One-Carbon Chain Extension of Esters to α-Chloroketones: A Safer Route without Diazomethane

Article abstract of DOI:10.1021/jo035733r

The reaction of a variety of methyl esters with dimethylsulfoxonium methylide at 0-25 °C affords the chain-extended β-keto dimethylsulfoxonium ylides. Subsequent treatment with hydrogen chloride in THF proceeds with loss of DMSO to afford the corresponding α-chloroketones. This sequence has been utilized to convert the methyl esters of CBZ-protected alanine and valine to the anti N-protected α-amino epoxides, which are important pharmaceutical intermediates. When the same protocol is applied to BOC-protected phenylalanine methyl ester, epimerization occurs so that the use of a more reactive aryl ester is required. This chemistry provides a practical route to α-chloroketones that avoids the use of toxic and explosive diazomethane.

Full text of DOI:10.1021/jo035733r

On e-Ca r bon Ch a in Exten sion of Ester s to r-Ch lor ok eton es:
A Sa fer Rou te w ith ou t Dia zom eth a n e
Dengjin Wang, Mark D. Schwinden,† Lilian Radesca,‡ Bharat Patel, David Kronenthal,
Ming-Hsing Huang, and William A. Nugent*
Process Research and Development Department, Bristol-Myers Squibb Pharmaceutical Research Institute,
Princeton, New J ersey 08543
william.nugent@bms.com
Received November 25, 2003
The reaction of a variety of methyl esters with dimethylsulfoxonium methylide at 0-25 °C affords
the chain-extended â-keto dimethylsulfoxonium ylides. Subsequent treatment with hydrogen
chloride in THF proceeds with loss of DMSO to afford the corresponding R-chloroketones. This
sequence has been utilized to convert the methyl esters of CBZ-protected alanine and valine to the
anti N-protected R-amino epoxides, which are important pharmaceutical intermediates. When the
same protocol is applied to BOC-protected phenylalanine methyl ester, epimerization occurs so
that the use of a more reactive aryl ester is required. This chemistry provides a practical route to
R-chloroketones that avoids the use of toxic and explosive diazomethane.
In tr od u ction
nelfinavir.7 Moreover, even achiral R-chloroketones are
of increasing interest as pharmaceutical building blocks.
Ikariya, Noyori, and co-workers recently reported8 that
these compounds undergo highly catalytic asymmetric
transfer hydrogenation to the â-chlorohydrins (eq 3),
opening an efficient route to enantiopure styrene oxides.
Organic chemists have a love/hate relationship with
diazomethanesand for good reason. On one hand, diaz-
omethane has proven to be a uniquely useful reagent.1
For example, it has been widely utilized for the chain
extension of carboxylic acids (via the mixed anhydrides)
to the corresponding R-chloroketones, including the
important amino acid-derived chloroketones2,3 exempli-
fied in eq 1. The products from eq 1 are readily converted
Balanced against the utility of diazomethane are the
unique hazards associated with its use. This gaseous
reagent is highly toxic, shock-sensitive, and explosive.9
For intermediate-scale manufacturing, engineering solu-
tions can be applied which limit the steady-state amount
of CH2N2 in a reaction vessel and make these hazards
manageable.10 However, when faced with the need for
a large-scale manufacturing process for N-protected
R′-amino-R-chloroketones, we felt an urgent obligation to
develop a safer alternative to eq 1.
to N-protected R-amino epoxides (eq 2), which are ideal
precursors for the hydroxyethylamine class of protease
inhibitors. Consequently, these building blocks are of
interest in the fields of both antihypertensives4 and HIV
protease inhibitors such as saquinavir,5 amprenavir,6 and
In this regard, we were attracted to a 1987 report of
Elkik and Imbeaux-Oudotte.11 Building on earlier obser-
vations by others,12,13 these researchers demonstrated a
(6) Kim, E. E.; Baker, C. T.; Dwyer, M. D.; Murcko, M. A.; Rao, B.
G.; Tung, R. D.; Navia, M. A. J . Am. Chem. Soc. 1995, 117, 1181.
(7) Kaldor, S. W.; Kalish, V. J .; Davies, J . F.; Shetty, B. V.; Fritz, J .
E.; Appelt, K.; Burgess, J . A.; Companele, K. M.; Chirgadze, N. Y.;
Clawson, D. K.; Dressman, B. A.; Hatch, S. D.; Khalil, D. A.; Kosa, M.
B.; Lubbehusen, P. P.; Muesing, M. A.; Patick, A. K.; Reich, S. H.; Su,
K. S.; Tatlock, J . H. J . Med. Chem. 1997, 40, 3979.
(8) Hamada, T.; Torii, T.; Izawa, K.; Noyori, R.; Ikariya, T. Org. Lett.
2002, 4, 4373. See also: Ma, Y.; Liu, H.; Chen, L.; Cui, X.; Zhu, J .;
Deng, J . Org. Lett. 2003, 5, 2103.
(9) (a) Dangerous Prop. Ind. Mater. Rep. 1992, 12, 530. (b) de Boer,
Th. J .; Backer, H. J . Organic Syntheses; Wiley: New York, 1963;
Collect. Vol. IV, p 250.
(10) Archibald, T. G. Chim. Oggi 2000, 18, 34.
(11) Elkik, E.; Imbeaux-Oudotte, M. Bull. Soc. Chim. Fr. 1987, 5,
861.
† Current address: SCYNEXIS, Inc., Research Triangle Park, NC.
‡ Current address: Millenium Pharmaceuticals, Cambridge, MA.
(1) (a) Black, T. H. Aldrichim. Acta 1983, 16, 3. (b) Hopps, H. B.
Aldrichimica Acta 1970, 3, 9.
(2) Penke, B.; Czombos, J .; Balaspiri, L.; Petres, J .; Kovacs, K. Helv.
Chim. Acta 1970, 53, 1057.
(3) Albeck, A.; Persky, R. Tetrahedron 1994, 50, 6333.
(4) Delaria, J . F.; Maki, R. G.; Bopp, B. A.; Cohen, J .; Kleinert, H.
D.; Luly, J . R.; Merits, I.; Plattner, J . J .; Stein, H. H. J . Med. Chem.
1987, 30, 2137.
(5) Roberts, N. A.; Martin, J . A.; Kinchington, D.; Broadhurst, A.
V.; Craig, J . C.; Duncan, I. B.; Galphin, S. A.; Handa, B. K.; Krohn,
A.; Lambert, R. W.; Merrett, J . H.; Mills, J . S.; Parkes, K. E. B.;
Redshaw, S.; Ritchie, A. J .; Taylor, D. L.; Thomas, G. J .; Machin, P. J .
Science 1990, 248, 358.
10.1021/jo035733r CCC: $27.50 © 2004 American Chemical Society
Published on Web 02/11/2004
J . Org. Chem. 2004, 69, 1629-1633
1629

Wang et al.
novel two-step conversion of several fluorine-containing
R,â-unsaturated esters to the corresponding R-chloroke-
tones (eq 4). This sequence involved treatment of the
While 4a -c could be isolated in high yield (typically
>80%) by flash chromatography, it is generally more
convenient to crystallize the crude products from hot
ethyl acetate. After drying, the â-keto sulfur ylides are
obtained as analytically pure, white crystalline solids.
Yields for compounds 4a -c were 65%, 53%, and 80%,
respectively. In the case of completely unactivated methyl
caprylate, 3d , the reaction was significantly slower. Even
with 5 equiv of reagent and increasing the reaction time
to 72 h, 4d was isolated after chromatography in only
25% yield.
Attempts to replace the excess sulfur ylide in eq 5 with
other bases were generally unsuccessful. These experi-
ments are discussed in the Supporting Information.
We then examined the cleavage reaction of 4a and 4b
with anhydrous hydrogen chloride. In both cases, treat-
ment of a THF solution of the â-keto sulfur ylide at room
temperature results in formation of a sparingly soluble
salt (eq 6). Heating the resultant suspension at 65 °C
ester with dimethylsulfoxonium methylide to afford an
intermediate â-keto sulfur ylide followed by protonation
with HCl and thermolysis of the resultant salt.
If this transformation could be extended to N-protected
R-amino esters, it would provide an attractive alternative
to eq 1. However, two significant issues needed to be
addressed. Would the N-protected R-amino esters survive
exposure to the basic sulfur ylide without significant
epimerization? Beyond that, would the addition of dim-
ethylsulfoxonium methylide to an ester occur in the
absence of a strongly activating fluorine substituent?
The latter question was of especial concern in light of
two literature reports. Corey and Chaykovsky14 examined
the reaction of dimethylsulfoxonium methylide with ester
1, which lacks the fluorine substituent of eq 4. They
demonstrated that 1 undergoes Michael addition to the
CdC bond while the ester group is recovered unchanged.
Moreover, Nagao and co-workers15 treated a series of
methyl esters of structure 2 with dimethylsulfoxonium
methylide and observed cleavage of the C-N bond rather
than attack at the ester moiety.
Given these concerns, we elected to first delineate the
scope of the reaction of achiral methyl esters with
dimethylsulfoxonium methylide.
for 2 (for 4a ) or 16 h (for 4b) results in clean conversion
to the R-chloroketone. Extractive workup affords 5a and
5b in 97% and 92% yield, respectively.
Rhodium-catalyzed transfer hydrogenation of 5b af-
fords the â-chlorohydrin in 90% yield and 92% ee.8a
Chloroketone 5a has been reduced enzymatically.17
Ch a in Exten sion of r-Am in o Ester s. Encouraged
by the preceding results, we turned our attention to the
more sensitive N-protected methyl esters derived from
alanine, valine, and phenylalanine (6a -c, R′ ) methyl).
In each case, treatment with dimethylsulfoxonium me-
thylide, generated as above, resulted in conversion to the
corresponding â-keto sulfur ylide (eq 7). Results are
Resu lts a n d Discu ssion
Syn th esis of Ach ir a l r-Ch lor ok eton es. A series of
four diverse methyl esters, 3a -d , was selected for study.
Dimethylsulfoxonium methylide16 was generated by treat-
ment of trimethylsulfoxonium chloride with potassium
tert-butoxide in refluxing THF. Treatment of 3a or 3c
with 3 equiv of the resulting solution at room tempera-
ture overnight resulted in complete conversion to the
corresponding â-keto sulfur ylide according to eq 5. In
the case of 3b, 5 equiv of the reagent and a longer
reaction time (48 h) were required for complete conver-
sion. (When dimethylsulfoxonium methylide was gener-
ated from trimethylsulfoxonium iodide, the reaction was
not as clean and yields were diminished by 10-15%.)
summarized in Table 1. Again generating the methylide
from trimethylsulfoxonium iodide rather than the chlo-
ride resulted in a lower yield, although the reduction in
yield was not as severe in the absence of visible light.18
Nominal yields in Table 1 are for crude products which
were 85-95% pure by NMR. The remaining material
appears to be oligomeric sulfur compounds as indicated
(12) For the reaction of dimethylsulfoxonium methylide with lac-
tones, see: Mukaiyama, T.; Higo, M.; Takei, H. Bull. Chem. Soc. J pn.
1970, 43, 2566.
(13) For reactions of related sulfur ylides with HCl, see: (a) DeGraw,
J . I.; Cory, M. Tetrahedron Lett. 1968, 2501. (b) Koenig, H.; Metzger,
H. Chem. Ber. 1965, 98, 3733.
(14) Corey, E. J .; Chaykovsky, M. J . Am. Chem. Soc. 1964, 86, 1640.
(15) Nagao, Y.; Ikeda, T.; Yagi, M.; Fujita, E.; Shiro, M. J . Am. Chem.
Soc. 1982, 104, 2079.
1
by broad singlets at δ 0.85 and 1.25 in the H NMR.
(16) Corey, E. J .; Chaykovsky, M. J . Am. Chem. Soc. 1965, 87, 1353.
These authors note that phenyl esters would be expected to react at
the ester functionality although no examples are given.
(17) Martinez, F.; Del Campo, C.; Sinisterra, J . V.; Llama, E. F.
Tetrahedron: Asymmetry 2000, 11, 4651.
(18) We thank Prof. Barry M. Trost for this suggestion.
1630 J . Org. Chem., Vol. 69, No. 5, 2004

Chain Extension of Esters to R-Chloroketones
TABLE 1. Rea ction of N-P r otected r-Am in o Ester s w ith
TABLE 3. Red u ction of Ala n in e-Der ived Ch lor ok eton e
8a Accor d in g to Eq 9a
Dim eth ylsu lfoxon iu m Meth ylid e Accor d in g to Eq 7a
ester
R′
X
yieldb
eec
reductant
NaBH4/EtOH
NaBH4/EtOH
(R)-(BINAP)RuCl2/H2
(S)-BINAP)RuCl2/H2
temp (°C)
t (h)
drb
6a
6b
6c
6c
6a
6b
6c
Me
Me
Me
Me
Npf
Npf
Npf
Cl
Cl
Cl
I
I
I
97
98d
87
99
90
<2
<2
>99
>99
>99
0
-70
50
0.5
1c
24
2.3/1
2.9/1
23/1
59e
67
50
48
1/2.2
a
Catalytic hydrogenations were carried out in methanol under
100
99
65 psi of hydrogen at substrate/catalyst ) 100:1; see the Experi-
I
b
mental Section. Diastereomer ratio of anti to syn chlorohydrins
a
Reactions were 0.1 M in THF at 0 °C, using 3 equiv of sulfur
by HPLC. c 15% of unreacted chloroketone remained after HOAc
quench.
b
ylide; see the Experimental Section. Nominal yield of crude
product after extractive workup and vacuum drying. c Determined
d
by HPLC after conversion to R-chloroketone. 5 equiv of sulfur
reaction. With alternative HCl sources (tetrabutylam-
monium chloride/MsOH or concentrated aqueous HCl),
cleavage of the methyl C-S bond became a significant
competing pathway (up to 37% in polar solvents such as
DMSO).
ylide was used. e Reaction run with exclusion of light. f Np )
4-nitrophenyl.
TABLE 2. Con ver sion of Su lfu r Ylid es to Ch lor ok eton es
Accor d in g to Eq 8a
a n ti-r-N-P r otected Am in o Ep oxid es. Although it
is not the principal focus of this paper, we wish to record
several observations regarding the reduction of chloroke-
tones 8a -c to the corresponding â-chlorohydrins (eq 9)
R′
PG
product
yield (%)b
ee (%)
Me
iPr
Bn
CBZ
CBZ
BOC
8a
8b
8c
70
>99
>99
>99
75c,d
81
a
Conditions: substrate/MsOH/LiCl ) 1.0:1.1:1.2 in solvent
b
THF, 65 °C, 4 h; see the Experimental Section. Isolated yield of
chloroketone after crystallization. c Reaction time 48 h. Yield
d
after flash chromatography.
Determination of the ee of the products from eq 7 was
most readily accomplished by supercritical fluid chroma-
tography after conversion to the corresponding R-chlo-
roketones. As shown in Table 1, the optical purity for 7a
was excellent and remained in the synthetically useful
range for 7b. However, for the phenylalanine-derived 7c
we observed essentially complete racemization.
To address this problem, we examined the use of
4-nitrophenyl (Np) esters in place of the methyl esters
as starting materials. The requisite Np esters are readily
available and the 4-nitrophenoxide group functions as an
excellent nucleofuge. With these alternative substrates,
we observed clean conversion to the â-keto sulfur ylides
without detectable epimerization even when using tri-
methylsulfoxonium iodide as precursor to the methylide
(Table 1). This is, of course, advantageous since the iodide
is the inexpensive reaction product of methyl iodide with
DMSO.
and their subsequent cyclization to the epoxides. Both
of these steps have been studied in some detail by
others.19 When sodium borohydride is used as reducing
agent for eq 9, the anti diastereoselectivity of the reduc-
tion is sensitive to the steric bulk of the R group and
i
decreases along the series R ) Pr > Bn > Me.
Consistent with this, treatment of valine-derived chlo-
roketone 8b with sodium borohydride in ethanol at 0 °C
proceeded in 92:8 dr and the anti â-chlorohydrin could
be isolated by crystallization in 84% yield. In contrast,
sodium borohydride reduction of phenylalanine-derived
8c proceeded with only 4:1 selectivity for the anti
â-chlorohydrin. Nevertheless, crystallization of the crude
product from hot ethyl acetate afforded the pure (S,S)-
chlorohydrin in 50% yield without the need for chroma-
tography.
Reduction of the sterically small alanine-derived chlo-
roketone 8a represents a particular challenge and results
are summarized in Table 3. Reduction with sodium
borohydride proceeded in 2.3:1 diastereoselectivity at 0
°C and this improved to only 2.9:1 at -70 °C. It occurred
to us that higher diastereoselectivity might be achieved
by hydrogenation with an asymmetric homogeneous
catalyst. Entries 3 and 4 of Table 3 provide evidence for
“matched versus mismatched” interaction between (BI-
NAP)RuCl2 as catalyst20 and the chloroketone substrate.
With use of (R)-BINAP as the ligand and the chloroke-
tone derived from natural L-alanine as substrate, 23:1
anti/syn diastereoselectivity could be achieved. Formation
of the desired anti-(S,S)-isomer was confirmed by conver-
Conversion of the â-keto sulfur ylides to the R-chlo-
roketones (eq 8) proved straightforward in all cases. The
results shown in Table 2 are for the â-keto sulfur ylides
prepared from the Np esters. Chloroketones 8a and 8c
were crystallized from 80:20 hexanes/ethyl acetate. Chlo-
roketone 8b was an oil, which was purified by flash
chromatography. The lower yields in Table 2 versus those
for cleavage of 4a and 4b reflect, at least in part, the
use of crude (uncrystallized) â-keto sulfur ylides as
starting materials.
(19) Rotella, D. P. Tetrahedron Lett. 1995, 36, 5453 and references
therein.
(20) For the asymmetric hydrogenation of an achiral R-chloroketone
with (BINAP)RuCl2, see: (a) Kitamura, M.; Ohkuma, T.; Takaya, H.;
Noyori, R. Tetrahedron Lett. 1988, 29, 1555. (b) Davies, S. G.; Ichihara,
O. Tetrahedron: Asymmetry 1996, 7, 1919.
As a practical source of HCl, the combination of lithium
chloride and methanesulfonic acid provided the cleanest
J . Org. Chem, Vol. 69, No. 5, 2004 1631

Wang et al.
sion to the corresponding epoxide and comparison with
an authentic sample.
approach and the hazards associated with diazomethane,
we would certainly recommend that synthetic chemists
also consider this methodology for the laboratory scale
preparation of R-chloroketones.
Ring-closure of the N-protected â-chlorohydrins is
again well precedented in the literature21 and proceeded
uneventfully. For example, epoxide 10 was formed in 96%
yield and >99% ee upon treatment with potassium tert-
butoxide in a 2-propanol/THF mixed solvent system (eq
10).
Exp er im en ta l Section
Dim eth ylsu lfoxon iu m 2-Oxo-3-(ph en oxy)pr opylide, 4a.
A 100-mL flask was charged with trimethylsulfoxonium
chloride (3.90 g, 30 mmol), tetrahydrofuran (30 mL), and 1.0
M potassium tert-butoxide in THF solution (31.5 mL, 31.5
mmol). The suspension was heated at reflux for 2 h then cooled
to room temperature. Methyl phenoxyacetate (1.66 g, 10 mmol)
was added dropwise over 10 min whereupon the mixture was
stirred overnight at room temperature. After distillation of the
solvent, water (10 mL) was added and the product was
extracted into ethyl acetate (2 × 50 mL). The organic phase
was washed with brine (2 × 10 mL), dried over sodium sulfate,
and concentrated in a vacuum to afford the crude product (2.1
g). Recrystallization from ethyl acetate (10 mL) afforded the
sulfur ylide (1.47 g, 65%) as a white crystalline solid. Anal.
Calcd for C11H14O3S: C, 58.38; H, 6.34. Found: C, 58.50; H,
Although our primary interest was in the large-scale
manufacturing of R-amino epoxides, this chemistry works
quite well on a laboratory scale. As a demonstration of
this, we have carried out the (R)-(BINAP)RuCl2 reduc-
tion/cyclization sequence on a 30-g scale (eq 11). The
1
6.10. H NMR (400 MHz, CDCl3): δ 3.40 (s, 6H), 4.38 (s, 2H),
4.90 (s, 1H), 6.83-6.96 (m, 3H), 7.24 (m, 2H). 13C NMR (100
MHz, CDCl3): δ 41.8, 69.3, 70.3, 114.5, 121.0, 129.4, 158.0,
185.5.
1-Ch lor o-3-p h en oxy-2-p r op a n on e, 5a . A glass tube was
charged with 4a (0.91 g, 4.0 mmol), THF (15 mL), and 4 M
HCl in dioxane solution (1.15 mL, 4.6 mmol). The tube was
sealed with a Teflon stopper and stirred at room temperature
for 10 min then for 2 h in a 70 °C heated block at which time
the mixture was homogeneous. After cooling, the solvent was
removed at reduced pressure. The residue was added to water
(15 mL) and extracted into methyl tert-butyl ether (2 × 10 mL).
Removal of the solvent afforded 5a as a colorless oil (0.72 g,
97%) which crystallized upon standing to a low melting solid.
Anal. Calcd for C9H9ClO2: C, 58.55; H, 4.91. Found: C, 58.57;
H, 5.00. 1H NMR (400 MHz, CDCl3): δ 4.27 (s, 2H), 4.60 (s,
2H), 6.77 (d, J ) 8.0 Hz, 2H), 6.90 (t, J ) 7.4 Hz, 1H), 7.19
(m, 2H). 13C NMR (100 MHz, CDCl3): δ 46.7, 71.3, 114.6, 121.6,
129.6, 157.1, 198.8.
Dim eth ylsu lfoxon iu m (3S)-2-Oxo-3-(ben zyloxyca r bo-
n yla m in o)bu tylid e, 7a . A solution of 1.0 M potassium tert-
butoxide in THF (60 mL, 60 mmol) was added at room
temperature to a suspension of trimethylsulfoxonium chloride
(7.72 g, 60 mmol) in THF (40 mL). The mixture was heated at
reflux for 2 h and was then cooled to 0 °C. A solution of CBZ-
alanine methyl ester (4.74 g, 20 mmol) in THF (10 mL) was
added dropwise at 0 °C and the resultant solution was stirred
for 4 h at 0 °C. The reaction was quenched with water (20
mL). The organic layer was separated and washed with brine
(2 × 20 mL). The solvent was removed at reduced pressure to
afford the crude sulfur ylide (5.76 g, 97%) as an off-white solid.
1H NMR (400 MHz, CDCl3): δ 1.30 (d, J ) 16.6 Hz, 3H), 3.34
(s, 3H), 3.36 (s, 3H), 4.16 (m, 1H), 4.61 (S, 1H), 5.08 (s, 2H),
5.81 (d, J ) 7.0 Hz, 1H), 7.28 (m, 5H). 13C NMR (100 MHz,
CDCl3): δ 20.3, 42.0, 42.2, 53.2, 66.7, 68.7, 109.8, 128.2, 128.7,
136.9, 155.9, 188.7.
(3S)-1-Ch lor o-3-(b en zyloxyca r b on yla m in o)-2-b u t a n -
on e, 8a . Crude 7a (6.18 g, 22.0 mmol) was dissolved in THF
(100 mL) and the solution was cooled to 0 °C. Lithium chloride
(1.6 g, 25 mmol) and methanesulfonic acid (1.6 mL, 24 mmol)
were added. The temperature was slowly raised to 70 °C and
stirring was continued at that temperature for 2 h. After
cooling, the reaction was quenched by addition of water (100
mL). The phases were separated and 2:1 heptane/ethyl acetate
(100 mL) was added. To remove DMSO, the organic layer was
washed with saturated NaHCO3 (20 mL), water (2 × 20 mL),
and brine (20 mL) and dried over Na2SO4. Removal of solvent
afforded the crude chloroketone (5.0 g). The enantiomeric
excess was determined to be >99% by supercritical fluid
chromatography (Chiralpak AD-H, 150 × 4.6 mm, 5 µ particle
methanolic solution of â-chlorohydrin obtained from the
hydrogenation was filtered through Celite and treated
with a 10% excess of 1 M sodium hydroxide overnight.
The overall yield of crystalline 11 (>99% ee) for this two-
step sequence was 67%.
Con clu sion
We have demonstrated a straightforward two-step
chain extension of esters to R-chloroketones, which avoids
the use of diazomethane. We have found that even
modestly activated methyl esters will react with dimeth-
ylsulfoxonium methylide to afford the chain-extended
â-keto sulfur ylides in synthetically useful yields. Where
applicable, this is advantageous since methyl esters can
generally be prepared without the intermediacy of a
moisture-sensitive acid chloride. Nevertheless, for unac-
tivated or base-sensitive substrates, the use of the more
reactive aryl esters (such as the 4-nitrophenyl esters) may
be required.
As one application of this methodology, we have
demonstrated a novel and practical route to anti N-
protected R-amino epoxides. It should be noted that the
same chloroketones 8a -c can also be reduced enzymati-
cally22 to the corresponding syn chlorohydrins so that our
route can be used to access both sets of diastereomeric
epoxides. An attractive feature of this chemistry as a
practical route23 to the anti epoxides is that, with the
exception of compound 8b, all of the intermediates are
crystalline solids and hence no chromatographic purifica-
tion is required.
In our opinion, this chemistry represents a useful
advance in the manufacture of R-chloroketones as phar-
maceutical intermediates. Given the simplicity of this
(21) Chen, P.; Cheng, P. T. W.; Spergel, S. H.; Zahler, R.; Wang, X.;
Thottathil, J .; Barrish, J . C.; Polniaszek, R. P. Tetrahedron Lett. 1997,
38, 3175. See also refs 2 and 3.
(22) Patel, R. N.; Banerjee, A.; McNamee, C. G.; Brzozowski, D. B.;
Szarka, L. J . Tetrahedron: Asymmetry 1997, 8, 2547.
(23) Kronenthal, D.; Schwinden, M. D. U.S. Patent 6,399,793, 2002.
1632 J . Org. Chem., Vol. 69, No. 5, 2004

Chain Extension of Esters to R-Chloroketones
size, 15% methanol in CO2 mobile phase, 40 °C, 2 mL/min,
150 bar). Retention times for the enantiomers were assigned
by using an authentic racemate as 3.00 and 3.18 min. Crystal-
lization of the product from hot 5:1 heptane/ethyl acetate (50
mL) afforded pure 8a (2.9 g, 70%) as a snow-white solid. Anal.
Calcd for C12H14ClNO3: C, 56.37; H, 5.52; N, 5.48. Found: C,
56.47; H, 5.30; N, 5.35. 1H NMR (400 MHz, CDCl3): δ 1.44 (d,
J ) 7.1 Hz, 3H), 4.32 (s, 2H), 4.64 (m, 1H), 5.12 (s, 2H), 5.48
(br s, 1H), 7.36 (m, 5H). 13C NMR (100 MHz, CDCl3): δ 17.8,
46.2, 53.7, 67.4, 128.4, 128.5, 128.8, 136.2, 155.9, 201.8.
(2S,3S)-1-Ch lor o-3-(b en zyloxyca r b on yla m in o)-2-b u -
ta n ol, 9a . Chloroketone 8a (30.0 g, 117 mmol) was dissolved
in methanol (300 mL). The solution was purged with N2.
Hydrochloric acid in isopropyl alcohol 5-6 N solution (2.4 mL)
and [(R)-BINAP]RuCl2 (2.6 g, 3.3 mmol) was then added and
the mixture was purged with N2 and H2. The mixture was
heated to 40 °C and hydrogenation was performed at 2.6 bar
of pressure. After reaction was complete (2.64 L of H2
absorbed), the mixture was allowed to cool to 25 °C and was
filtered through Celite. The anti/syn ratio was determined to
be 23:1 with a 250 mm Zorbax EclipseXDB-C18 column (1.5
mL/min, 25-min runs), using a solvent gradient as follows.
Solvent 1 ) 0.1% trifluoroacetic acid in water, solvent 2 ) 0.1%
trifluoroacetic acid in acetonitrile; the initial ratio water/
acetonitrile was 95:5 and this was ramped to 5:95 over the
course of 23 min. A small analytical sample of the product was
prepared by crystallization from hot hexane/ethyl acetate.
Anal. Calcd for C12H16ClNO3: C, 55.93; H, 6.26; N, 5.44.
Found: C, 55.83; H, 6.07; N, 5.63. 1H NMR (400 MHz,
CDCl3): δ 1.15 (d, J ) 6.6 Hz, 3H), 3.29 (d, J ) 4.3 Hz, 1H),
3.48 (dd, J ) 11.2, 7.3 Hz, 1H), 3.50 (dd, J ) 11.2, 3.8 Hz,
1H), 3.84 (m, 1H), 3.90 (m, 1H), 5.09 (m, 2H), 5.25 (d, J ) 8.1
Hz, 1H), 7.30 (s, 5H). 13C NMR (100 MHz, CDCl3): δ 14.1, 46.0,
48.2, 66.0, 73.2, 127.1, 127.2, 127.6, 135.2, 155.0.
[(1S)-1-(2S)-Oxir a n yleth yl]ca r ba m ic Acid Ben zyl Es-
ter , 11. The remainder of the filtered solution of 9a from the
preceding example was treated with 1 M aqueous NaOH
solution (130 mL, 0.13 mol) and stirred at room temperature
overnight. The solution was concentrated under vacuum and
toluene (225 mL), tert-butyl methyl ether (75 mL), and water
(30 mL) were added. Layers were separated. The aqueous
phase was twice extracted with a mixture of tert-butyl methyl
ether (150 mL) and toluene (75 mL). All organic layers were
combined and washed twice with water (100 mL), and then
concentrated to an oil. The oil was dissolved in isopropyl
acetate (120 mL) and purified by filtration through a silica
gel column and crystallization obtaining 17.0 g of epoxide 11
(67% overall yield). Anal. Calcd for C12H15NO3: C, 65.14; H,
6.83; N, 6.33. Found: C, 65.19; H, 6.71; N, 6.42. 1H NMR (400
MHz, CDCl3): δ 1.20 (d, J ) 6.9 Hz, 3H), 2.70 (br d, 2H), 2.90
(br s, 1H), 3.72 (m, 1H), 4.9-5.1 (m, 3H), 7.38 (m, 5H). 13C
NMR (100 MHz, CDCl3): δ 16.6, 46.3, 48.1, 54.6, 67.0, 128.3,
128.4, 128.8, 136.5, 156.0.
Ack n ow led gm en t . The authors thank Professors
J ames M. Mayer, Barry M. Trost, and Edwin Vedejs for
helpful discussions.
Su p p or tin g In for m a tion Ava ila ble: Additional experi-
mental details and a summary of preliminary mechanistic
studies. This material is available free of charge via the
Internet at http://pubs.acs.org.
J O035733R
J . Org. Chem, Vol. 69, No. 5, 2004 1633