A scalable asymmetric synthesis of (R)-2-amino-1-(3-pyridinyl)ethanol dihydrochloride via an oxazaborolidine catalyzed borane reduction

Article abstract of DOI:10.1021/op0340208

This report describes a scalable process for the asymmetric synthesis of (R)-2-amino-1-(3-pyridinyl)ethanol dihydrochloride. The stereochemistry of the product is set via a reduction of 3-chloroacetyl pyridine with 2 equiv of borane-dimethyl sulfide and a catalytic amount of an in situ generated oxazaborolidine. The enantiomeric excess (ee) of the reductive step depends on the addition rate of the substrate and the temperature. The authors hypothesize that the low ee observed during a fast addition of the substrate or at low temperatures is due to the slow regeneration of the active catalyst from the catalyst-product complex.

Full text of DOI:10.1021/op0340208

Organic Process Research & Development 2003, 7, 285−288
A Scalable Asymmetric Synthesis of (R)-2-Amino-1-(3-pyridinyl)ethanol
Dihydrochloride via an Oxazaborolidine Catalyzed Borane Reduction
Jason Duquette,* Mingbao Zhang, Lei Zhu, and Raymond S. Reeves
Department of Chemistry Research, Bayer Research Center, Bayer Pharmaceuticals Corporation, 400 Morgan Lane,
West HaVen, Connecticut 06516, U.S.A.
Scheme 1. Enantioselective Synthesis of
(R)-2-Amino-1-(3-pyridinyl)ethanol
Abstract:
This report describes a scalable process for the asymmetric
synthesis of (R)-2-amino-1-(3-pyridinyl)ethanol dihydrochloride.
The stereochemistry of the product is set via a reduction of
3-chloroacetyl pyridine with 2 equiv of borane-dimethyl sulfide
and a catalytic amount of an in situ generated oxazaborolidine.
The enantiomeric excess (ee) of the reductive step depends on
the addition rate of the substrate and the temperature. The
authors hypothesize that the low ee observed during a fast
addition of the substrate or at low temperatures is due to the
slow regeneration of the active catalyst from the catalyst-
product complex.
Introduction
Recently we had a need to produce kilogram quantities
of amino alcohol 4 (Scheme 1). Previously, our group
obtained this intermediate enantiomerically pure via chiral
chromatography of racemic 31 and subsequent Gabriel
synthesis.2,3 To obviate the need for chromatography and to
facilitate the rapid chemical development of a preclinical
candidate, we investigated the asymmetric synthesis of 4 via
an oxazaborolidine catalyzed borane reduction. Oxazaboro-
lidines have demonstrated tremendous scope and efficiency
in asymmetric catalytic borane reductions of unsymmetrical
ketones.4 For example, Quallich5 and Masui6 have reported
their use in catalytically reducing ketones containing het-
erocyclic functionality. Specifically, Quallich obtained 96%
ee and Masui obtained 99% ee with 3-acetylpyridine, in
excellent yields.7 Quallich,5 Corey,8 and Hett9 also reported
the reduction of the R-haloacetophenone moiety, with Hett
further demonstrating the scalability of this reaction.10 Chiral
stoichiometric reductions of R-haloacetylpyridines have been
reported,11,12 as has a poor-yielding catalytic reduction.13
However, this report describes the first example of an
efficient catalytic asymmetric reduction on 3-chloroacetyl
pyridine and the subsequent large-scale synthesis of its
corresponding amino alcohol.
Results and Discussion
Earlier work demonstrated that high yields are achievable
for the chlorination of the hydrochloride salt of 3-acetylpyr-
idine (Scheme 1).12 In the referenced procedure, 3-acetylpyr-
idine was precipitated as the HCl salt by treatment with 1
M HCl in ether. The product was filtered, added to 1 M
HCl in acetic acid, treated with NCS, and filtered again. To
avoid the use of a highly flammable solvent on a large scale
and to eliminate a time-consuming isolation step, the in situ
generation of the 3-acetylpyridine HCl salt was investigated.
* To whom correspondence should be addressed. E-mail: jason.duquette.b@
bayer.com.
(1) Chromatography was performed on borane-free chlorohydrin.
(2) Bayer Co. U.S. Patent 6,051,586, April 18, 2000.
(3) The enantiomerically pure chlorohydrin is also obtainable via a fermentation
process. Kaneka Co. WO Patent 00/48997, August 24, 2000.
(4) Singh, V. K. Synthesis 1992, 605-617.
(5) Quallich, G. J.; Woodall, T. M. Tetrahedron Lett. 1993, 34, 785-788.
(6) Masui, M.; Shioiri, T. Synlett 1997, 273-274.
(7) Quallich used 5 mol % and Masui used 10 mol % catalyst.
(8) Corey, E. J.; Shibata, S. J.; Bakshi, R. K. J. Org. Chem. 1988, 53, 2861-
2863.
(11) Naylor, E. M.; Colandrea, V. J.; Candelore, M. R.; Cascieri, M. A.; Colwell,
L. F.; Deng, L.; Feeney, W. P.; Forrest, M. J.; Hom, G. J.; MacIntyre, D.
E.; Strader, C. D.; Tota, L.; Wang, P.; Wyvratt, M. J.; Fisher, M. H.; Weber,
A. E. Bioorg. Med. Chem. Lett. 1998, 8, 3087-3092.
(12) Merck & Co. U.S. Patent 5,561,142, October 1, 1996.
(13) (a) Hu, B.; Ellingboe, J.; Gunawan, I.; Han, S.; Largis, E.; Li, Z.; Malamas,
M.; Mulvey, R.; Oliphant, A.; Sum, F.; Tillett, J.; Wong, V. Bioorg. Med.
Chem. Lett. 2001, 11, 757-760. (b) American Home Products Co. WO
Patent 02/0623, January 24, 2002. (c) The optically active bromohydrin
obtained in the previous references was converted into the epoxide and
opened with ammonia/methanol to yield the optically active amino alcohol.
No ee was reported.
(9) Hett, R.; Senanayake, C. H.; Wald, S. A. Tetrahedron Lett. 1998, 39, 1705-
1708.
(10) Hett, R.; Fang, Q. K.; Gao, Y.; Wald, S. A.; Senanayake, C. H. Org. Process
Res. DeV. 1998, 2, 96-99.
10.1021/op0340208 CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/15/2003
Vol. 7, No. 3, 2003 / Organic Process Research & Development
•
285

Scheme 2. Putative Catalytic Cycle
In the current method, an ∼2 M HCl solution in acetic acid
was easily prepared with gaseous HCl and glacial acetic
acid.14 3-Acetylpyridine was then added to the solution,
followed by the addition of NCS, and the product was
isolated by filtration with yields and purity comparable to
those of the previous method.15
Chloroketone 2 was then converted to the free base in
preparation for the reduction. Initially, the HCl salt was
dissolved in water, slowly treated with solid sodium bicar-
bonate, and extracted with dichloromethane. The organic
extracts were then dried with sodium sulfate, filtered,
concentrated in Vacuo, and redissolved in THF. Two
concerns about the scalability of this step prompted us to
explore alternative methods. First, the procedure seemed
unnecessarily lengthy for such a simple step. Second, earlier
work in our lab showed that the free base of the chloroketone
had poor solution stability. This instability might lead to
decomposition during the removal of dichloromethane on a
large scale. Two other methods more amenable to large-scale
synthesis were developed. The free base of the chloroketone
is insoluble in water, so it could be isolated by a simple
filtration and a solution prepared as needed. The other
approach was to prepare a slurry of the HCl salt in THF,
basify the salt with triethylamine, and remove the resulting
triethylamine hydrochloride solid by filtration. The latter
approach was chosen for scale-up because it permitted the
direct use of the filtrate for the subsequent step. Despite the
heterogeneity of this procedure, only small amounts of the
product were found in the solid triethylamine salt. Addition-
ally, only a trace of triethylamine was found in the filtrate.16
This new procedure eliminated the need to perform a slow
solid addition, an extraction, and a concentration and thus
improved the efficiency of the process.
The oxazaborolidine catalyzed borane reduction was
performed next. The active catalytic species for the reduction
was prepared in situ by mixing (R)-R,R-diphenyl-2-prolinol
and trimethylborate in THF for 1 h. By using trimethyl
borate, the active catalyst has a methoxy group attached to
boron (Scheme 2). Masui6 proposed that, by modifying the
Lewis acidity of the boron, the binding rate of the catalyst
to the ketone should increase and so should the relative rate
of the catalytic reduction versus the background reduction.
Thus, this procedure was chosen because it provided excel-
lent results with 3-acetylpyridine and precedent for an
increase in ee versus catalysts with other17 boron substituents.
After the generation of the active catalyst, 2 equiv of borane-
dimethyl sulfide in THF were then charged to the reaction.
Excess borane was necessary to facilitate the reaction, as 1
Chart 1. Percent of Enantiomeric Excess versus Addition
Time of the Free Base of 2a-c
a Scale shown is for the HCl salt. bThe temperature was maintained between
c
23 and 25 °C during the addition. The catalyst appears saturated at less than
170 min. The authors believe that the lower ee’s for the 10 g runs at 170 min
(89% ee) and 226 min (89% ee) versus the 1.5 kg run (96% ee) are due to the
presence of adventitious water (ref 22) that was introduced into the chloroketone
solution during the open-to-air filtration. A closed filter, inerted with dry nitrogen,
was used for the 1.5 kg run.
Table 1. Effect of Temperature on Percent of Enantiomeric
Excessa
addition time
(min:s)
% ee
tempb (°C)
48:55
49:53
47:47
7.3
53.9
88.9
2-3
24-25
43-44
a Runs were performed on 10 g of 2. bInternal temperature.
equiv coordinates to the pyridine nitrogen.5,18 The chloroke-
tone solution was then added slowly to the reaction solution
to effect the asymmetric reduction.
Small-scale experiments demonstrated that the addition
time (Chart 1) and reaction temperature (Table 1) were
critical in determining the enantiomeric excess. To the best
of our knowledge, it has never been reported in the literature
that the enantiomeric excess can depend on the substrate
addition time for an oxazaborolidine catalyzed reduction of
a ketone.19 Most efforts at improving the ee have focused
on modifying the ligand structure and the boron substituent.9
(14) In the preliminary experiments, the flask containing acetic acid and the HCl
cylinder were tared to determine that the desired amount was transferred
into the solution. On the kilogram scale, only the cylinder tare was used.
No other method was used to determine the precise molarity of the solution.
(15) The impurities in the reaction mixture for both methods are 3-acetylpyridine,
the bis-chlorinated byproduct, and succinimide. However, the bis-chlorinated
byproduct and succinimide are more soluble in the reaction mixture and
are not detected in the isolated solid. Based on the relative ratio of the
methyl (2.69 ppm) and methylene (5.30 ppm) protons in the 1H NMR
(DMSO-d6), there was 3% 3-acetylpyridine hydrochloride in the solid.
(16) By 1H NMR, less than 2 mol % of the product was found in the triethylamine
hydrochloride solid and less than 2 mol % of triethylamine was found in
the filtrate.
(18) After quenching with methanol, a broad peak is observed in the 1H NMR
(17) Quallich, G. J.; Woodall, T. M. Synlett 1993, 929-930.
at 2.5 ppm, which is consistent with a borane-pyridine complex.
286
•
Vol. 7, No. 3, 2003 / Organic Process Research & Development

Presumably, an optimized ligand should offer greater enan-
tioselectivity during the reductive step in the catalytic cycle.
In our case, since the ee was high during a slow addition,
the enantioselectivity of the catalytic species must be high,
thus suggesting that the low ee at high addition rates is due
to a low catalyst turnover frequency. Previous experiments
in our lab showed that the background reduction of the
chloroketone was fast. If the putative catalytic cycle is correct
(Scheme 2),20 then the binding of the substrate and the
hydride transfer must be even faster. This conclusion is
reached because the ee was high with a slow addition in the
presence of a large excess of borane. Other data suggest that
borane binding is also fast. For example, the reduction of
acetophenone reported by Masui, with 0.05 equiv of the
enantiomer of 6, yielded a high ee with an addition time of
10 min at 25-30 °C. This duration is much less than the
addition time needed to obtain a high ee for 2.21-23 Thus,
we hypothesize that the low ee at high addition rates was
due to the slow release of the product from the catalyst
relative to the background reduction.24 Temperature also had
a dramatic effect on the ee, with higher temperatures yielding
a higher ee for a constant addition time. We reason that the
rate of product release from the catalyst depends more on
temperature than does the background reduction, resulting
in a higher relative rate at higher temperatures and thus a
higher ee.
reaction was stirred for 18 h. No effort was made to reduce
the number of equivalents of ammonia, but the excess was
used to avoid significant formation of the dialkylated
byproduct.25 The solvent was then swapped with n-butanol
via a combination of ambient and azeotropic vacuum
distillation. This also removed most of the ammonia.26
Concentrated aqueous hydrochloric acid was then added to
form the dihydrochloride salt.27 The water was then removed
via azeotropic vacuum distillation,28 and the resulting white
solid was filtered to give crude amino alcohol 4 in 66% yield
over three steps with 96% ee and 96% purity by HPLC
area.29 The product can be used without further purification.
Alternatively, enrichment by a hot reslurry in 190 proof
ethanol delivers a 79% recovery with 99.5% ee and 99.8%
purity by area.
Summary
An efficient and scalable asymmetric synthesis of enan-
tiomerically enriched 2-amino-1-(3-pyridinyl)ethanol dihy-
drochloride was developed using commercially available
starting materials. The desired compound is produced in 3
steps, in high purity and with an overall yield of 66% and
an ee of 96%. The catalytic cycle appears to be limited by
the regeneration of the active catalyst from the catalyst-
product complex.
After the completion of the chloroketone addition, the
reaction mixture was slowly quenched with methanol18 to
liberate 3 and then was concentrated in Vacuo. Methanol and
30 equiv of ammonium hydroxide were added, and the
Experimental Section
General. 1H NMR and 13C NMR spectra were measured
with a Varian NMR spectrometer with residual protonated
solvent (1H NMR, DMSO-d6 δ 2.50; CDCl3 δ 7.27; 13C
NMR, DMSO-d6 δ 39.51) as the reference standard. Chemi-
cal shifts are reported in part per million. Elemental analysis
was performed by Robertson Microlit Laboratories in
Madison, NJ. Unless otherwise noted, all reagents were
obtained from Acros, Aldrich, or Lancaster and were used
without further purification. All solvents were obtained from
EM Science or Pharmco. The pump system used to examine
the effect of addition rate was the MasterFlex Digital Console
L/S Pump Drive (Cole-Parmer #7524-50) with a PTFE tubing
pump head (Cole-Parmer #77390-00) and PTFE tubing
assembly (2 mm ID, 4 mm OD, Cole-Parmer #77390-50).
For the kilogram scale run, a Fluid Metering Inc. pump
(19) Hett showed (ref 9) that the same ee was obtained in the reduction of
2-bromo-4′-(benzyloxy)-3′-nitroacetophenone when the substrate addition
time was 2 h or 30 min.
(20) Corey, E. J.; Bakshi; R. K.; Shibata, S. J. Am. Chem. Soc. 1987, 109, 5551-
5553.
(21) High ees have been obtained with oxazaborolidine catalyzed borane
reductions during which borane was added into a solution of the catalyst
and ketone (ref 22). These results also suggest that borane binding is fast.
(22) Jones, T. K.; Mohan, J. J.; Xavier, L. C.; Blacklock, T. J.; Mathre, D. J.;
Sohar, P.; Jones, E. T. T.; Reamer, R. A.; Roberts, F. E.; Grabowski, E. J.
J. J. Org. Chem. 1991, 56, 763-769.
(23) For the addition of ketone, excess borane is always in solution with the
catalyst and intermediate 7 can form upon regeneration of 6. In this addition
sequence, the borane binding rate required for a high ee is less than the
rate required for a high ee under the addition of borane.
(24) This low turnover frequency might be due to the stabilization of intermediate
9 via the electron withdrawing effect of the chloro substituent and the
consequent delocalization of the negative charge on boron. For example,
without the chloro substituent, only a 1 h addition time at 0-5 °C was
needed to reduce 3-acetylpyridine in 99% ee with 10 mol % of the
enantiomer of catalyst 6 and with borane-dimethyl sulfide as the reducing
agent (ref 6). Furthermore, it has been reported that the simultaneous addition
of ketone and reducing reagent was necessary to ensure high enantioselec-
tivity in the reduction of 2-bromo-4′-(benzyloxy)-3′-nitroacetophenone (ref
10), R-chloroacetophenone (ref 8, 31), and 2-bromo-cyclohex-2-eneone (ref
32). The only explanation provided on the need for simultaneous addition
is that it "suppress[es] uncatalyzed reduction pathways" (ref 31). However,
there are also examples on the reduction of 2-bromo-4′-(benzyloxy)-3′-
nitroacetophenone (ref 9) and R-chloroacetophenone (ref 17, 32) for which
simultaneous addition was not necessary, using a different catalyst, a
different borane source, or both a different catalyst and a different borane
source. This may be due to the role the catalyst and borane source play in
the stabilization or destabilization of intermediate 9. Another possible
counterexample is that trihalomethyl ketones were reduced with excellent
enantioselectivity (ref 33). However, no description of the addition mode,
addition time, or time for complete conversion was provided. Also, the
trihalomethyl group inverts the stereo configuration of the catalyst complex
relative to the corresponding monohalomethyl ketone, which may in itself
have a consequence on the rate of catalyst regeneration.
(25) A minor amount of the ion (MH+ ) 260.1) believed to correspond to the
dialkylated product was observed in the LCMS (APCI) of one of the
impurities detected in the isolated solid. The impurity (tr ) 8.57 min) was
present in 2 area % based on the achiral method outlined in ref 29. The
major ion detected was 224.1. Also a minor amount of 265.1 was detected.
1
(26) Because of its insolubility, ammonium chloride has no H NMR spectrum
in DMSO-d6, but when a sample of pure 4 is spiked with ammonium
chloride, a triplet is observed at 7.4 ppm. This triplet was observed by H
NMR in the final product, but its intensity was reduced by the reslurry. See
the Experimental Section for more information.
1
(27) The monohydrochloride salt is partially soluble in n-butanol and also can
be isolated by filtration, albeit in lower yields.
(28) It was later observed that water appears to aid the removal of some of the
organic impurities. Future work might include eliminating the final
azeotrope.
(29) Chiral conditions: Crownpak CR(+) 4.0 mm × 150 mm; A ) 0.5% HClO4
in H2O, B ) MeOH; 0.5 mL/min; isocratic run with 1% B for 10 min;
room temp; UV 265 nm; tr ) 5.2 min. Achiral conditions: Metachem Diol.
5.0 mm × 60 mm; A ) 0.1% TFA in hexane, B ) 0.1% TFA in 1:1 MeOH/
EtOH; 3.0 mL/min; ramp from 1% B to 30% B over 9.5 min; room temp;
UV 254 nm; tr ) 7.7 min.
Vol. 7, No. 3, 2003 / Organic Process Research & Development
•
287

(QSY-1) was used. The kilogram scale synthesis was
performed in a Bu¨chi 100-L reactor train.
(R)-2-Amino-1-(3-pyridinyl)ethanol Dihydrochloride
(4). To the solution previously mentioned was added
methanol (13 L) and 30% ammonium hydroxide (30 L, 220
mol). The mixture was stirred at 25 °C for 3 days, but a
sample after 18 h indicated the reaction was complete. Next,
methanol was distilled under ambient pressure. About one-
half of the water and ammonia were then removed via
vacuum distillation. The remaining water and most of the
ammonia were removed via azeotropic vacuum distillation
with n-butanol (53 L). HCl (37% aq, 1.5 L) was then added
at 10 °C over 10 min. The mixture was again distilled under
vacuum until the distillate contained a single phase, and the
final volume was ∼18 L. The white solid was filtered, rinsed
with n-butanol (4 L), and dried in the vacuum oven at 35
°C to yield 1.30 kg (79%).
2-Chloro-1-(3-pyridinyl)ethanone Hydrochloride (2).
Glacial acetic acid (24 L) was charged to the reactor, and
gaseous HCl (1.7 kg, 47 mol) was introduced above the
surface of the liquid at 17 °C over 1 h. The temperature was
raised to 20 °C, and 3-acetylpyridine (2.3 L, 21 mol) was
added dropwise in 40 min, resulting in the formation of a
vapor cloud and a clear solution. The sides of the reactor
were rinsed with more glacial acetic acid (2 L). N-
chlorosuccinimide (3.0 kg, 22.5 mol) was then dissolved in
the solution at 22 °C. After 1 h, the product began to
crystallize from the reaction mixture, resulting in a rise in
temperature. The mixture was stirred overnight at 20 °C,
cooled to 15 °C the next morning, and filtered. The cake
was washed with acetic acid (2 L), washed with ethyl acetate
(4 L), and dried under vacuum at 25 °C to yield 3.3 kg (83%)
The solid previously mentioned (1.30 kg) was charged
to a flask with 5.2 L of ethanol/water (95:5), heated to reflux,
and allowed to cool to room temperature overnight. The solid
was filtered, rinsed with ethanol (500 mL), and dried in a
vacuum oven at 35 °C to yield 1.03 kg (79%) with 99.8%
purity by area and 99.5% ee. The solid was determined to
be partially crystalline by the observation of some birefrin-
gence under an optical microscope with a differential
1
of white powder. Compound is an irritant. H NMR (400
MHz, DMSO-d6) δ 9.26 (s, 1H), 8.95 (d, J ) 5.2 Hz, 1H),
8.59 (d, J ) 8.3 Hz, 1H), 7.86 (dd, J ) 5.2, 8.3 Hz, 1H),
5.30 (s, 2H). 13C NMR (100 MHz, DMSO-d6) δ 188.6, 147.8,
144.2, 140.4, 131.0, 125.6, 48.1.
Borane-(R)-2-chloro-1-(3-pyridinyl)ethanol (3). 2 (1.50
kg, 7.81 mol) and THF (11.3 L) were charged to the reactor.
A solution of triethylamine (1.09 L, 7.81 mol) in THF (1.25
L) was then added dropwise over 10 min. The mixture was
stirred for 1 h at 17 °C and filtered. The reactor was rinsed
with THF (2.5 L), and the rinse was filtered over the solid.
The filtrate (14 L) was kept under inert atmosphere at room
temperature during the addition, and no precipitation was
observed. A sample of the filtrate was concentrated and
1
interface contrast filter. H NMR (400 MHz, DMSO-d6) δ
8.87 (s, 1H), 8.84 (d, J ) 5.7 Hz, 1H), 8.54 (d, J ) 8.2 Hz,
1H), 8.28 (b, 2H), 8.01 (dd, J ) 5.7, 8.2 Hz, 1H), 5.15 (dd,
J ) 3.9, 7.8 Hz, 2H), 3.21 (m, 1H), 3.05 (m, 1H). 13C NMR
(100 MHz, DMSO-d6) δ 142.5, 140.5, 140.4, 139.5, 126.2,
66.0, 44.6. HRMS +esi (m/z): [M + H]+ calcd for C7H12-
Cl2N2O, 139.086589; found, 139.08697. Anal. Calcd for
C7H12Cl2N2O‚0.15H4ClN: C, 38.37; H, 5.80; Cl, 34.79; N,
13.74. Found: C, 38.35; H, 5.89; Cl, 34.66; N, 13.77. Based
on the elemental analysis, the estimated ammonium chloride
content is 4 wt %.
An analytical sample was obtained by an additional
recrystallization. Anal. Calcd for C7H12Cl2N2O: C, 39.83;
H, 5.73; Cl, 33.59; N, 13.27. Found: C, 39.80; H, 5.64; Cl,
33.95; N, 13.23. Melting point 205-208 °C.
1
1
analyzed by H NMR. H NMR (400 MHz, CDCl3) δ 9.13
(s, 1H), 8.79 (d, J ) 4.7 Hz, 1H), 8.23 (d, J ) 7.9 Hz, 1H),
7.45 (dd, J ) 4.7, 7.9 Hz, 1 H), 4.70 (s, 2H).
To the same reactor previously mentioned, without further
rinsing, was added THF (7.8 L), (R)-(+)-R,R-diphenylpro-
linol (Yunichem, 98.9 g, 391 mmol) and trimethyl borate
(58.7 mL, 523 mmol). The mixture was stirred for 1 h at 25
°C. A bleach scrubber was connected to the vent line of the
reactor, and 2 M borane-dimethyl sulfide in THF (6.87 kg,
16.1 mol) was then charged to the solution. The chloroketone
solution previously mentioned was then added at 23 mL/
min for a total addition time of 11.3 h. After the addition
was complete, the solution was stirred for 1 h and quenched
with methanol (2.5 L) over a period of 2 h to avoid rapid
hydrogen evolution. Solvent (24 L) was then removed in
Vacuo. A sample of the reaction mixture was analyzed prior
to distillation. Chiral HPLC30 indicated 95.7% ee. 1H NMR
(400 MHz, DMSO-d6) δ 8.58 (s, 1H), 8.49 (d, J ) 5.7 Hz,
1H), 8.15 (d, J ) 8.0 Hz, 1H), 7.69 (dd, J ) 5.7, 8.0 Hz,
1H), 6.20 (d, J ) 4.7 Hz, 1H), 5.05 (q, J ) 4.9 Hz, 1H),
3.85 (dq, J ) 5.1, 8.5 Hz, 2H), 2.50 (b, 3H).
Acknowledgment
The authors thank Romulo Romero for analytical support
and Dr. Andreas Lender for fruitful discussions about the
chemistry and for providing a chiral HPLC method for the
final product.
Supporting Information Available
Table of data for Chart 1. This material is available free
of charge via the Internet at http://pubs.acs.org.
(30) Chiralpak AD 4.6 mm × 250 mm; 1:9 (v/v) ethanol/hexane; 1.0 mL/min;
isocratic 25 min; room temp; UV 254 nm.
(31) Corey, E. J.; Helal, C. J. Angew. Chem., Int. Ed. 1998, 37, 1986-2012.
(32) Corey, E. J.; Bakshi, R. K.; Shibata, S.; Chen, C.; Singh, V. K. J. Am.
Chem. Soc. 1987, 109, 7925-7926.
(33) Corey, E. J.; Link, J. O.; Bakshi, R. K. Tetrahedron Lett. 1992, 33, 7107-
7110.
Received for review February 4, 2003. Accepted March 14,
2003.
OP0340208
288
•
Vol. 7, No. 3, 2003 / Organic Process Research & Development