Separation of enantiomers by lipase-catalyzed fluorous-phase delabeling

Article abstract of DOI:10.1016/S0040-4020(02)00258-2

Simultaneous enantiomer-selective fluorous-phase delabeling and kinetic resolution was achieved by lipase-catalyzed alcoholysis of highly fluorinated esters of racemic alcohols. The separation of the fast-reacting delabeled enantiomers and the slow-reacting fluorous-phase labeled enantiomers was performed very efficiently by partition in an organic/fluorous biphasic solvent system.

Full text of DOI:10.1016/S0040-4020(02)00258-2

TETRAHEDRON
Pergamon
Tetrahedron 58 ;2002) 4085±4089
Separation of enantiomers by lipase-catalyzed
¯uorous-phase delabeling
Sauda M. Swaleh, Benno Hungerhoff, Helmut Sonnenschein and Fritz Theilp
È
ASCA GmbH Angewandte Synthesechemie Adlershof, Richard-Willstatter-Straûe 12, D-12489 Berlin, Germany
Dedicated to Professor Dr Hans Schick on the occasion of his 65th birthday
Received 3 November 2001; revised 21 December 2001; accepted 16 January 2002
AbstractÐSimultaneous enantiomer-selective ¯uorous-phase delabeling and kinetic resolution was achieved by lipase-catalyzed
alcoholysis of highly ¯uorinated esters of racemic alcohols. The separation of the fast-reacting delabeled enantiomers and the slow-reacting
¯uorous-phase labeled enantiomers was performed very ef®ciently by partition in an organic/¯uorous biphasic solvent system. q 2002
Elsevier Science Ltd. All rights reserved.
1. Introduction
usually separated from each other by chromatography.
This chromatographic step might be regarded as a major
drawback for using this type of reactions on large scale in
the pharmaceutical industry or in high-throughput kinetic
resolutions.
Per¯uorinated and non-¯uorinated organic solvents are very
often immiscible at ambient temperature. Based on this fact,
in the search for chemical transformations integrating
innovative workup procedures ¯uorous techniques1,2 are
becoming increasingly popular in organic synthesis for the
separation of temporarily ¯uorous labeled from non-labeled
compounds. If one or more of the components of a reaction
mixture, e.g. homogeneous catalysts,3±5 reagents or
products6±10 are equipped with a per¯uorinated auxiliary,
the organic and the ¯uorous components can be separated
easily by either liquid±liquid extraction, solid±liquid
extraction or ¯uorous chromatography.
It was our aim to combine ¯uorous-phase separation with
lipase-catalyzed kinetic resolution in order to avoid chro-
matography. Recently, we were able to demonstrate that it is
possible to use a highly ¯uorinated acyl donor for the kinetic
resolution of racemic alcohols in the presence of a
lipase.14,15 We found that the ¯uorinated ester 116 is a very
useful acylating agent that is recognized by a lipase as its
substrate and most importantly, the lipase transfers the
¯uorinated acyl residue with high enantioselectivity onto
the fast-reacting enantiomer of a racemic alcohol affording
a mixture of a highly ¯uorinated ester ;fast-reacting
enantiomer) and an alcohol ;slow-reacting enantiomer)
;Scheme 1). Indeed, the enantiomer tagged with the ¯uori-
nated residue could be separated from the untagged one by a
simple liquid±liquid partition between a ¯uorous and an
organic phase with high ef®cacy demonstrating that the
labeled ¯uorous enantiomer was recognized selectively
from the per¯uorinated solvent, whereas the unlabeled
organic enantiomer remains in the organic solvent. Using
this technique the enantiomeric products could be separated
without chromatography.14,15
Due to the increasing demand of enantiomerically pure
intermediates enzyme-catalyzed kinetic resolutions particu-
larly lipase-catalyzed acylation, hydrolysis or alcoholysis
represent an easy access to enantiomerically pure or
enriched alcohols and their carboxylic esters.11±13 Lipases
are relatively inexpensive, highly selective, very robust and
easy to handle biocatalysts; many of them are commercially
available. Lipase-mediated acylation of a racemic alcohol
and alcoholysis or hydrolysis of its corresponding ester is
enantioconvergent. It means that always one and the same
enantiomer reacts faster. Selection of the conditions affords
the certain enantiomer either as the ester or the alcohol.
However, the degree of selectivity may be different.
In order to prevent side reactions, it is important that the
acyl donor requires a spacer consisting of at least two
methylene groups between the carbonyl group and the
per¯uorinated residue.
The enantiomer obtained as an alcohol and the other
obtained as an ester in lipase-catalyzed resolutions are
Keywords: ¯uorous phase; kinetic resolution; lipase; per¯uorinated
solvents.
Corresponding author. Tel.: 149-30-6392-2076; fax: 149-30-6392-
4103; e-mail: theil@asca-berlin.de; url: http://www.asca-berlin.de
After having demonstrated that lipase-catalyzed acylation
under kinetic resolution simultaneously labels the faster
reacting enantiomer with a ¯uorous tag very ef®ciently,
p
0040±4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020;02)00258-2

4086
S. M. Swaleh et al. / Tetrahedron 58 +2002) 4085±4089
Scheme 1. Lipase-catalyzed enantiomer-selective tagging with a ¯uorous acyl residue.
our next aim was to investigate the possibility of an enan-
tiomer-selective detagging procedure by subjecting highly
¯uorinated esters of racemic alcohols to a lipase-catalyzed
alcoholysis under deacylation of the fast-reacting
enantiomer and leaving the slow-reacting enantiomer
attached to the ¯uorous tag.
respectively ;Scheme 3). For rac-3f due to the higher rate
of conversion tert-butyl methyl ether was used instead of
acetonitrile as the solvent. After ®ltering off the lipase and
evaporation of the solvent under reduced pressure the ester
and the alcohol were separated by partition between
per¯uoro-n-hexane and methanol. This biphasic ¯uorous/
organic solvent system was found to be very useful in the
separation of the highly ¯uorinated esters from non-
¯uorinated alcohols.14,15
2. Results and discussion
The results of the kinetic resolutions are summarized in
Table 1. The outcomes from these reactions con®rm that
apart from rac-3c the other ¯uorinated esters were accepted
as substrates for CAL-B and resolved with excellent
enantiomer-selectivity. For all conversions, the E-value17
of the reaction was .200. The extractive separation by
partition between methanol and per¯uoro-n-hexane as
expected works very well, because for the resolution of
rac-2a, e, f by lipase-catalyzed acylation the separation of
the products in the biphasic system was already
achieved.14,15 In a control experiment ;S)-3d and ;R)-2d
were separated conventionally by ¯ash chromatography
showing almost identical results compared with the
The racemic ¯uorinated esters rac-3a±f were prepared by
acylation of the corresponding alcohols rac-2a±f with the
acid chloride 416 ;Scheme 2) which already has been used
for the synthesis of the ¯uorous acyl donor 1 depicted in
Scheme 1.14,15
The racemic esters rac-3a±e were subjected to a lipase-
catalyzed alcoholysis in acetonitrile or tert-butyl methyl
ether for rac-3f in the presence of Candida antarctica B
lipase ;CAL-B) and four equivalents of n-butanol yielding
except for the bromoindanol derivative rac-3c the ¯uori-
nated esters ;S)-3a, b, d, e or ;1S,4R)-3f and the non-
¯uorinated alcohols ;R)-2a, b, d, e or ;1R,4S)-2f,
Scheme 2. Synthesis of the racemic esters.

S. M. Swaleh et al. / Tetrahedron 58 +2002) 4085±4089
4087
Scheme 3. Lipase-catalyzed enantiomer-selective detagging.
extractive separation. Remarkably, the alcoholysis of rac-3f
shows a much higher selectivity than the acylation of rac-2f
where the E-value was only 7.6 as demonstrated recently.15
per¯uoroundecanoate in the aqueous phase, respectively.
Due to instability of silylalkynes under alkaline conditions
cleavage of the ester ;S)-3e was performed by acid-
catalyzed methanolysis and subsequent partition between
methanol containing ;S)-2e and per¯uoro-n-hexane contain-
ing the ¯uorinated species.
The reason for the resistance of rac-3c towards alcoholysis
in the presence of CAL-B is very likely the sterical
hindrance at the reacting stereogenic center preventing the
formation of the necessary transition state leading to the
formation of the acyl±enzyme intermediate which ®nally
must be attacked by n-butanol. Changing of the standard
conditions ;acetonitrile, room temperature) by using other
solvents such as chloroform, toluene, diisopropyl ether and
tert-butyl methyl ether at room temperature or in some cases
at 508C did not convert rac-3c.
The ¯uorine content of the slow-reacting enantiomers in all
cases was suf®ciently high for an ef®cient extractive
separation from the non-¯uorinated enantiomer as already
demonstrated previously.15
3. Conclusion
The ¯uorous phases contain besides the esters ;S)-3a, b, d, e
or ;1S,4R)-3f butyl 2H,2H,3H,3H-per¯uoroundecanoate
which was removed in the cases of ;S)-3a, b, d or
;1S,4R)-3f by saponi®cation with lithium hydroxide and
subsequent partition of the reaction mixture in an organic/
aqueous biphasic system affording the alcohols ;S)-2a, b, d
or ;1S,4R)-2f in the organic and lithium 2H,2H,3H,3H-
We were able to demonstrate that highly ¯uorinated
carboxylic esters of racemic alcohols are substrates for
lipase from C. antarctica B. Enantiomer-selective
alcoholysis of this type of esters selectively cleaves the
¯uorous label from the fast-reacting enantiomer affording
a mixture of an alcohol and the corresponding enantiomeric
¯uorinated ester. The products can be separated very
Table 1. Lipase-catalyzed kinetic resolution of the esters rac-3a±f by alcoholysis with n-butanol
Substrate
Time ;d)
Ester
% ee
Alcohol
% ee
E-value
Conversion ;%)
Con®guration
% Yielda
Con®guration
% Yield
rac-3a
rac-3b
rac-3c
rac-3d
rac-3e
rac-3f
0.8
8
No conversion
Sb
Sb
97
90
43
41
Rb
Rb
99
99
48
44
.200
.200
49
48
7
8
4
Sb
94 ;96c)
68
74
43
25e
42
Rb
99 ;99c)
99
99
44
17e
41
.200
.200
.200
49
41
43
Sd
Rd
1S,4Rd
1R,4Sd
a
Determined after saponi®cation or transesteri®cation ;cf. Section 4).
Assigned by comparison with one of the commercially available enantiomers.
The numbers in brackets correspond to the ee values determined after separation by ¯ash chromatography.
Assigned on the basis of the known [a]D-values of the free alcohols, cf. Ref. 18 for 2e and Ref. 19 for 2f.
The yields are low due to the high volatility of the alkynol 2.
b
c
d
e

4088
S. M. Swaleh et al. / Tetrahedron 58 +2002) 4085±4089
1
2.65 ;m, 4H,); 2.75 ;m, H); 4.72 ;m, 1H); 5.48 ;m, 1H);
5.98 ;m, 1H); 6.02 ;m, 1H).
ef®ciently by partition in the biphasic solvent system
methanol/per¯uoro-n-hexane avoiding chromatography.
These results represent another example for the successful
combination of ¯uorous techniques with lipase-catalyzed
kinetic resolutions where the ®nal separation procedure is
already integrated in the initial chemical transformation.
4.2. General procedure for the kinetic resolution
Solutions of the esters rac-3a±e ;5 mmol) in acetonitrile
;120 mL) or tert-butyl methyl ether for rac-3f ;120 mL)
were treated with n-butanol ;1.48 g, 1.83 mL, 20 mmol)
and CAL-B ;Chirazyme L-2, c.-f., lyo. from Roche
Diagnostics, Mannheim) ;8.0 g). The reaction mixture was
stirred at ambient temperature until the conversion reached
ca. 50% ;estimated by TLC). The lipase was removed by
®ltration and washed with acetone ;2£40 mL). The
combined ®ltrates were evaporated under reduced pressure.
The residue was dissolved in MeOH ;25 mL) and the result-
ing solution was extracted with n-C6F14 ;6£25 mL). The
organic phase was concentrated to dryness yielding the
pure ;R)-alcohols 2a, b, d, e and ;1R,4S)-2f.
4. Experimental
Starting materials, reagents and solvents were of com-
mercial grade. C. antarctica B lipase ;Chirazyme L-2,
c.-f., lyo.) was purchased from Roche Diagnostics,
Mannheim. The acid chloride 4 was prepared according to
Refs. 14,15. The alcohols rac-2e and rac-2f were prepared
according to Refs. 18,19, respectively. The 1H NMR spectra
were recorded on a Varian Gemini 300 at 300 MHz. HPLC
was carried out on a Merck-Hitachi system consisting of
L-6200A pump, L-4000 UV detector or Differential
Refractometer RI-71, and Chromato-Integrator D-2500.
The enantiomeric excesses of the alcohols were determined
by chiral HPLC ;see conditions below).
4.1. General procedure for the synthesis of the racemic
esters rac-3a±f
The ¯uorous phase was concentrated yielding a mixture of
butyl 2H,2H,3H,3H-per¯uoro undecanoate and the corre-
sponding ;S)-esters 3a, b, d, e or ;1S,4R)-3f, respectively.
To an ice-cold solution of the acid chloride 4 ;2.55 g,
5.0 mmol) and the alcohols rac-2a±f ;5.25 mmol) in
anhydrous THF ;5 mL) containing a catalytic amount of
4-DMAP ;20 mg) was added drop-wise anhydrous pyridine
;415 mg, 5.25 mmol) over 5 min. The cooling bath was
removed and the reaction mixture was allowed to reach
room temperature. After stirring for 2 h at room tempera-
ture, the precipitate was ®ltered, the ®ltrate was concen-
trated under reduced pressure and the residue was
partitioned between tert-butyl methyl ether ;10 mL) and
2N HCl ;2.5 mL). The organic layer was washed with
water ;2.5 mL), dried with Na2SO4 and concentrated.
Flash chromatography ;cyclohexane/ethyl acetateˆ10:1)
of the residue on silica gel yielded the pure esters rac-3a±f.
The esters ;S)-3a, b, d and ;1S,4R)-3f were cleaved by
saponi®cation as follows. The residue from the ¯uorous
phases were dissolved in a 1:1 mixture of THF/H2O
;40 mL) containing LiOH ;0.36 g, 15 mmol) and re¯uxed
for 3 h. Subsequently, the reaction mixture was diluted with
a mixture of cyclohexane ;70 mL) and tert-butyl methyl
ether ;30 mL), and the two distinct phases separated. The
aqueous phase was washed with a mixture of cyclohexane
;7 mL) and tert-butyl methyl ether ;3 mL). The combined
organic phases were concentrated to dryness to yield the
alcohols ;S)-2a, b, d or ;1S,4R)-2f, respectively.
4.1.1. rac-3a. Yield 78%; pale yellow oil; 1H NMR
;CDCl3): 1.56 ;d, Jˆ6.6 Hz, 3H); 2.47 ;m, 2H); 2.65 ;m,
2H); 5.96 ;q, Jˆ6.6 Hz, 1H); 7.26±7.39 ;m, 5H).
The ester ;S)-3e was cleaved as follows. The ¯uorous phase
was concentrated under reduced pressure. The residue was
dissolved in MeOH ;200 mL) containing a catalytic amount
of p-toluenesulfonic acid ;500 mg) and re¯uxed for 68 h.
The solvent was removed under reduced pressure and the
residue was dissolved in cyclohexane/tert-butyl methyl
ether ;200:50 mL). The resulting solution was ®ltered and
concentrated to dryness yielding a mixture of ;S)-2e and
methyl 2H,2H,3H,3H-per¯uoro undecanoate which were
separated after dissolution in MeOH ;25 mL) and extraction
with n-C6F14 ;6£25 mL). From the organic phase ;S)-2e was
isolated.
1
4.1.2. rac-3b. Yield 78%; white wax; mp 31±328C; H
NMR ;CDCl3): 2.10 ;m, 1H); 2.37±2.68 ;m, 5H); 2.90
;m, 1H); 3.13 ;m, 1H); 6.25 ;dd, Jˆ6.9 and 3.6 Hz, 1H);
7.20±7.45 ;m, 4H).
1
4.1.3. rac-3c. Yield 78%; white powder; mp 63±658C; H
NMR ;CDCl3): 2.40±2.70 ;m, 4H); 3.29 ;dd, Jˆ16.6 and
3.9 Hz, 1H); 3.72 ;dd, Jˆ16.6 and 6.6 Hz, 1H); 4.50 ;m,
1H); 6.37 ;d, Jˆ3.6 Hz, 1H); 7.25±7.42 ;m, 4H).
The enantiomeric excesses of the esters from the ¯uorous
phases were determined after saponi®cation or transesteri-
®cation, respectively, to the corresponding alcohols by
chiral HPLC.
1
4.1.4. rac-3d. Yield 82%; white powder; mp 66±678C; H
NMR ;CDCl3): 1.65 ;d, Jˆ6.6 Hz, 3H); 2.50 ;m, 2H); 2.68
;m, 2H); 6.10 ;q, Jˆ6.6 Hz, 1H); 7.40±7.92 ;m, 7H).
4.1.5. rac-3e. Yield 80%; pale yellow wax; mp 26±278C; 1H
NMR ;CDCl3): 0.17 ;s, 9H); 1.48 ;d, Jˆ6.6 Hz, 3H); 2.50
;m, 2H); 2.68 ;m, 2H); 5.53 ;q, Jˆ6.6 Hz, 1H).
4.1.6. rac-3f. Yield 72%; pale yellow oil; 1H NMR ;CDCl3):
0.08 and 0.09 ;2 s, 6H); 0.89 ;s, 9H); 1.60 ;m, 1H); 2.38±
Conditions for the determination of the enantiomeric excess
by HPLC on chiral phases:
+R)- and +S)-2a. Chiralpak ADw, n-heptane/n-
propanolˆ95:5, UV-detection at lˆ254 nm, ¯ow rateˆ
1.00 mL/min.

S. M. Swaleh et al. / Tetrahedron 58 +2002) 4085±4089
4089
+R)- and +S)-2b. Chiralcel ODw, n-heptane/n-propanolˆ
90:10, UV-detection at lˆ254 nm, ¯ow rateˆ1.00 mL/
min.
Â
5. Soos, T.; Bennett, B. L.; Rutherford, D.; Barthel-Rosa, L. P.;
Gladysz, J. A. Organometallics 2001, 20, 3079±3086.
6. Curran, D. P.; Hoshino, M. J. Org. Chem. 1996, 61, 6480±
6481.
+R)- and +S)-2d. Chiralcel OJw, n-hexane/2-propanolˆ
90:10, UV-detection at lˆ254 nm, ¯ow rateˆ1.50 mL/
min.
7. Wipf, P.; Reeves, J. T. Tetrahedron Lett. 1999, 40, 4649±
4652.
È
8. Rover, S.; Wipf, P. Tetrahedron Lett. 1999, 40, 5667±5670.
9. Luo, Z.; Williams, J.; Read, R. W.; Curran, D. P. J. Org.
Chem. 2001, 66, 4261±4266.
+R)- and +S)-2e. Chiralpak ADw, n-heptane/n-propa-
nolˆ99:1, RI-detection, ¯ow rateˆ0.60 mL/min.
+1S,4R)- and +1R,4S)-2f. Chiralcel ODw, n-heptane/n-propa-
10. Crich, D.; Neelamkavil, S. J. Am. Chem. Soc. 2001, 123,
7449±7450.
11. Theil, F. Chem. Rev. 1995, 95, 2203±2227.
12. Bornscheuer, U. T.; Kazlauskas, R. J. Hydrolases in Organic
Synthesis; Wiley/VCH: Weinheim, 1999.
nolˆ99:1, RI-detection, ¯ow rateˆ1.00 mL/min.
13. Carrea, G.; Riva, S. Angew. Chem. 2000, 112, 2312±2341
Angew. Chem. Int. Ed. Engl. 2000, 39, 2226±2254.
14. Hungerhoff, B.; Sonnenschein, H.; Theil, F. Angew. Chem.
2001, 113, 2550±2552 Angew. Chem. Int. Ed. Engl. 2001,
40, 2492±2494.
Acknowledgements
This work was supported by the Deutsche Forschungs-
gemeinschaft ;grant: TH 562/3-1) and the Fonds der
Chemischen Industrie.
15. Hungerhoff, B.; Sonnenschein, H.; Theil, F. J. Org. Chem.
2002, 67, 1781±1785.
16. Synthesized from the commercially available 1H,1H,2H,2H-
hepta¯uorodecanol according to Refs. 14,15.
17. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C.-J. J. Am.
Chem. Soc. 1982, 104, 7294±7299.
References
1. Curran, D. P. Green Chem. 2001, G3±G7.
2. Curran, D. P. Angew. Chem. 1998, 110, 1231±1255 Angew.
Chem. Int. Ed. Engl. 1998, 37, 1174±1196.
18. Burgess, K.; Jennings, L. D. J. Am. Chem. Soc. 1991, 113,
6129±6139.
Â
Â
3. Horvath, I. T.; Rabai, J. Science 1994, 266, 72±75.
4. Richter, B.; Spek, A. L.; van Koten, G.; Deelman, B.-J. J. Am.
Chem. Soc. 2000, 122, 3945±3951.
19. Curran, T. T.; Hay, D. A.; Koegel, C. P.; Evans, J. C.
Tetrahedron 1997, 53, 1983±2004.