Resolution of a chiral alcohol through lipase-catalyzed transesterification of its mixed carbonate by poly(ethylene glycol) in organic media

Article abstract of DOI:10.1016/S0957-4166(00)00068-9

A simple approach to the resolution of chiral alcohols through a lipase- catalyzed transesterification of one enantiomer of the corresponding trifluoroethyl carbonate by a low molecular weight poly(ethylene glycol), PEG, is described. The method was demonstrated through resolution of (RS)- sec-phenethyl alcohol. The alcohol was converted to its 2,2,2-trifluoroethyl carbonate, 2, and the (R)-enantiomer was selectively transesterified with PEG in warm diisopropyl ether using porcine pancreas lipase, PPL, as the catalyst. The two carbonate enantiomers were easily separated by cooling and filtering off the solid PEG having the (R)-alcohol covalently attached. Hydrolysis of the unchanged (S)-carbonate was achieved in dilute aqueous base, and the enantiomeric excess of the (S)-alcohol was found to be 80% by NMR in the presence of the chiral shift reagent Eu(hfc)₃. Methanolysis of the modified (R)-PEG carbonate yielded (R)-sec-phenethyl alcohol having enantiomeric excess=96% by NMR with Eu(hfc)₃. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00068-9

Tetrahedron: Asymmetry 11 (2000) 1279±1288
Resolution of a chiral alcohol through lipase-catalyzed
transesteri®cation of its mixed carbonate by
poly(ethylene glycol) in organic media
Lisa J. Whalen and Cary J. Morrow*
Department of Chemistry, University of New Mexico, Albuquerque, NM 87131, USA
Received 4 January 2000; accepted 3 February 2000
Abstract
A simple approach to the resolution of chiral alcohols through a lipase-catalyzed transesteri®cation of
one enantiomer of the corresponding tri¯uoroethyl carbonate by a low molecular weight poly(ethylene
glycol), PEG, is described. The method was demonstrated through resolution of (RS)-sec-phenethyl alcohol.
The alcohol was converted to its 2,2,2-tri¯uoroethyl carbonate, 2, and the (R)-enantiomer was selectively
transesteri®ed with PEG in warm diisopropyl ether using porcine pancreas lipase, PPL, as the catalyst. The
two carbonate enantiomers were easily separated by cooling and ®ltering o€ the solid PEG having the (R)-
alcohol covalently attached. Hydrolysis of the unchanged (S)-carbonate was achieved in dilute aqueous
base, and the enantiomeric excess of the (S)-alcohol was found to be 80% by NMR in the presence of the
chiral shift reagent Eu(hfc)3. Methanolysis of the modi®ed (R)-PEG carbonate yielded (R)-sec-phenethyl
alcohol having enantiomeric excess=96% by NMR with Eu(hfc)3. # 2000 Elsevier Science Ltd. All rights
reserved.
1. Introduction
The resolution of enantiomers by enantioselective hydrolysis with an enzyme has been exploited
extensively for many years.1 5 However, the discovery that many enzymes work well in low
polarity organic solvents6 10 has allowed enzymatic resolution to be extended to reactions such
as esteri®cation and transesteri®cation. These approaches have been reviewed by Koskinen and
Klibanov4 and by Drauz and Waldmann.5 Resolution by enzymatic transesteri®cation in an
organic solvent solves some of the problems associated with the enzymatic hydrolyses such as the low
solubility of many organic compounds in water, the diculty of recovering the enzyme for reuse,
* Corresponding author. Tel: 505-277-0279; fax: 505-277-2609; e-mail: cmorrow@unm.edu
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00068-9

1280
L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
and the requirement for adjusting the pH as the reaction progresses.11 Also, an increase in enzyme
stability has been reported when it is used in an organic solvent, so higher reaction temperatures
are tolerated.7 Resolution of a racemic ester may be carried out by having one enantiomer of the
ester transesteri®ed with an achiral alcohol so that a di€erence in chemical properties between
the two ester enantiomers is established.4,5 More commonly, a racemic alcohol has been resolved
by having one enantiomer of the alcohol transesterify an achiral ester, so that the enantiomer is
converted to an ester and its enantiomer is left unchanged as an alcohol.4,5 The di€erence in the
chemical and physical properties that is established enables separation of the two enantiomers
by chromatography.
Nearly a decade ago, we showed that, if a high molecular weight polymeric alcohol such as
poly(ethylene glycol) is used as the achiral transesterifying alcohol, one enantiomer of a chiral
ester can be converted to an ester of poly(ethylene glycol) (PEG),12 while its mirror image is left
as the original ester.13 A signi®cant di€erence in chemical properties is established between the
enantiomeric pair and separation is readily achieved without the need for chromatography. If the
enzymatic transesteri®cation takes place in an appropriate solvent, such as diisopropyl ether,
separation of the enantiomers can be achieved by methodology as simple as cooling the mixture
to 0ꢀC and ®ltering o€ the solid PEG carrying the modi®ed enantiomer. The unchanged ester
enantiomer remains dissolved in the ®ltrate.13 We have now shown that this approach, which is
termed a `phase switch' by Curran,14 can be easily extended to the resolution of chiral alcohols if
the racemic alcohol is converted to a mixed carbonate functional group.
Limited work has been reported on enzyme-catalyzed reactions of carbonates. In 1989,
Abramowicz and Keese reported lipase-catalyzed transesteri®cation of diphenyl carbonate with a
variety of alcohols.15 Gotor and co-workers subsequently demonstrated the use of enzyme-
catalyzed transformations utilizing carbonates leading to progress on the kinetic resolution of
chiral aminoalcohols16 and chiral vinyl carbonates,17 and in the enzymatic synthesis of chiral
carbonates18 20 and chiral carbamates.18,21 Similarly, the use of enzyme-catalyzed enantio-
selective hydrolysis of carbonates22 24 in the synthesis of optically active diols25 and diol deriva-
tives26 has been reported by various groups.
For the demonstration study reported here, an extensively studied chiral alcohol, (RS)-sec-
phenethyl alcohol, was chosen. This alcohol has been previously resolved using enzyme-catalyzed
enantioselective transesteri®cation involving a variety of acyl sources.27 However, each of these
methods required a tedious separation such as chromatography to achieve the actual resolution.
2. Results and discussion
In developing the current strategy for resolution of (RS)-sec-phenethyl alcohol, incorporation
of two features was sought: (1) modi®cation of the chiral alcohol into a functional group that
could be transesteri®ed enantioselectively by porcine pancreas lipase (PPL); (2) use of a low
molecular weight PEG as the transesterifying alcohol so that a strong di€erence in solubility
properties could be imparted to one of the enantiomers.
The (RS)-sec-phenethyl alcohol was initially converted to urethane 1 by treating it with 1,10-
carbonyldiimidazole as shown in the ®rst reaction of Eq. (1). It was hoped that PPL would
replace the imidazole moiety of this urethane with PEG enantioselectively. However, control
experiments run without the enzyme showed that displacement of the remaining imidazole in 1 by
PEG occurred non-catalytically, and this avenue was abandoned.

L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
1281
ꢁ1†
Enzyme-catalyzed resolutions of chiral esters in which the stereogenic center is located on the
carbonyl moiety of the ester at a site to the carbonyl functional group are relatively rare but often
very successful, especially when PPL or porcine liver esterase (PLE) is used as the catalyst.11,28 41
Our previous success using a 2,2,2-trihaloethyl ester of a b-chiral carboxylic acid in an enantio-
selective PPL-catalyzed transesteri®cation reaction13 suggested conversion of 1 to the mixed
carbonate (RS)-2 in which one side of the carbonate is the chiral alcohol to be resolved and the
other side is the 2,2,2-trihaloethyl leaving group. This structure places the stereogenic center of
the alcohol b to the carbonyl of the carbonate and takes advantage of PPL's ability to catalyze
the transesteri®cation of such substrates enantioselectively. Carbonate (RS)-2 was synthesized
directly from (RS)-1 by treating it with 2,2,2-tri¯uoroethanol in the presence of DMAP as shown
in the second step of Eq. (1). The presence of chirality in 2 and its impact on a site at some
distance from the center of chirality was readily observed by the appearance of two doublets of
quartets for the diastereotopic methylene protons in the tri¯uoroethyl group.
The resolution was achieved by stirring a mixture of (RS)-2 dissolved in diisopropyl ether with
PEG having a molecular weight of about 1500 and dried PPL at 45ꢀC as shown in Eq. (2). Under
these conditions, the PEG is molten, but only partially dissolved, while the PPL is a separate solid
phase. To allow the reaction to be monitored by VPC, o-dichlorobenzene (o-DCB) was added as
an internal standard. Aliquots were drawn for analysis at the beginning of the reaction and
periodically afterwards. The reaction was allowed to continue until the ratio of the VPC peaks for
2 and for o-DCB ceased changing at about half of its original value, approximately nine days
being required. A control experiment run in parallel, but lacking the PPL, con®rmed that the
catalyst was required for transesteri®cation to take place. A mixture containing (R)-3, the PPL,
and any unchanged PEG was separated from the unchanged (S)-2 by cooling the mixture to 0ꢀC
to freeze the PEG and acylated PEG then ®ltering. Dissolving the PEG and (R)-3 from the solid
phase in CH2Cl2 and again ®ltering removed the enzyme.
Repetition of the experiment without o-DCB had an interesting result. The reaction described
by Eq. (2) was stopped at nine days, assuming that the previous analysis conducted with o-DCB was
valid. However, the yield of the unchanged carbonate phase was higher than expected, indicating that
the reaction had not reached completion. It is suspected that o-DCB may be acting as a cosolvent
with isopropyl ether, possibly facilitating import of PEG to the active site of the enzyme.
ꢁ2†
The 1H NMR spectrum of the fraction containing the (R)-3 and unchanged PEG displayed two
signi®cant peaks: (1) a complex multiplet at 4.26 ppm, down®eld of the PEG methylene envelope,

1282
L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
showed the presence of acylated PEG in the insoluble fraction. At 500 MHz this signal is resolved
into two doublets of triplets due to the diastereotopic methylene Hs that result from bonding the
chiral carbonate group to the PEG; (2) a pair of doublets of quartets at 4.5 ppm which was
known to arise from the diastereotopic methylene hydrogens of the tri¯uoroethyl group in 2,
and which was expected to be absent from the spectrum of this fraction. Two possibilities were
suggested: either the separation had not been complete or the transesteri®cation had replaced the
sec-phenethyl group rather than the tri¯uoroethyl group. The separation step using isopropyl
1
ether was repeated, and H NMR con®rmed that the source of the quartets was unreacted
carbonate, presumably (S)-2, that had become entrained in the PEG/enzyme fraction.
1
The H NMR spectrum of product in the combined ®ltrates from the separations con®rmed
that this fraction contained unreacted carbonate 2. A very tiny peak at 3.6 ppm, corresponding to
the envelope of signals from the many PEG methylene groups, was the only evidence of PEG in
the ®ltrate fraction. Thus, the separation of the PEG having bound (R)-3 from the ®ltrate con-
taining unchanged carbonate (S)-2 was excellent.
Signals indicative of a small amount of sec-phenethyl alcohol were also present in the spectrum
of the product from the ®ltrate fraction expected to contain (S)-2. Hydrolysis rather than trans-
esteri®cation can occur if the water content of the enzyme is too high.15,42 If enzymatic hydrolysis
is the source of the sec-phenethyl alcohol, it is expected that the enantiomeric excess of the
alcohol should be high and consistent with the selectivity of PPL. Thus, it should have the (R)
con®guration rather than the (S) con®guration expected for the unchanged carbonate presumed to
comprise the rest of this fraction, and separating the two was essential to evaluating the enzyme's
selectivity in transesterifying (R)-2. Using column chromatography, the alcohol was separated
from (S)-2 but the presence of another minor side product, suspected to be the sec-phenethyl
carbonate of ethylene glycol, interfered with the analysis of the alcohol's con®guration and
enantiomeric excess. E€orts to prevent the hydrolysis side reaction have not yet been successful.
Conversion of both products (R)-3 and (S)-2 back to the corresponding sec-phenethyl alcohol
enantiomers was carried out to determine the enantiomeric excess, the selectivity of the enzyme,
and the overall success of the method for e€ecting an alcohol resolution. Hydrolysis of the
unchanged carbonate (S)-2 was easily achieved with aqueous alkali and 1H NMR analysis of the
product proved that, as expected, (S)-sec-phenethyl alcohol was the product formed. Methan-
1
olysis of (R)-3 was also successful as demonstrated by TLC analysis and con®rmed by H NMR
analysis of the isolated product. The methanolysis did not yield the methyl carbonate of sec-
phenethyl alcohol, as was expected, but instead gave back (R)-sec-phenethyl alcohol directly.
Surprisingly, there was no evidence of the methyl carbonate of PEG having formed. Rather, VPC
analysis of the crude reaction mixture prior to evaporating the methanol showed that methano-
lysis on both sides of the carbonate had occurred, and that dimethyl carbonate was a product of
the reaction along with the (R)-sec-phenethyl alcohol and unacylated PEG.
1
The enantiomeric excess of the two alcohols was determined by their H NMR analysis in the
presence of the chiral lanthanide shift reagent (tris-3-hepta¯uorobutyryl-d-camphorato)europium(III)
[Eu(hfc)3].43 The enantiomeric composition of non-racemic sec-phenethyl alcohol has been
determined previously with this shift reagent.43 In the present work, control experiments were
carried out with racemic and partially resolved sec-phenethyl alcohol to verify the technique.
Resolution of the methine proton of (R)- and (S)-sec-phenethyl alcohol was achieved with a ÁÁd
of 0.3 ppm at approximately 13 ppm by using a molar ratio of Eu(hfc)3 to alcohol of 0.4.
Approximately 2% of the minor enantiomer could be detected. The recovered (R)-alcohol from
methanolysis of (R)-3 showed no detectable amount of the (S)-enantiomer, so its enantiomeric

L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
1283
excess is at least 96%. To prove that the (S)-enantiomer was missing, alcohol known to be enriched
in (S)-enantiomer was added to the NMR sample and the NMR experiment repeated. The
appearance of a small absorption in the region where the signal from the (S)-enantiomer was
expected con®rmed that PPL had selected the (R)-enantiomer of the racemic carbonate for reac-
tion with the PEG to give 3 and left behind the (S)-enantiomer as carbonate 2. This result is
consistent with con®gurational selectivities reported in the literature for PPL-catalyzed trans-
esteri®cation of a b chiral ester.13,36,37,40 The E value of the reaction was found to be 35 with a
54.5% conversion of (RS)-2 to (R)-3.y
The enantiomeric excess of (S)-sec-phenethyl alcohol was determined to be 80%. The alcohol
sample used was prepared from carbonate that was hydrolyzed only after removing the presumed
(R)-sec-phenethyl alcohol side product by column chromatography. Similar analysis of a sample
from which the (R)-sec-phenethyl alcohol had not been removed prior to hydrolysis showed an
enantiomeric excess of 54%. This result provides evidence that the observed sec-phenethyl
alcohol side product arises from PPL-catalyzed hydrolysis of (R)-2 to (R)-sec-phenethyl alcohol.
3. Conclusion
This study has demonstrated the viability of enantioselective, enzyme-catalyzed transesteri®ca-
tion of a mixed carbonate with poly(ethylene glycol) as a method of resolving a racemic chiral
alcohol. The separation when isopropyl ether is used as the solvent is excellent, but there is some
tendency for unchanged carbonate enantiomer to be entrained in the PEG-containing fraction
requiring that the simple separation protocol be repeated with fresh diisopropyl ether. The source
of the water that leads to some hydrolysis rather than transesteri®cation is almost certainly the
matrix in which the enzyme is supplied. PPL is an attractive enzyme for the method because of its
high selectivity in transesterifying b chiral esters as well as its stability and very low cost. How-
ever, attempts to reduce the available water content have not yet been fully successful.
4. Experimental
4.1. General
Anhydrous diisopropyl ether, iPr2O, was purchased from Aldrich Chemical Co. in a Sure-
TM
Seal bottle (stabilized with BHT, packaged under N2) and used as received (caution: unstabil-
ized iPr2O can form explosive peroxides upon exposure to air,44 but this has not been a problem
in the inert atmosphere reactions we have been carrying out). As a further precaution, a sample of
the iPr2O being used was periodically tested for peroxide formation with moist KI/starch paper
(available from Aldrich). All tests were negative. Poly(ethylene glycol), PEG, of average molec-
ular weight 1500 (Aldrich cat. no. 20,243-6) and porcine pancreas lipase, PPL (25% protein,
activity=50 units/mg using triacetin at pH 7.4, 179 units/mg using olive oil at pH 7.7, Sigma cat.
no. L-3126), were treated as described below. All reactions were stirred magnetically under a dry
N2 atmosphere. Other chemicals were the best available grade, and were used as received.
y
The E value reported is based on the e.e.s for sec-phenethyl alcohol and requires the assumption that there is no
change in e.e. in steps subsequent to the enzymatic step.

1284
L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
Organic extracts were dried with a 1:1 mixture of Na2SO4 and MgSO4. NMR spectra were
recorded at 250 MHz except where otherwise noted. VPC analyses were carried out on a fused
silica column having a bonded liquid phase of cross-linked methyl silicone. TLC analyses were
carried out on silica gel coated plastic plates using 4:1 (v/v) hexanes:ethyl acetate as eluent.
4.2. (RS)-sec-Phenethyl 1H-imidazole-1-carboxylate [(RS)-1]
In a modi®cation of the method of Ohta et al.45 25 mL methylene chloride was added to a dry
250 mL ¯ask followed, with stirring, by 17.8 g (110 mmol) 1,10-carbonyldiimidazole, then an
additional 10 mL methylene chloride, and, ®nally, 12.2 g (100 mmol) (RS)-sec-phenethyl alcohol
dissolved in 10 mL methylene chloride. Upon addition of the alcohol the solution became light
yellow and boiled for a few minutes. A yellow precipitate of imidazole could be seen in the reac-
tion ¯ask after 10 min. After 4 h stirring at room temperature, VPC analysis showed that the
alcohol had been consumed. The mixture was extracted with three 20 mL portions of distilled
water to remove the imidazole and dried (note: if the imidazole is not removed by the aqueous
extraction, it interferes with the next reaction). Evaporation of the methylene chloride yielded
21.12 g (98% yield) of a light yellow oil that solidi®ed into a brittle white solid while stored in the
1
freezer. H NMR (CDCl3): ꢀ 1.73 (d, J=6.6 Hz, 3H), 6.07 (q, J=6.6 Hz, 1H), 7.04 (s, 1H), 7.37
(m, 6H), 8.15 (s, 1H); 13C NMR (CDCl3): ꢀC 147.8, 139.5, 136.9, 130.4, 128.6, 128.5, 125.9, 116.9,
76.9, 21.7.
4.3. (RS)-sec-Phenethyl 2,2,2-tri¯uoroethyl carbonate [(RS)-2]
Following the procedure given above, 8.9 g (55 mmol) 1,10-carbonyldiimidazole and 6.1 g (50
mmol) (RS)-sec-phenethyl alcohol were used to prepare (RS)-1. After extracting with three 15 mL
portions of distilled water and drying, the solution was transferred to a new 250 mL ¯ask, and
12.31 g (123 mmol) 2,2,2-tri¯uoroethanol dissolved in 10 mL methylene chloride was added
followed by 0.61 g (5 mmol) DMAP. After 20 h at rt VPC analysis showed that the mixture was
>95% carbonate (RS)-2. The mixture was then stirred with 20 mL of 0.05 M HCl and 2.9 M
HCl was added dropwise with stirring until pH=2, 17 mL being required. The phases were
separated and the organic phase was extracted successively with 20 mL each of 0.05 M HCl,
distilled water, saturated aqueous sodium bicarbonate, and two 20 mL portions of saturated
sodium chloride solution. After drying, the methylene chloride was evaporated to give 10.52 g
(84.8% yield) of a clear liquid that could be further puri®ed by vacuum distillation, bp 44ꢀC/
1
0.001 mm Hg. H NMR (CDCl3): ꢀ 1.62 (d, J=6.6 Hz, 3H), 4.36 (dq, JH,H=12.6 Hz, JH,F=8.3
Hz, 1H), 4.55 (dq, JH,H=12.6 Hz, JH,F=8.3 Hz, 1H), 5.75 (q, J=6.6 Hz, 1H), 7.34 (m, 5H); 13C
NMR (CDCl3): ꢀC 153.3, 140.2, 128.6, 128,4, 125.9, 122.6 (q, JC,F=277 Hz), 77.9, 63.15 (q,
JC,F=37 Hz), 21.9. Mass spectrum: m/z 248 (12.7), 105 (60.0), 104 (100), 83 (20.9), 77 (22.0), 51
(13.1). Anal. calcd for C11H11F3O3: C, 53.23; H, 4.47; Found: C, 53.03; H, 4.48.
4.4. Preparation of the poly(ethylene glycol) (PEG) substrate
To a dried 500 mL round-bottomed boiling ¯ask was added 82.4 g PEG having an average
molecular weight of 1500. Approximately 250 mL toluene was added and the mixture distilled at
an atmospheric pressure of 630 mm Hg. Toluene and water co-distilled at 103ꢀC, with the
distillate appearing cloudy until approximately 1 h later, when a few drops collected in a separate

L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
1285
¯ask were clear. Upon cooling overnight, the PEG precipitated from the toluene. The toluene was
decanted and was replaced with 150 mL of iPr2O. To remove low molecular weight oligomers, the
mixture was heated at 45ꢀC with stirring in a water bath for 30 min. The ¯ask was then placed in
the freezer for 30 min and the iPr2O was decanted. The process was repeated with 150 mL of fresh
iPr2O except that the ¯ask allowed to return to room temperature slowly with rapid stirring to
fracture the PEG into a ®ne white powder. The powder was ®ltered o€ and dried in vacuo for 2 h,
1
yielding 79.8 g (97%) of PEG. H NMR (CDCl3): ꢀ 1.7 (s), 3.6 (br s).
4.5. Resolution of (RS)-sec-phenethyl 2,2,2-tri¯uoroethyl carbonate [(RS)-2]
To a dry 250 mL ¯ask was added 20 mL iPr2O followed by 1.25 g (5.0 mmol) (RS)-2 dissolved
in 5 mL iPr2O and 0.046 g (0.31 mmol) ortho-dichlorobenzene (o-DCB). A sample of the mixture
was withdrawn and the ratio of the area below the (RS)-2 peak in the VPC to the area under the
o-DCB peak was measured at time zero. To the ¯ask were then added 4.87 g (3.25 mmol) of the
PEG prepared above and 2.51 g PPL that had been dried in vacuo for three days over P2O5.
The mixture was heated to 45ꢀC with stirring in an oil bath. The PEG and PPL formed a separate
phase which collected at the bottom of the ¯ask once the polymer had melted. Periodically, the
¯ask was cooled to 0ꢀC in an ice/water bath and an aliquot withdrawn for VPC analysis of the
(RS)-2:o-DCB peak ratio. After nine days the analysis showed that the ratio had decreased
by half. The reaction mixture was placed in a freezer for 30 min, then the solid phase comprising
the PPL, PEG, and acylated PEG was separated from the ®ltrate containing unchanged carbonate
and iPr2O by suction ®ltration.
4.6. Recovery of (R)-3
The solid collected in the ®ltration was washed with cold iPr2O then was stirred with 20 mL
CH2Cl2 and the insoluble material, presumed to be PPL, was removed by centrifugation followed
by suction ®ltration of the supernatant with a fritted glass funnel. The CH2Cl2 ®ltrate had a
yellowish±orange color that remained in the waxy solid isolated after evaporation of the solvent.
1
In the H NMR spectrum of the solid, the characteristic pair of doublets of quartets at 4.5 ppm
from the tri¯uoroethyl group of 2 showed that further puri®cation was needed. The waxy solid
was dissolved in 12.5 mL iPr2O, and the solution was heated to 45ꢀC for 10 min with stirring so
that the PEG could melt and unchanged 2 could dissolve. The ¯ask was then placed in a freezer
for 1 h and the solid phase separated from the ether phase by suction ®ltration. This process of
heating, stirring, cooling, and ®ltering was repeated three times. The combined ®ltrates were
1
concentrated by evaporation and con®rmed by H NMR to consist of unreacted 2. The puri®ed
solid PEG fraction was dissolved in methylene chloride and ®ltered through a fritted glass funnel,
1
then the solvent removed to obtain 4.75 g of (R)-3 and unchanged PEG. H NMR analysis indi-
cated that the tri¯uoroethyl group was absent but showed a strong signal from the PEG at 3.6
ppm. Signals for the sec-phenethyl group appeared at 1.6 (d), 5.7 (q), and 7.3 (m). At 4.2 was a
multiplet arising from the diastereotopic methylene Hs in the esteri®ed PEG.
4.7. Recovery of (S)-2
The 1H NMR spectrum of the product recovered by evaporating the original iPr2O ®ltrate was
identical with that of the original, racemic carbonate but VPC analysis showed that the carbonate

1286
L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
contained 7% sec-phenethyl alcohol. However, only a small peak at 3.6 ppm was seen as evidence
of PEG in this fraction. After column chromatography on 200 mesh silica gel using methylene
chloride as the solvent to remove the sec-phenethyl alcohol from the mixture, 0.75 g carbonate
(S)-2 having 1H NMR data identical with those of (RS)-2 was isolated.
4.8. Hydrolysis of (S)-sec-phenethyl 2,2,2-tri¯uoroethyl carbonate [(S)-2]
In a 100 mL ¯ask were placed 0.75 g (3.0 mmol) of the (S)-2 isolated above and 5 mL THF.
Dry N2 was bubbled through the solution with a Pasteur pipette while 9.0 mL of 1 M NaOH was
added. The ¯ask was then ¯ushed with N2 and the mixture allowed to stir at rt. After 18 h, TLC
showed the disappearance of (S)-2, so 1.5 mL of 2.9 M HCl was added to neutralize the alkali.
Evaporation of the THF and tri¯uoroethanol gave a bright yellow mixture which was dissolved
in 20 mL CH2Cl2. The solution was extracted with three 5 mL portions of half-saturated NaCl
solution, then dried. Evaporation of the CH2Cl2 yielded 0.204 g (66% yield) of a bright yellow
oil. 1H NMR analysis showed peaks from sec-phenethyl alcohol: 1.4 (d, J=5.7 Hz), 4.9 (q, J=6.4
1
Hz), 7.3 (m). For H NMR experiments with Eu(hfc)3, a 0.40 M solution of the isolated (S)-sec-
phenethyl alcohol in CDCl3 was prepared and 0.5 mL of the solution was added to an NMR tube
followed by 0.1 mL of a 0.194 M solution of Eu(hfc)3 in CDCl3. After allowing 15 min for equi-
1
libration, a H NMR spectrum was obtained at 250 MHz and examined for evidence of signal
resolution. Additional 0.1 mL increments of the Eu(hfc)3 solution were added, allowed to reach
1
equilibrium, and H NMR spectra obtained after each addition until baseline resolution of the
signal for the methine proton of sec-phenethyl alcohol was achieved at ꢂ13 ppm with a ÁÁd of
0.3 ppm.
4.9. Methanolysis of PEG-carbonate [(R)-3]
To a 100 mL ¯ask containing 4.75 g (R)-3 was added 15 mL CH3OH. The contents were stirred
magnetically to dissolve the solid then 0.304 g (5.6 mmol) NaOCH3 was added with stirring.
Immediately the mixture turned light yellow and cloudy, but after 15 min, the cloudiness dis-
appeared. Dry N2 was bubbled through the solution for 2 min using a Pasteur pipette, then the
¯ask was ¯ushed with dry N2 and the mixture allowed to stir at rt for two days. The reaction was
stopped by neutralizing the mixture with 1 M H2SO4 in CH3OH, 0.9 mL being required. The
1
CH3OH was evaporated and the residual CH3OH removed in vacuo. H NMR analysis of the
residue showed peaks indicative of non-acylated PEG and sec-phenethyl alcohol. The absence of
two doublets of triplets at 4.26 shows removal of PEG from the carbonate. The solid was dis-
solved in 10 mL CH2Cl2 then gravity ®ltered to remove the Na2SO4 from the neutralization step.
The CH2Cl2 was evaporated and replaced with 10 mL iPr2O. The mixture was heated at 45ꢀC
with stirring, during which time most of the PEG melted and formed a separate phase at the
bottom of the ¯ask. After 10 min of heating and stirring, the ¯ask was placed in the freezer for 20
min. The solidi®ed PEG was ®ltered o€ and the procedure repeated with 10 mL of fresh iPr2O.
The ®ltrates were combined in a tared 100 mL ¯ask and the iPr2O was evaporated with the last
traces being removed on the vacuum manifold to yield 0.193 g (63%) of a clear, yellowish oil. 1H
NMR analysis of the product showed peaks from sec-phenethyl alcohol but no singlets indicative
of a methyl carbonate. For 1H NMR experiments with Eu(hfc)3, the same procedure as described
above was followed. When the methine proton signal had been shifted to ꢂ13 ppm and no reso-
lution had been observed, 46 mL of a 1.3 M solution of (R)-sec-phenethyl alcohol having

L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
1287
e.e.=54% was added to the NMR tube containing the recovered (R)-alcohol and shift reagent,
and allowed to stand for 3 h. The 1H NMR experiment was repeated and a new quartet 0.25 ppm
up®eld of the original quartet appeared.
Acknowledgements
This research was supported in part by the NSF Research Experiences for Undergraduates
program, grant CHE-9531414 (L.J.W.), by NSF grant no. DMR-8912065, and by the University
of New Mexico Research Allocations Committee. The authors also wish to acknowledge the NSF
for contributing to the purchase of the 500 MHz NMR spectrometer required for this research
through grant no. CHE-9709174.
References
1. Jones, J. B. Tetrahedron 1986, 42, 3351.
2. Biocatalysts in Organic Synthesis; Tramper, J.; van der Plas, H. C.; Linko, P., Eds.; Elsevier: New York, 1985.
3. Enzyme Chemistry: Impact and Applications; Suckling, C. J., Ed.; Chapman and Hall: New York, 1984.
4. Enzymatic Reactions in Organic Media; Koskinen, A. M. P.; Klibanov, A. M., Eds.; Blackie: New York, 1996.
5. Enzyme Catalysis in Organic Synthesis: A Comprehensive Handbook; Drauz, K.; Waldmann, H., Eds.; VCH: New
York, 1995.
6. Cambou, B.; Klibanov, A. M. J. Am. Chem. Soc. 1984, 106, 2687.
7. Zaks, A.; Klibanov, A. M. Science 1984, 224, 1249.
8. Gat®eld, I. L. Ann. N.Y. Acad. Sci. 1984, 568.
9. Zaks, A.; Klibanov, A. M. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 3192.
10. Klibanov, A. M. Chemtech 1986, 354.
11. Wilson, W. K.; Baca, S. B.; Barber, Y. J.; Scallen, T. J.; Morrow, C. J. J. Org. Chem. 1983, 48, 3960.
12. For more applications of PEG in organic synthesis, see: Gravert, D. J.; Janda, K. D. Chem. Rev. 1997, 97, 489.
13. Wallace, J. S.; Reda, K. B.; Williams, M. E.; Morrow, C. J. J. Org. Chem. 1990, 55, 3544.
14. Curran, D. P. Angew. Chem., Int. Ed. 1998, 37, 1174.
15. Abramowicz, D. A.; Keese, C. R. Biotechnol. Bioeng. 1989, 33, 149.
16. Maestro, A.; Astorga, C.; Gotor, V. Tetrahedron: Asymmetry 1997, 8, 3153.
17. Pozo, M.; Gotor, V. Tetrahedron 1993, 49, 10725.
18. Pozo, M.; Gotor, V. Tetrahedron: Asymmetry 1995, 6, 2797.
19. Pozo, M.; Pulido, R.; Gotor, V. Tetrahedron 1992, 48, 6477.
20. Moris, F.; Gotor, V. J. Org. Chem. 1992, 57, 2490.
21. Pozo, M.; Gotor, V. Tetrahedron 1993, 49, 4321.
22. Kawashima, M.; Hasegawa, T. Biotechnol. Lett. 1993, 15, 465.
23. Yang, Y. L.; Ramaswamy, S. G.; Jakoby, W. B. J. Biol. Chem. 1998, 273, 7814.
24. (a) Matsumoto, K.; Fuwa, S.; Kitajima, H. Tetrahedron Lett. 1995, 36, 6499. (b) Matsumoto, K.; Fuwa, S.;
Shimojo, M.; Kitajima, H.; Hatanaka, M. Synlett 1996, 11, 1085.
25. Matsumoto, K.; Shimojo, M.; Hatanaka, M. Chem. Lett. 1997, 11, 1151.
26. Matsumoto, K.; Fuwa, S.; Shimojo, M.; Kitajima, H. Bull. Chem. Soc. Jpn. 1996, 69, 2977.
27. See: Ref. 4, pp. 194±195.
28. Kanerva, L. T.; Sundholm, O. J. Chem. Soc., Perkin Trans. 1 1993, 1385.
29. Gou, D.-M.; Liu, Y.-C.; Chen, C. S. J. Org. Chem. 1993, 58, 1287.
30. Gentile, A.; Giordano, C.; Fuganti, C.; Ghirotto, L.; Servi, S. J. Org. Chem. 1992, 57, 6635.
31. Preparative Biotransformations; Roberts, S. M.; Casy, G., Eds.; John Wiley & Sons: Chichester, 1994.
32. Bianchi, D.; Cabri, W.; Cesti, P.; Francacalanci, F.; Ricci, M. J. Org. Chem. 1988, 53, 104.
33. Mohr, P.; Rosslein, L.; Tamm, C. Tetrahedron Lett. 1989, 30, 2513.

1288
L. J. Whalen, C. J. Morrow / Tetrahedron: Asymmetry 11 (2000) 1279±1288
34. Ramaswamy, S.; Hui, R. A. F.; Jones, J. B. J. Chem. Soc., Chem. Commun. 1986, 1545.
35. Burgess, K.; Henderson, I. Tetrahedron Lett. 1989, 30, 3633.
36. Guibe-Jampel, E.; Rousseau, G.; Salaun, J. J. Chem. Soc., Chem. Commun. 1987, 1080.
37. Salaun, J.; Karkour, B. Tetrahedron Lett. 1987, 28, 4669.
38. Barnier, J. P.; Blanco, L.; Guibe-Jampel, E.; Rousseau, G. Tetrahedron 1989, 45, 5051.
39. Knezovic, S.; Sunjic, V. Tetrahedron: Asymmetry 1993, 4, 313.
40. Blanco, L.; Rousseau, G.; Barnier, J. P.; Guibe-Jampel, E. Tetrahedron: Asymmetry 1993, 4, 783.
41. Boaz, N. W. J. Org. Chem. 1992, 57, 4289.
42. Brazwell, G. Ph.D. Dissertation, University of New Mexico, 1992.
43. Goering, H.; Eikenberry, J.; Koermer, G.; Lattimer, C. J. Am. Chem. Soc. 1974, 96, 1493.
44. The Merck Index: An Encyclopedia of Chemicals, Drugs, and Biologicals, 11th ed.; Budavari, S., Ed.; Merck & Co.,
Inc.: Rahway, NJ, 1989; p. 820.
45. Ohta, S.; Yamashita, M.; Nagai, N.; Kawasaki, I.; Maeda, K.; Miyano, Y. Heterocycles 1995, 41, 1683.