Asymmetric fluorination of α-aryl acetic acid derivatives with the catalytic system NiCl2-binap/R3SiOTf/2,6-lutidine

Article abstract of DOI:10.1002/anie.200701071

(Chemical Equation Presented) Not binary, but trinary: A binary system consisting of Ni(OTf)₂-binap complex and 2,6-lutidine failed to promote the asymmetric fluorination of ester equivalents. However, upon the addition of a substoichiometric a

Full text of DOI:10.1002/anie.200701071

Angewandte
Chemie
DOI: 10.1002/anie.200701071
Asymmetric Catalysis
Asymmetric Fluorination of a-Aryl Acetic Acid Derivatives with the
Catalytic System NiCl2–Binap/R3SiOTf/2,6-Lutidine**
Toshiaki Suzuki, Yoshitaka Hamashima, and Mikiko Sodeoka*
Dedicated to Professor Masakatsu Shibasaki on the occasion of his 60th birthday
As a reflection of the importance of organofluorine com-
pounds in medicinal and agricultural chemistry and materials
science,[1] catalytic asymmetric fluorination has been attract-
ing increasing attention.[2,3] We and others have developed
elegant systems for the catalytic asymmetric fluorination of
carbonyl compounds, including b-ketoesters,[4] b-ketophos-
phonates,[5] cyanoacetates,[6] oxindoles,[7] and primary alde-
hydes.[8] In contrast, the preparation of chiral a-fluorocarbox-
ylic acids has not been explored as thoroughly, even though
such compounds have great potential in medicinal chemis-
try.[9] Only a few synthetically useful diastereoselective
fluorination reactions of ester equivalents are known.[10,11]
No catalytic method has been developed so far, probably
owing to the difficulty in the in situ generation of metal
enolates (or enols) of ester equivalents under catalytic
conditions.
Although N-acyl thiazolinethiones were used successfully in
a direct catalytic asymmetric aldol reaction,[14] 2 and the
related compound 3 were found to be unstable in the presence
of N-fluorobenzenesulfonimide (NFSI), probably owing to
=
the high reactivity of the C S group towards the electrophilic
fluorine atom. In contrast, substrates 4a and 4b with an
oxycarbonyl group were insensitive to NFSI. As we could not
isolate the fluorinated product derived from 4b, and one of
the enantiomers of the product was inseparable from 4b by
HPLC on a chiral phase, we initially selected 4a as a model
compound.
We reported recently a unique reaction in which tandem
asymmetric fluorination–methanolysis of an N-Boc-protected
oxindole yielded a chiral monofluorinated aryl acetate (Boc =
tert-butoxycarbonyl).[12] To our knowledge, this reaction is the
only catalytic asymmetric synthesis of an a-fluorinated ester
through a fluorination reaction so far. However, the ester
moiety was formed indirectly through opening of the oxindole
ring; the structures of available starting materials are thus
highly restricted. As the aryl acetic acid moiety is found in
various important medicines, such as nonsteroidal anti-
inflammatory drugs, the availability of chiral a-monofluori-
nated aryl acetic acids is expected to be useful for drug
synthesis. As part of our research on the synthesis of chiral
organofluorine compounds, we present herein a novel cata-
lytic asymmetric fluorination of (2-aryl acetyl)thia- and
oxazolidin-2-ones with the unique reaction system NiII–
binap/R3SiOTf/2,6-lutidine (Scheme 1; binap = 2,2’-bis(di-
phenylphosphanyl)-1,1’-binaphthyl, Tf = trifluoromethane-
sulfonyl).
Scheme 1. Catalytic asymmetric fluorination.
We anticipated that cooperative activation by a cationic
metal complex and an organic base would promote the
formation of a metal enolate from 4a (Scheme 2)[15] and
We first tested compounds 2–4, expected to react via a
bidentate metal enolate, in the fluorination reaction.[13]
[*] T. Suzuki, Dr. Y. Hamashima, Prof. M. Sodeoka
RIKEN (Institute of Physical and Chemical Research)
2-1 Hirosawa, Wako
Saitama 351-0198 (Japan)
Fax: (+81)48-467-9373
E-mail: sodeoka@riken.jp
[**] This research was supported financially through a Grant-in-Aid on
Priority Areas, a Grant-in Aid for the Encouragement of Young
Scientists (B), and Special Project Funding for Basic Science.
Binap=2,2’-bis(diphenylphosphanyl)-1,1’-binaphthyl, Tf=trifluoro-
methanesulfonyl.
Scheme 2. Concept for the catalytic fluorination of 4a.
Angew. Chem. Int. Ed. 2007, 46, 5435 –5439
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5435

Communications
examined a variety of reaction conditions. We varied the
the reaction was carried out at À208C, the uncatalyzed
racemic pathway was suppressed considerably, and the
ee value of 5a was improved to 68% (Table 1, entries 5 and
6). Interestingly, the neutral NiCl2–binap complex 1b could
also be used as an effective catalyst precursor and gave
comparable results (95%, 68% ee; Table 1, entry 7). Thus, the
same active species appears to be operative in the reactions
with 1a and 1b. The use of 1b is convenient in practical terms,
as 1b is bench stable.
While the nonstereoselective reaction to give the racemic
product was promoted by Me3SiOTf and 2,6-lutidine in
CH2Cl2 even at À208C, no reaction was observed in toluene.
When Et3SiOTf and toluene were used in the presence of 1b
and 2,6-lutidine, the reaction proceeded smoothly to give 5a
in quantitative yield with the highest enantioselectivity
observed in this study (88% ee; Table 1, entry 9). It was
possible to apply a substoichiometric amount of Et3SiOTf
(about 0.25 equiv; Table 1, entries 9–11), however, a stoichio-
metric amount of the base was required for a high chemical
yield.[21,22] Furthermore, the quantity of the catalyst could be
decreased to 5 mol% without significant deterioration of the
reaction efficiency. A drop in the selectivity was finally
observed when 2.5 mol% of 1b was used (Table 1, entries 12–
14). It is interesting that no formation of a difluorinated
compound and no racemization of the monofluorinated
product were observed in these reactions.[23]
metal complex, the organic base, the fluorine source, and the
solvent. Unfortunately, however, no reaction occurred with
any combination of reagents tested. Previously, we had
observed that independent simultaneous activation of both
the nucleophile and the electrophile is highly effective in
[16]
À
promoting C C bond-forming reactions. We hypothesized
that the desired fluorination reaction might occur if the NFSI
reagent was activated strongly, even though the amount of the
enolate formed in situ might be very small. Consequently, we
examined the effect of supplementary Lewis acids
(Scheme 2). It was expected that a chiral metal complex
with two vacant binding sites would activate 4a preferentially
through bidentate complexation, whereas a monodentate
secondary Lewis acid would interact with NFSI. A recent
report by Evans and Thomson supported the validity of this
idea.[17]
We tested several chiral metal complexes[18] and supple-
mentary Lewis acids[19] and found that the concurrent use of
the NiII complex 1a and Me3SiOTf was effective.[20] When
Me3SiOTf (1 equiv) was added as a supplementary Lewis
acid, the reaction of 4a proceeded smoothly in the presence of
1a (10 mol%) and 2,6-lutidine (1 equiv) to afford the desired
product 5a in 99% yield with 28% ee (Table 1, entry 1). The
With these optimized conditions established, we next
examined the generality of this reaction. As shown in Table 2,
the reaction is applicable to a variety of a-aryl acetic acid
derivatives. The electronic nature of the aromatic ring did not
appear to affect the reactivity of the para-substituted
substrates 4c and 4d, which underwent the fluorination
reaction smoothly with high enantioselectivity (Table 2,
entries 2 and 3). When 5 mol% of 1b was used, decreased
chemical yield and enantioselectivity were observed with the
Table 1: Optimization of the reaction conditions.
Entry Catalyst
R3SiOTf
Base
Solvent
T
Yield[a] ee[b]
([mol%]) [equiv] (R) [equiv]
[8C] [%]
[%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
1a (10)
1a (10)
1a (10)
–
1 (Me)
1 (Me)
–
1
–
3
3
CH2Cl2 RT[c] 99
CH2Cl2 RT[c] n.r.
CH2Cl2 RT[c] n.r.
CH2Cl2 RT[c] 85
CH2Cl2 À20 64
CH2Cl2 À20 90
CH2Cl2 À20 95
CH2Cl2 À20 99
toluene À20 99
toluene À20 98
toluene À20 77
toluene À20 99
toluene À20 96
toluene À20 91
28
–
–
2 (Me)
1 (Me)
1.5 (Me)
1.5 (Me)
1.5 (Et)
1.5 (Et)
0.5 (Et)
0.25 (Et)
0.75 (Et)
0.5 (Et)
0.5 (Et)
–
Table 2: Catalytic asymmetric fluorination.
1a (10)
1a (10)
1b (10)
1b (10)
1b (10)
1b (10)
1b (10)
1b (5)
1b (5)
1b (2.5)
1
68
68
68
70
88
88
86
88
86
76
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
Entry
4
X
R
1b
Et3SiOTf Yield[a] ee[b]
[mol%] [equiv]
[%]
[%]
1
2
3
4
5
4a SPh
5
0.75
99
88
4c
4d
4e
4e
4 f
4 f
S
S
S
S
S
S
p-FC6H4
p-MeOC6H4
m-MeOC6H4
5
5
5
0.75
0.75
0.75
1.5
0.75
1.5
90
92
56
95
73
87
83
81
69
82
61
78
[a] Yield of the isolated product. [b] The ee value was determined by
HPLC on a chiral phase. [c] Room temperature: 23–258C.
m-MeOC6H4 10
6
7
o-MeOC6H4
o-MeOC6H4
5
10
results shown in entries 2–4of Table 1 indicate that the triad
NiII/Me3SiOTf/2,6-lutidine is necessary for high conversion
and enantioselectivity. However, the observation that the
fluorination reaction proceeds in the absence of the Ni
complex (Table 1, entry 4) indicates that the low ee value of
the product in entry 1 is associated with the spontaneous
fluorination of the silyl enolate of 4a generated in situ. When
8
9
4g S2-naphthyl
4h S1-naphthyl
10
5
1.5
0.75
1.5
1.5
99
94
83
87
10
4b
4i
O
S
Ph
n-propyl
10
10
95
15
87
11
11[c]
[a] Yield of the isolated product. [b] The ee value was determined by
HPLC on a chiral phase. [c] The reaction was carried out at room
temperature.
5436
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5435 –5439

Angewandte
Chemie
corresponding ortho- and meta-methoxy-substituted sub-
strates (Table 2, entries 4and 6), but satisfactory results
were obtained when 10 mol% of the catalyst was used
(entries 5 and 7). The reactions of the bulkier naphthyl-
substituted substrates 4g and 4h also afforded the desired
products in excellent yield with high enantioselectivity
(Table 2, entries 8 and 9). Under these optimized conditions,
4b was converted completely into the desired product 5b,
which was isolated in 95% yield with 87% ee (Table 2,
entry 10). Unfortunately, however, this method for the
catalytic asymmetric a-monofluorination of carboxylic acid
derivatives is not applicable to (3-alkanoyl)thiazolin-2-ones,
such as 4i, probably as a result of their insufficient acidity
(Table 2, entry 11).
Scheme 4. Transformation of 5a.
A plausible catalytic cycle is outlined in Scheme 3.
31P NMR spectroscopic studies indicated that two Cl anions
on the complex 1b were replaced by TfO anions when 1b was
comparing the optical rotation of 6 with the reported value.
The sense of face selectivity observed in this reaction can be
explained by postulating the involvement of the bidentate
nickel enolate.[17]
To confirm the utility of our fluorination reaction, 5a was
converted into the corresponding carboxylic acid
7
(Scheme 4). The recrystallization of 5a from CH2Cl2/Et2O
(1:9) yielded racemic prisms, and almost optically pure 5a was
obtained from the mother liquid (88%, 99% ee). When
exposed to hydrolytic conditions, 5a was converted quantita-
tively into the carboxylic acid 7. To our delight, analysis of the
corresponding methyl ester by HPLC on a chiral phase
confirmed that no loss of optical purity had occurred.[25] For
the synthesis of a-fluorocarboxylic acids, the organocatalytic
fluorination of primary aldehydes is likely to be an appro-
priate choice. However, the conversion of the products into
the corresponding acids has not yet been demonstrated,
probably as a result of the instability of a-fluoroaldehydes.[8]
Furthermore, chiral a-fluoroketones and 2-fluoroalcohols can
be synthesized in principle from 5 and 6 according to the
procedure of Davis and co-workers.[10] Thus, we expect the
present methodology to be useful in providing a variety of
optically active fluorine-containing compounds.
In summary, we have developed the first catalytic
asymmetric fluorination reaction for ester equivalents. In
the presence of a NiCl2–binap/R3SiOTf/2,6-lutidine triad, the
monofluorinated compounds were formed with high enantio-
selectivity (88% ee) and up to 99% yield. Furthermore, the
conversion of the fluorinated product into the corresponding
carboxylic acid was achieved. Although the exact role of
R3SiOTf remains to be elucidated, we believe that the present
three-component activation system has the potential for
application to novel reactions. Further investigations into the
substrate scope and reaction mechanism are in progress.
Scheme 3. Proposed catalytic cycle.
treated with excess R3SiOTf.[23] The cationic complex 1a
formed in situ and 2,6-lutidine act cooperatively to give the
chiral Ni enolate. This Ni enolate with the Z configuration
reacts with NFSI, and the dissociation of the product from the
Ni complex completes the catalytic cycle. H and 19F NMR
1
spectroscopic studies showed an interaction between
Et3SiOTf and NFSI. Although the exact role of R3SiOTf is
unclear, it is likely that it enhances the electrophilicity of
NFSI. Alternatively, R3SiOTf might reduce the acidity of the
substrate. However, a control experiment with the silyl
enolate preformed in situ indicated that the silyl enolate
was not involved in the reaction.[24] It appears that the
formation of the Ni enolate is more favorable than the
formation of the achiral silyl enolate. The putative role(s) of
the supplementary Lewis acid in our system should be distinct
from the action of Me3SiOTf in the Evans aldol reaction, in
which it is proposed to trap the nickel aldolate intermedi-
ate.[14]
Compound 5a was converted into the Weinreb amide 6
according to a literature procedure (Scheme 4).[10b] The
absolute configuration of 5a was determined to be R by
Angew. Chem. Int. Ed. 2007, 46, 5435 –5439
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5437

Communications
Chem. 2005, 117, 3772; Angew. Chem. Int. Ed. 2005, 44, 3706;
d) T. D. Beeson, D. W. C. MacMillan, J. Am. Chem. Soc. 2005,
127, 8826; e) Y. Huang, A. M. Walji, C. H. Larsen, D. W. C.
MacMillan, J. Am. Chem. Soc. 2005, 127, 15051.
Experimental Section
Representative procedure: Compound 4a (22 mg, 0.1 mmol), the
nickel complex 1b (3.8 mg, 0.005 mmol, 5 mol%), and NFSI (47.4 mg,
0.15 mmol) were placed in a dry reaction tube. Toluene (0.1 mL) and
triethylsilyl triflate (17 mL, 0.75 equiv) were added at À208C under an
atmosphere of dry nitrogen. After 10 min, 2,6-lutidine (18 mL,
0.15 mmol) was added, and the resulting mixture was stirred at
À208C for 24h. Saturated aqueous NaCl was then added to quench
the reaction, and the aqueous layer was extracted with ethyl acetate
(3 ” 5 mL). The combined organic layers were washed with brine and
dried over Na2SO4, the solvent was evaporated, and the crude product
was purified by flash column chromatography (SiO2, hexane/chloro-
form/ethyl acetate 5:1:1) to afford 5a as a white solid. The ee value of
the product was determined by HPLC on a chiral phase. [a]3D1 = À92.1
(c = 1.0, CHCl3; 82% ee); HPLC (Daicel Chiralcel AD-H, n-hexane/
isopropyl alcohol 9:1, 1.0 mLminÀ1, 254nm): tR(major) = 14.3 min,
tR(minor) = 16.5 min; IR (neat): n˜ = 2953, 1687, 1358, 1286, 1240,
[9] For fluorocitric acid, see: a) D. B. Harper, D. OꢀHagan, Nat.
Prod. Rep. 1994, 11, 123; for fluoroglutamic acids, see: b) B. P.
Hart, W. H. Haile, N. J. Licato, W. E. Bolanowska, J. J. McGuire,
J. K. Coward, J. Med. Chem. 1996, 39, 56.
[10] a) F. A. Davis, W. Han, Tetrahedron Lett. 1992, 33, 1153; b) F. A.
Davis, P. V. N. Kasu, Tetrahedron Lett. 1998, 39, 6135.
[11] For enantioselective reactions in which a stoichiometric amount
of a chiral fluorinating reagent is used, see: a) E. Differding,
R. W. Lang, Tetrahedron Lett. 1988, 29, 6087; b) F. A. Davis, P.
Zhou, C. K. Murphy, G. Sundarababu, H. Qi, W. Han, R. M.
Przeslawski, B.-C. Chen, P. J. Carroll, J. Org. Chem. 1998, 63,
2273.
[12] For a short review, see: Y. Hamashima, M. Sodeoka, Synlett
2006, 1467; see also reference [7a].
1178, 1154, 1067, 1016 cmÀ1; H NMR (400 MHz, CDCl3): d = 3.25–
1
3.35 (m, 2H), 4.12–4.27 (m, 2H), 6.81 (d, J = 48.5 Hz, 1H), 7.37–7.41
(m, 3H), 7.51–7.53 ppm (m, 2H); 13C NMR (100 MHz, CDCl3): d =
25.6, 46.7, 89.6 (d, J = 178.6 Hz), 128.5 (d, J = 5.0 Hz), 128.8 (d, J =
1.6 Hz), 130.0 (d, J = 3.2 Hz), 133.5 (d, J = 19.8 Hz), 168.1 (d, J =
27.2 Hz), 172.5 ppm; 19F NMR (376 Hz, CDCl3, CF3COOH): d =
À95.2 ppm (d, J = 48.5 Hz); FABMS (m-nitrobenzylalcohol):
m/z 262 [M+Na]+; HRMS: m/z calcd for C11H10FNNaO2S: 262.0314
[M+Na]+; found: 262.0311.
[13] a) T. Izawa, T. Mukaiyama, Bull. Chem. Soc. Jpn. 1979, 52, 555;
b) M. T. Crimmins, B. W. King, E. A. Tabet, K. J. Chaudhary, J.
Org. Chem. 2001, 66, 892; c) D. A. Evans, J. Bartroli, T. L. Shih,
J. Am. Chem. Soc. 1981, 103, 2127.
[14] D. A. Evans, C. W. Downey, J. L. Hubbs, J. Am. Chem. Soc. 2003,
125, 8706.
[15] For the concurrent use of a Lewis acid and an organic base in
catalysis, see: a) K. Itoh, S. Kanemasa, J. Am. Chem. Soc. 2002,
124, 13394; b) D. M. Barnes, J. Ji, M. G. Fickes, M. A. Fitzgerald,
S. A. King, H. E. Morton, F. A. Plagge, M. Preskill, S. H. Wagaw,
S. J. Wittenberger, J. Zhang, J. Am. Chem. Soc. 2002, 124, 13097;
c) D. A. Evans, J. S. Tedrow, J. T. Shaw, C. W. Downey, J. Am.
Chem. Soc. 2002, 124, 392; d) N. Kumagai, S. Matsunaga, M.
Shibasaki, J. Am. Chem. Soc. 2004, 126, 13632; we also reported
a similar combination in asymmetric fluorination reactions: e) T.
Suzuki, T. Goto, Y. Hamashima, M. Sodeoka, J. Org. Chem.
2007, 72, 246; f) K. Moriya, Y. Hamashima, M. Sodeoka, Synlett
2007, 1139.
[16] The simultaneous activation of electrophiles, such as enones,
imines, and N,O acetals, by protonation was important for their
reaction with chiral palladium enolates: a) Y. Hamashima, D.
Hotta, M. Sodeoka, J. Am. Chem. Soc. 2002, 124, 11240; b) Y.
Hamashima, N. Sasamoto, D. Hotta, H. Somei, N. Umebayashi,
M. Sodeoka, Angew. Chem. 2005, 117, 1549; Angew. Chem. Int.
Ed. 2005, 44, 1525; c) N. Sasamoto, C. Dubs, Y. Hamashima, M.
Sodeoka, J. Am. Chem. Soc. 2006, 128, 14010.
Received: March 12, 2007
Published online: June 14, 2007
Keywords: asymmetric catalysis · carboxylic acid derivatives ·
.
fluorination · Lewis acids · nickel
[1] P. Kirsch, Modern Fluoroorganic Chemistry: Synthesis, Reac-
tivity, Applications, Wiley-VCH, Weinheim, 2004.
[2] K. Mikami, Y. Itoh, M. Yamanaka, Chem. Rev. 2004, 104, 1.
[3] For reviews of (catalytic) enantioselective fluorination reactions,
see: a) C. Bobbio, V. Gouverneur, Org. Biomol. Chem. 2006, 4,
2065; b) H. Ibrahim, A. Togni, Chem. Commun. 2004, 1147; c) J.-
A. Ma, D. Cahard, Chem. Rev. 2004, 104, 6119.
[4] a) L. Hintermann, A. Togni, Angew. Chem. 2000, 112, 4530;
Angew. Chem. Int. Ed. 2000, 39, 4359; b) D. Y. Kim, E. J. Park,
Org. Lett. 2002, 4, 545; c) Y. Hamashima, K. Yagi, H. Takano, L.
Tamµs, M. Sodeoka, J. Am. Chem. Soc. 2002, 124, 14 530; d) Y.
Hamashima, T. Suzuki, H. Takano, Y. Shimura, Y. Tsuchiya, K.
Moriya, T. Goto, M. Sodeoka, Tetrahedron 2006, 62, 7168; e) J.-
A. Ma, D. Cahard, Tetrahedron: Asymmetry 2004, 15, 1007; f) N.
Shibata, T. Ishimaru, T. Nagai, J. Kohno, T. Toru, Synlett 2004,
1703; g) S. Suzuki, H. Furuno, Y. Yokoyama, J. Inanaga,
Tetrahedron: Asymmetry 2006, 17, 504.
[5] a) Y. Hamashima, T. Suzuki, Y. Shimura, T. Shimizu, N.
Umebayashi, T. Tamura, N. Sasamoto, M. Sodeoka, Tetrahedron
Lett. 2005, 46, 1447; b) L. Bernardi, K. A. Jørgensen, Chem.
Commun. 2005, 1324; c) S. M. Kim, H. R. Kim, D. Y. Kim, Org.
Lett. 2005, 7, 2309.
[6] H. R. Kim, D. Y. Kim, Tetrahedron Lett. 2005, 46, 3115.
[7] a) Y. Hamashima, T. Suzuki, H. Takano, Y. Shimura, M.
Sodeoka, J. Am. Chem. Soc. 2005, 127, 10164; b) N. Shibata, J.
Kohno, K. Takai, T. Ishimaru, S. Nakamura, T. Toru, S.
Kanemasa, Angew. Chem. 2005, 117, 4276; Angew. Chem. Int.
Ed. 2005, 44, 4204.
[17] D. A. Evans, R. J. Thomson, J. Am. Chem. Soc. 2005, 127, 10506;
BF3·OEt2 (3 equiv) was used as an external activator to form an
oxonium intermediate from methyl orthoformate.
[18] Metal–binap complexes of PdII, PtII, and AgI and metal–
bisoxazoline complexes of CuII, ZnII, NiII, PdII, and PtII gave
almost racemic products under similar conditions.
[19] The reaction with Bu2BOTf as a supplementary Lewis acid
proceeded with no enantioselectivity, and the reaction with
BF3·OEt2 did not proceed at all.
[20] The nickel complexes used in these studies were prepared
according to the procedure described by Evans and Thomson.[17]
[21] When 0.5 equivalents of 2,6-lutidine were used, the reaction did
not proceed to completion: The product was formed in 42%
yield with 43% ee when the reaction was carried out in CH2Cl2 at
room temperature.
[22] The use of other bases, including Et3N, iPr2NEt, pyridine, and
2,6-di(tert-butyl)pyridine, led to less satisfactory results.
[23] Details of these experiments will be discussed in a full paper.
[24] The treatment of 4a with Et3SiOTf and 2,6-lutidine at room
temperature gave the corresponding silyl enolate in 78% yield
with 22% unchanged 4a. When this reaction mixture was
subjected to the fluorination conditions at À208C for 24h, 5a
was obtained in 29% yield with 64% ee. The enantioselective
[8] a) D. Enders, M. R. M. Hüttl, Synlett 2005, 991; b) M. Marigo, D.
Fielenbach, A. Braunton, A. Kjærsgaard, K. A. Jørgensen,
Angew. Chem. 2005, 117, 3769; Angew. Chem. Int. Ed. 2005,
44, 3703; c) D. D. Steiner, N. Mase, C. F. Barbas III, Angew.
5438
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 5435 –5439

Angewandte
Chemie
fluorination of the remaining 4a and the uncatalyzed reaction of
a small amount of the silyl enolate account for the observed low
chemical yield and moderate enantioselectivity. If the formation
of the silyl enolate and transmetalation are involved, a higher
chemical yield and a comparable ee value should be observed.
[25] In contrast, when chiral 2-acyl oxazolidin-2-ones were used as
substrates, the optical purity of the fluorinated compounds
decreased considerably during hydrolysis.[10a]
Angew. Chem. Int. Ed. 2007, 46, 5435 –5439
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5439