ, 1H), 7.20–7.37 (m, 5H); MS (EI+)
m/z 123 (MH+). Qualitative determinations of alcohols were
done via TLC analysis, purity and yield by GC, and ee by
HPLC.
Analysis. Gas chromatography was used for the determi-
nation of reaction conversion (Shimadzu GC-14C, SE-54 quartz
capillary column, injector 280 °C, detector 350 °C, column 150
°C, splitting ratio 1:50). The retention times for 1-indanol,
1-phenylethanol, 1-indanyl acetate and 1-phenylethyl acetate
were 4.7, 2.9, 7.4 and 3.9 min, respectively. HPLC (Shimadzu)
and chiralcel OD-H column (Daicel, Japan) were used for
determination of the enantiomeric excess of remaining substrate
alcohols and product acetates (hexane/i-PrOH 98:2 (v/v), flow
rate 1.0 mL/min, UV detector wavelength 212 nm). The
retention times (all in minutes) for the alcohols and their esters
are as follows: 1-indanol, t(S) ) 17.0, t(R) ) 20.0; 1-phenyl-
ethanol, t(S) ) 12.9, t(R) ) 16.3; 1-indanyl acetate, t(S) ) 5.0,
t(R) ) 6.6; 1-phenylethyl acetate, t(S) ) 4.9, t(R) ) 6.2.
Enzyme and Reagents. This work was conducted with
lipase (Novozym 435) procured from Novozymes Co. (Den-
mark). 1-Indanol (98% purity) and 1-phenylethanol (97% purity)
were purchased from Alfa-Aesar. Other chemicals were from
local reagent suppliers and used directly without any further
treatment.
Typical Procedure for Transesterification Catalyzed by
Novozym 435. Racemic alcohol 1 (1-indanol) or 2 (1-
phenylethanol) was added to vinyl acetate (the molar ratio of
substrate to acyl donor was 1:5) in a 10-mL test tube (A), and
2 mL of solvent and 4 mg of enzyme were added into another
10-mL test tube (B). After both test tubes were preheated to
the required temperature, the contents of test tube A was poured
into tube B, and the latter was shaken at the required temperature
for an appropriate time. Samples were taken at certain times
followed by dilution with ethanol of equal volume for detection
of the degree of conversion by gas chromatography or dilution
with hexane and isopropyl alcohol (98:2) of 10 times volume
for measurement of ees and eep by HPLC.
Repeated-Batch Reaction. Into a 100-mL Erlenmeyer flask
with vinyl acetate (4.60 mL, 50 mmol) dissolved in 4.18 mL
of DIPE was added racemic alcohol 1 (1.34 g, 10 mmol) or 2
(1.22 mL, 10 mmol). After these reactants were preheated to
50 °C, 60 mg of carrier-bound lipase (Novozym 435) was
added. The mixture was magnetically stirred at 50 °C for 3 h
(for alcohol 1) or 10 h (for alcohol 2). The reaction was stopped
by filtering off the solid enzyme. After washing the enzyme
with DIPE (10 mL × 2), fresh substrate was added into the
flask for the next reaction, and in total 10 batches of reactions
were carried out.
Acknowledgment
This research was financially supported by National Natural
Science Foundation of China (nos. 20402005 and 20576037)
and Ministry of Science and Technology, P. R. China (no.
2006AA02Z205). We are indebted to Mr. Lei Chang for his
kind assistance.
Typical Procedure for Mitsunobu Inversion and Saponi-
fication Reaction. After filtering off the enzyme and evaporat-
ing the solvent in transesterification mixture, the residue was
Received for review November 7, 2007.
OP700253T
Vol. 12, No. 2, 2008 / Organic Process Research & Development
•
195