Crystallization-induced asymmetric transformations. Enantiomerically pure (-)-(R)- and (+)-(S)-2,3-dibromopropan-1-ol and epibromohydrins. A study of dynamic resolution via the formation of diastereoisomeric esters

Article abstract of DOI:10.1002/1522-2675(200211)85:11<3785::AID-HLCA3785>3.0.CO;2-A

(S)-2,3-Dibromopropan-1-ol of high enantiomer excess was obtained by crystallization-induced asymmetric transformations of racemic 2,3-dibromopropan-1-ol esterified with N-([1,1′-biphenyl]-4-ylcarbonyl-L-alanine; in particular, an asymmetric transformatio

Full text of DOI:10.1002/1522-2675(200211)85:11<3785::AID-HLCA3785>3.0.CO;2-A

Helvetica Chimica Acta ± Vol. 85 (2002)
3785
Crystallization-Induced Asymmetric Transformations. Enantiomerically
pure (À)-(R)- and (‡)-(S)-2,3-Dibromopropan-1-ol and Epibromohydrins. A
Study of Dynamic Resolution via the Formation of Diastereoisomeric Esters
a
a
a
a b
1
by Gabriella Brunetto ), Susanna Gori ), Rita Fiaschi ), and Elio Napolitano* ) ) )
a) Dipartimento di Chimica Bioorganica e Biofarmacia, Via Bonanno 33, I-56100 Pisa
b
)
Scuola Normale Superiore, Piazza dei Cavalieri, I-56100 Pisa
Dedicated to Professor Dieter Seebach on the occasion of his 65th birthday
(
S)-2,3-Dibromopropan-1-ol of high enantiomer excess was obtained by crystallization-induced asym-
metric transformations of racemic 2,3-dibromopropan-1-ol esterified with N-([1,1'-biphenyl]-4-ylcarbonyl-l-
alanine; in particular, an asymmetric transformation of the first type (involving bromide exchange to equilibrate
the diastereoisomeric esters) and an asymmetric transformation of the second type (involving a trans-
esterification of diastereoisomeric esters with excess racemic alcohol) were devised.
Introduction. ± The 2,3-dibromopropan-1-ol (1) is an obvious and attractive reagent
for the generation of the dication 1-(hydroxymethyl)ethan-1,2-diylium (2); also, it can
be an effective precursor to epibromohydrin (3), which is itself a reagent for either 2 or
the dication 2-hydroxypropane-1,3-diylium (4). Most common reagents for the
enantiomerically pure C acceptor synthons 2 or 4 are glycidol or glycerol derivatives
3
that have been obtained by enantioselective epoxidation or dihydroxylation of allyl
alcohol derivatives [1], by stereoselective elaboration starting from d-mannitol or
serine [2], or by elaboration of halo-hydroxy-propane derivatives kinetically resolved
by means of chemical or enzymatic methods [3][4]. The enantiomerically enriched
dibromo alcohol 1 has been obtained from the corresponding dibromoamine partially
resolved by optical resolution [5] or by enantioselective addition of bromine to allyl
glycosides [6]. These methods are, however, inefficient and lead to poor enantiomer
excesses. As part of a program of systematic search for novel resolving agents suitable
for use in dynamic resolutions [7 ± 9], we wish to report here a productive approach to
enantiomerically pure (R)- and (S)-1 based on asymmetric transformation of
appropriate ester derivatives.
Results and Discussion. ± Synthesis of Diastereoisomeric Esters 9 and 10. We set out
by tayloring the homochiral acid to be conjugated with racemic 2,3-dibromopropan-1-
ol. We adopted amino acids as the chiral source, because of some attractive features of
this class of compounds as chiral auxiliaries: amino acids are often inexpensive,
available in both chiral forms, and, as bifunctional substances, they allow easy
conjugation with the racemic substrate to be resolved as well as with other groups by
means of which the solubility properties of the diastereoisomeric conjugates can be
1)
Present address: Dipartimento di Chimica Bioorganica e Biofarmacia, via Bonanno 33, I-56127 Pisa
e-mail: eliona@farm.unipi.it; fax: ‡3905043321).
(

3
786
Helvetica Chimica Acta ± Vol. 85 (2002)
tuned and optimized. The control of the solubility of derivatized racemates is
particularly important in consideration of a possible dynamic resolution: the two
conjugates must not form a solid solution, and their solubility must be low under the
conditions under which their equilibration occurs efficiently. Among the N-acyl-
substituted amino acids tested (see 5 and 6, Fig.), we found that N-([1,1'-biphenyl]-4-
ylcarbonyl)-l-alanine provides esters (see 9 and 10) that do not co-crystallize and
exhibit appreciably different solubilities in various solvents such as toluene, MeCN, and
MeOH.
Figure. Resolving agents tested for the separation of 2,3-
dibromopropan-1-ol (Arˆ 3,4,5-triiodophenyl, 3,5-dinitro-
phenyl, [1,1'-biphenyl]-4-yl; X ˆ Cl, Br)
The details of the protocol by which racemic 2,3-dibromopropan-1-ol conjugates
with the resolving agent were obtained is shown in Scheme 1. Rather than preparing the
mixture of diastereoisomers 9 and 10 by condensation of N-([1,1'-biphenyl]-4-
ylcarbonyl)-l-alanine with racemic 2,3-dibromopropan-1-ol (which is commercially
available but somewhat expensive), we found it more practical to get the mixture by
bromination of allyl ester 8. For the first preparation of allyl ester 8, we condensed
alanine allyl ester p-toluenesulfonate (obtained from 7) with [1,1'-biphenyl]-4-carbonyl
13
chloride. Addition of Br2 to 8 gave a 1:1 mixture (HPLC and C-NMR) of
diastereoisomeric esters 9 and 10; as expected, no asymmetric induction was observed
in the bromination step (stereocontrol in the Br addition to allyl alcohol derivatives
2
has proved quite difficult, indeed [6]).
As mentioned above, rather than conducting the separation of the two diaster-
eoisomeric esters 9 and 10 as a traditional fractional crystallization, we were aiming to
devise conditions for dynamic resolution; this would require equilibration of 9 and 10
parallel to their precipitation as solids and would result in the conversion of the more-

Helvetica Chimica Acta ± Vol. 85 (2002)
3787
Scheme 1. Preparation of Diastereoisomeric Esters 9 and 10
a) Allyl alcohol, TsOH, hexane, reflux under continuous removal of H
chloride, Et N, CH Cl . c) Br , CHCl
2
O. b) [1,1'-Biphenyl]-4-carbonyl
3
2
2
2
3
.
soluble of the stereoisomers to the less-soluble one to be isolated as a virtually single
product. We attempted to accomplish this goal following two approaches that differ for
the kind of process involved in the equilibration of the two esters, i.e., via epimerization
by bromide exchange or via epimerization by transesterification in the presence of an
excess of racemic 2,3-dibromopropan-1-ol.
Resolution Involving Epimerization by Bromide Ion Exchange. As the most obvious
approach, we hoped to accomplish the dynamic resolution of 9 and 10 by digestion or
recrystallization of the two solids from a solution containing free bromide ion to induce
epimerization at the secondary bromide centers. So, when a 1:1 mixture of stereo-
isomers 9 and 10 was allowed to recrystallize from 3.5 parts of a 33% solution of
Bu NBr in MeCN, ester 10 was obtained in 25% yield and 75% de; however, analysis of
4
the mother liquors showed that the more-soluble diastereoisomer was prevailing and
little or no epimerization had occurred during the separation of 10. The high salt
concentration, although ineffective in producing a rapid bromide exchange at the
epimeric center, was, however, essential for differentiating the solubility of the two
stereoisomers to the extent that the less-soluble isomer could be obtained in reasonable
yield from a single recrystallization. The reasons for this result are not apparent but
might suggest another important use of concentrated salts solutions known as ionic
2
liquids [10] ). The dynamic resolution of 9 and 10 proved unpractical as a second-order
2)
The specific features of concentrated salt solutions in differentiating isomer solubility is under
investigation.

3
788
Helvetica Chimica Acta ± Vol. 85 (2002)
asymmetric transformation because, under the conditions of an effective bromide
exchange (higher salt concentration and temperature), the solubility of the two esters
was too high. Taking advantage of the substantially different solubilities of the two
diastereoisomers, we were able, however, to devise conditions for a practical
asymmetric transformation of the first type, whose net result is the clean conversion
of the 1:1 diastereoisomeric mixture to an equivalent amount of the single, less-soluble
diastereoisomer (Scheme 2). Thus, the relative amounts of solvent, salt, and esters were
adjusted to obtain the maximum amount of the less-soluble diastereoisomer of
acceptable purity from a single recrystallization: for example, the recrystallization of 9
and 10 from a solution of Bu NBr (5 parts) in MeCN (5 parts) afforded 10 (0.2 part,
4
40%) of 90% de. The mother liquors of this recrystallization (in which the more-
soluble stereoisomer was prevailing) were converted to a solution virtually identical to
that of the first recrystallization by adding 9 and 10 (1:1 mixture; 0.2 part, same as the
amount of 10 obtained as the first crop) and heating for a time long enough to allow the
bromide exchange to bring about complete equilibration. Upon cooling, a second crop
of pure diastereoisomer 10 was obtained as from the first recrystallization and the cycle
could be repeated. The mother liquors could be used to process more samples of the
mixture 9/10 with a constant yield of pure 10 at each step and without apparent loss of
diastereoisomer purity of the resolving agent.
Scheme 2. Production of Enantiomerically pure (2S)-2,3-Dibromopropan-1-ol ((S)-1) and d-Epibromohydrin
(
12) Involving Bromide Exchange
a) Recrystallization from 45% Bu
Aq. NaOH soln. e) Allyl alcohol, hexane, p-toluenesulfonic acid, continuous removal of H
4
NBr solution in MeCN. b) Addition of 9 ‡ 10, 908, 6 h. c) MeOH, HCl. d)
O. f) Br , CDCl
2
2
3
.
Acid-catalyzed methanolysis of the single diastereoisomer 10 afforded enantio-
merically pure 2,3-dibromopropan-1-ol (S)-1 along with methyl ester 11, which were
cleanly separated and quantitatively recovered. Dibromide (S)-1 was converted
quantitatively to epibromohydrine (12), whose enantiimer purity turned out to be
>
95% by gas chromatography (chiral column). Methyl ester 11 was recycled to give
more diastereoisomeric dibromo esters 9 and 10, by a quantitative transesterification

Helvetica Chimica Acta ± Vol. 85 (2002)
3789
with allyl alcohol followed by bromination of the intermediate allyl ester 8. No
detectable racemization of the resolving agent was produced when the above process
was repeated 5 times.
Dynamic Resolution Involving Transesterification. The equilibration of two
diastereoisomers (such as 9 and 10), which is necessary to obtain the selective
precipitation of the less-soluble one in a dynamic resolution, does not necessarily
require that the chiral center of the species to be resolved is directly involved in the
epimerization process. In the specific case of 9 and 10, we thought that equilibration
could also be accomplished by digesting the solid mixture 9/10 in the presence of an
excess of racemic alcohol under conditions in which the small portion of dissolved
esters can undergo transesterification with the free alcohol. Because of the rules
involved in the equilibration of two species forming separate solid phases, one would
expect to see a progressive increase of the less-soluble diastereoisomeric ester in the
solid at the expense of the more-soluble one, paralleled by a deracemization of the free
alcohol in solution. This result, perhaps surprising at first sight, is actually what would
be obtained by the resolution of the alcohol with a substoichiometric amount of
resolving agent; this method is frequently employed in the resolution of organic acids
and bases by formation of insoluble salts with enantiomerically pure bases or acids,
respectively, but it has never been adopted in resolutions with covalently bound chiral
auxiliary reagents. This technique proved quite successful in our case. Indeed, by
digesting a mixture 9/10 in an Et O solution containing an excess (1.5- to 2.5-fold) of
2
racemic 2,3-dibromopropan-1-ol in the presence of an acid catalyst, an increase of the
portion of more insoluble 10 to up to 92% de in the solid phase was observed; the most-
practical conditions to accomplish equilibration turned out to be the use of Et O as the
2
solvent and methanesulfonic acid as the transesterification catalyst. Other catalysts,
such as BF , AlMe (added to the alcohol to form the trialkoxide), or ZnCl , although
3
3
2
effective as well, made the final isolation of 10 unnecessarily more complicated.
Alternatively, we envisaged that the equilibration of solid esters 9 and 10 with a
deficiency of free dibromo alcohol would lead to enantiomerically pure free alcohol in
solution and to a mixture of stereoisomeric esters that, rather than being decomposed
to obtain an enantiomerically enriched alcohol, could be epimerized to a 1:1 mixture
by bromide exchange (as seen in the previous section) and used to treat another batch
of racemic alcohol in a cyclic process made by alternating steps of transesterification
and bromide exchange. One interesting feature of this approach is that the
enantiomerically pure alcohol is produced directly free from any derivatization.
According to our expectation, (S)-1 of 90% ee resulted by equilibrating a solution of
racemic alcohol with a four-fold excess of the diastereoisomer mixture 9/10 in Et O
2
containing methanesulfonic acid.
Scheme 3. Enantioselective Transformation Involving Transesterification
a) Racemic 2,3-dibromopropan-1-ol (0.25 mol), Et
2
O, methanesulfonic acid. b) 45% Bu
4
NBr in MeCN, 908, 6 h.
c) Racemic 2,3-dibromopropan-1-ol (2 mol), Et
2
O, methanesulfonic acid.

3
790
Helvetica Chimica Acta ± Vol. 85 (2002)
Conclusions. ± Two effective preparations of the hitherto elusive enantiomerically
pure 2,3-dibromopropan-1-ol were devised via crystallinity-induced asymmetric trans-
formations of diastereoisomeric esters. A suitable chiral auxiliary reagent, N-([1,1'-
biphenyl]-4-ylcarbonyl)-l-alanine, was found among five alanine derivatives designed
for this purpose. An asymmetric transformation of the first type involving bromide
exchange for the equilibration of the two diastereoisomeric conjugates was possible
because of their unexpectedly high difference in solubility in the concentrated solution
of Bu NBr originally meant to induce epimerization. In a successful asymmetric
4
transformation of the second type, equilibration of 9 and 10 was obtained by
transesterification with an excess of racemic alcohol; conceptually related to an optical
resolution with a substoichiometric amount of resolving agent (which allows only the
pure enantiomer giving rise to the less-soluble derivative to separate as a conjugate),
this technique of asymmetric transformation (or optical resolution) has never been
adopted before in cases in which the resolving agent and the racemic compound are
covalently bound. With the above results, we have demonstrated that, by a systematic
search of the resolving agent, enantiomer separation via diastereoisomer formation can
be efficiently applied to a preselected target.
This work was financially supported by Boehringer Ingelheim Pharmaceuticals Inc. (Ridgefield, CT).
Experimental Part
1. General. M.p.: uncorrected. B.p.: oven temp. of the bulb-to-bulb distillation apparatus (Buki KR200).
Gas chromatography: enantiomer-excess evaluation of 12 with an Astec Chiraldex GTA column (trifluor-
oacetylated cyclodextrins, 20 m, i.d. 0.25 mm). HPLC: Nucleosil-100 silica-gel column, 5 mm, 25 cm, i.d.
1
13
4
.6 mm). NMR Spectra: H at 200 MHz and C at 50 MHz; CHCl
3
solns., unless otherwise stated; d in ppm, J in
Hz.
2
. Allyl N-([1,1'-Biphenyl]-4-ylcarbonyl)-l-alaninate (8). A mixture of l-alanine (30 g, 0.34 mol), allyl
alcohol (230 ml, 3.4 mol), and TsOH ¥ H O (77.5 g, 0.4 mol) was heated to reflux in hexane (300 ml) under
continuous removal of H O. The resulting homogeneous mixture was evaporated and the residue triturated in
Et O: alanine allyl ester p-toluenesulfonate salt (96 g, 94%). Crystalline solid. M.p. 898 ([11]: 84 ± 858).
To a mixture of the above ester (41.5 g, 0.14 mol), and [1,1'biphenyl]-4-carbonyl chloride (30 g, 0.14 mol) in
Cl (300 ml), while cooled in an ice bath Et N (70 ml, 51 g, 0.505 mol) was added dropwise, and the reaction
was allowed to proceed overnight. The mixture was washed with sat. NaHCO soln., dried, and evaporated to
O. The anal. sample was
2
2
2
CH
2
2
3
3
afford a residue from which 8 (32.1 g, 75%) was obtained by trituration in Et
2
1
recrystallized from hexane. M.p. 118 ± 1208. H-NMR: 1.53 (d, J ˆ 7.3, 3 H); 4.64 (d, J ˆ 5.4, 2 H); 4.80 (q, J ˆ 7.3,
1
H); 5.21 ± 5.36 (m, 2 H); 5.80 ± 5.99 (m, 1 H); 6.96 (d, J ˆ 7.3, 1 H); 7.33 ± 7.45 (m, 3 H); 7.55 ± 7.63 (m, 4 H);
13
7
.86 (d, J ˆ 8.3, 2 H). C-NMR: 19.2; 46.4; 49.2; 66.6; 119.4; 127.8; 128.2; 128.6; 129.5; 132.1; 133.2; 140.6; 173.6.
Anal. calc. for C19 : C 73.77, H 6.19, N 4.53; found: C 73.2, H 6.5, N 4.5.
. (2R)- and (2S)-2,3-Dibromopropyl N-([1,1'-Biphenyl]-4-ylcarbonyl)-l-alaninate (9 and 10, resp.). To a
soln. of allyl ester 8 (27.5 g, 0.09 mol) in CHCl (200 ml), magnetically stirred and cooled in an ice bath, Br
5 ml) was added dropwise, and the mixture was further stirred for 2 h at r.t. The mixture was washed with 10%
H
19NO
3
3
3
2
(
sodium thiosulfate soln., dried (MgSO
4
), and evaporated to afford a residue from which 9/10 (38.5 g, 94%) was
1
obtained after trituration in cold Et
2
O. M.p. 138 ± 1408. H-NMR: 1.58 (d, J ˆ 7.3, 3 H); 3.71 ± 3.85 (m, 2 H); 4.36
(
(
m, 1 H); 4.56 ± 4.66 (m, 2 H); 4.88 (m, 1 H); 6.74 (d, J ˆ 7.3, 1 H); 7.37 ± 7.49 (m, 3 H); 7.58 ± 7.68 (m, 4 H); 7.87
13
d, J ˆ 8.3, 2 H). C-NMR: 19.2; 32.4; 32.6; 47.1; 49.2; 66.3; 66.7; 127.9; 128.2; 128.7; 129.6; 133.0; 140.6; 145.3;
67.4; 173.2. Anal. calc. for C19 NO : C 48.64, H 4.18, Br 34.06, N 2.99; found: C 48.4, H 4.3, Br 34.0, N 2.9.
. (2S)-2,3-Dibromopropyl N-([1,1'-Biphenyl]-4-ylcarbonyl)-l-alaninate (10). 4.1. By Bromide Exchange.
A mixture 9/10 (8 g) was recrystallized from a hot soln. of Bu NBr (20 g) in MeCN (20 ml). The precipitate was
1
H
19Br
2
3
4
4
collected by filtration, washed with MeCN (2 Â 5 ml), and dried: 10 (1.5 g, 90% de). A sample was recrystallized
1
from MeOH. M.p. 1608. H-NMR: 1.58 (d, J ˆ 7.3, 3 H); 3.71 ± 3.85 (m, 2 H); 4.36 (m, 1 H); 4.56 ± 4.66 (m, 2 H);
1
3
4
.88 (m, 1 H); 6.74 (d, J ˆ 7.3, 1 H); 7.37 ± 7.49 (m, 3 H); 7.58 ± 7.68 (m, 4 H); 7.87 (d, J ˆ 8.3, 2 H). C-NMR:

Helvetica Chimica Acta ± Vol. 85 (2002)
3791
1
1
9.2; 32.4; 47.1; 49.2; 66.3; 127.9; 128.2; 128.7; 129.6; 133.0; 140.6; 145.3; 167.4; 173.2. Recrystallization afforded
0 of 100% de.
A mixture 9/10 (1.5 g) was added to the filtrate which was heated under reflux for 6 h, while allowing ca.
0 ml of solvent to evaporate. Upon cooling, a second crop of 10 separated; the filtrate was processed as above
1
and the cycle repeated up to 5 times.
.2. By Transesterification. A suspension of 9/10 (3.0 g, 6.4 mmol) in a soln. containing racemic 2,3-
dibromopropan-1-ol (2.8 g, 12.8 mmol) and methanesulfonic acid (0.50 ml, 7.7 mmol) in Et O (15 ml) was
4
2
stirred at 408 in a screw-cap glass tube for 15 d. The solid phase was collected by filtration (2.7 g) and, upon
analysis, was discovered to be 92% pure (HPLC) ester 10.
5
. (2R)- or (2S)-2,3-Dibromopropan-1-ol ((R)-or (S)-1, resp.) 5.1. By Bromide Exchange. Pure
dibromopropyl ester 10 (obtained as described in Exper. 4.1; 1.5 g, 3.2 mmol) was added in a soln. obtained
by dissolving AcCl (0.75 ml, 81 mmol) in MeOH (15 ml) and the mixture was heated to reflux until 10 was
totally consumed (ca. 3 h, TLC monitoring; silica-gel; hexane/AcOEt 7:3). The mixture was cooled in a
refrigerator to produce the precipitation of methyl N-([1,1'-bipheny]-4-ylcarbonyl)-l-alaninate (11), which was
filtered off; H
2
O (15 ml) was added to the filtrate, an additional crop of ester 11 was collected. Total yield: 0.88 g
1
(
98%): M.p. 1608. H-NMR: 1.52 (d, J ˆ 7.3, 3 H); 3.78 (s, 3 H); 4.82 (q, 1 H); 6.87 (d, J ˆ 7.3, 1 H); 7.36 ± 7.47 (m,
1
3
3
1
H); 7.57 ± 7.65 (m, 4 H); 7.87 (d, J ˆ 8.3, 2 H). C-NMR: 19.3; 49.2; 53.2; 127.9; 128.2; 128.7; 129.6; 133.2;
40.6; 145.2; 167.2; 173.4. Anal. calc. for C17 : C 72.07, H 6.05, N 4.94; found: C 71.6, H 6.3, N 4.8.
The clear filtrate was saturated with NaCl and extracted with Et O; the extract was dried (CaCl ) and
H
17NO
3
2
2
evaporated to afford crude (S)-1 as a light tan oil (0.68 g, 95%). The anal. sample was purified by bulb-to-bulb
1
distillation at 1008/10 Torr. [a]
D
ˆ À7.6 (c ˆ 0.316, MeOH). H-NMR: 2.37 (s, 1 H); 3.77 (d, J ˆ 7.8, 2 H); 3.98 (d,
1
3
J ˆ 4.4, 2 H); 4.22 ± 4.33 (m, 1 H). C-NMR: 32.3; 54.1; 64.8.
.2. Transesterification. A suspension of 9/10 (5.8 g, 12.4 mmol) in a soln. of racemic 2,3-dibromopropan-1-
ol (0.7 g, 3.2 mmol) and methanesulfonic acid (0.5 ml, 7.7 mmol) in Et O (15 ml) was stirred in a screw-cap tube
5
2
at 408 for 15 d. The soln. was diluted with hexane, filtered, and evaporated: crude (R)-1 (0.7 g, 100%). Its
enantiomer excess (90%) was established by GC analysis of the corresponding epibromohydrin (see below).
(
R)-Epibromohydrine (ˆ(2R)-2-(Bromomethyl)oxirane (12). A soln. of (S)-1 (0.4 g, 1.82 mmol) in Et
10 ml) was stirred in the presence of 30% aq. NaOH soln. for 45 min. The Et O phase was dried (MgSO ) and
evaporated to give a residue from which pure 12 (0.21 g, 85%) was obtained by bulb-to-bulb distillation. [a]
2
O
(
2
4
D
ˆ
1
13
À15.0 (c ˆ 0.031, MeOH) [2b]. H-NMR: 2.66 (m, 1 H); 2.93 (m, 1 H); 3.13 ± 3.66 (m, 3 H). C-NMR: 33.1;
9.2; 51.9.
4
REFERENCES
[1] R. M. Hanson, Chem. Rev. 1991, 91, 437.
[
2] a) M. C. Pirrung, S. E. Dunlap, U. P. Trinks, Helv. Chim. Acta 1989, 72, 1301; b) Y. Kawakami, T. Asai, K.
Umeyama, Y. Yamashita, J. Org. Chem. 1982, 47, 3581; c) M. Bols, −Carbohydrate Building Blocks×, Wiley,
New York, 1996; J. J. Baldwin, A. W. Raab, K. Mensler, B. H. Arison, D. E. McLure, J. Org. Chem. 1978, 43,
4876.
[
3] M. K. Ellis, B. T. Golding, A. B. Maude, W. P. Watson, J. Chem. Soc., Perkin Trans. 1 1991, 747; M. E.
Furrow, S. E. Schaus, E. N. Jacobsen,J. Org. Chem. 1998, 63, 6776; S. E. Schaus, J. Branalt, E. N. Jacobsen,J.
Org. Chem. 1998, 63, 4876; D. A. Annis, E. N. Jacobsen, J. Am. Chem. Soc. 1999, 121, 4147.
4] C. A. G. M. Weijers, Tetrahedron: Asymmetry 1997, 8, 639; T. Suzuki, N. Kasai, N. Minamiura, Tetrahedron:
Asymmetry 1994, 5, 239; N. Kasai, K. Sakaguki, Tetrahedron Lett. 1992, 33, 1211; B. Cambou, A. M.
Klibanov, J. Am. Chem. Soc. 1984, 106, 2687; C. S. Chen, Y. C. Liu, M. Marsella, J. Chem. Soc., Perkin
Trans. 1 1990, 2559.
[
[
[
[
[
[
5] A. C. Huitrich, W. P. Gordon, S. D. Nelson, J. Chem. Eng. Data 1982, 27, 474.
6] G. Bellucci, C. Chiappe, F. D×Andrea, Tetrahedron Asymmetry 1995, 6, 221.
7] J. Jacques, A. Collet, S. H. Wilen, −Enantiomers, Racemates, and Resolutions×, Wiley, New York, 1981.
8] E. L. Eliel, S. H. Wilen, L. N. Mander, −Stereochemistry of Organic Compounds×, Wiley, New York, 1994.
9] E. J. Ebbers, G. J. A. Ariaans, J. P. M. Houbiers, A. Bruggink, B. Zwanenburg, Tetrahedron 1997, 53, 9417; S.
Caddick, K. Jenkins, Chem. Soc. Rev. 1996, 447.
[
[
10] P. Wassersheid, W. Keim, Angew. Chem. 2000, 112, 3926; T. Welton, Chem. Rev. 1999, 99, 2071.
11] H. Waldmann, H. Kunz, Liebigs Ann. Chem. 1983, 1912.
Received July 26, 2002