Synthesis of α-fluoro-β-hydroxy esters by an enantioselective Reformatsky-type reaction

Article abstract of DOI:10.1039/c2cc17985g

The first enantioselective Reformatsky-type reaction of ethyl iodofluoroacetate has been accomplished with alkyl aryl ketones. High diastereoselectivities and excellent enantioselectivities for the major diastereomer (93-95% ee) were achieved with large a

Full text of DOI:10.1039/c2cc17985g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 3500–3502
www.rsc.org/chemcomm
COMMUNICATION
Synthesis of a-fluoro-b-hydroxy esters by an enantioselective
Reformatsky-type reactionw
Michal Fornalczyk, Kuldip Singh and Alison M. Stuart*
Received 21st December 2011, Accepted 2nd February 2012
DOI: 10.1039/c2cc17985g
The first enantioselective Reformatsky-type reaction of ethyl
iodofluoroacetate has been accomplished with alkyl aryl ketones.
High diastereoselectivities and excellent enantioselectivities for
the major diastereomer (93–95% ee) were achieved with large
alkyl groups. For smaller alkyl groups the diastereoselectivities
were moderate, but excellent enantioselectivities were obtained
for both diastereomers (79–94% ee).
In stark contrast to these reports on the enantioselective
Reformatsky reaction of a-halogenated esters, there are only three
reports of an enantioselective reaction with the difluorinated
Reformatsky reagent5a,9 and no reports of an asymmetric
reaction with the monofluorinated Reformatsky reagent. The
Reformatsky reaction of ethyl bromofluoroacetate is one of
the oldest and most efficient methods for preparing a-fluoro-
b-hydroxy esters.10 Presumably, an enantioselective version
has not been developed because the reaction only proceeds in
the presence of both the zinc dust and the carbonyl substrate
in a one-step procedure whereas a two-step zinc insertion
procedure is normally used for the enantioselective Reformatsky
reaction with ethyl bromodifluoroacetate. In this paper we will
report the first example of an enantioselective Reformatsky-type
reaction of ethyl iodofluoroacetate with ketones using diethylzinc
to generate the Reformatsky reagent homogeneously and (1R,2S)-
1-phenyl-2-(pyrrolidin-1-yl)propan-1-ol as the chiral ligand.11
Previously, the Reformatsky reaction of ethyl bromofluoro-
acetate with carbonyl substrates had only been reported using
zinc dust under heterogeneous conditions.10 We now show
that the Reformatsky reaction of ethyl iodofluoroacetate and
ethyl bromofluoroacetate can also be performed under homo-
geneous conditions (Table 1, entries 1–3). The reaction with
ethyl iodofluoroacetate (entry 1) gave 100% conversion to
The incorporation of fluorine into the organic framework is an
important strategy for the development of new pharmaceutical
products because of the beneficial properties imparted by
fluorine.1 Recently, there has been a surge of interest in the
field of asymmetric fluorination because of the increasing
demand for enantiopure pharmaceutical products.2 We were
interested in designing an enantioselective synthesis of a-fluoro-
b-hydroxy esters which are important building blocks for the
preparation of monofluorinated analogues of amino acids and
carbohydrates.3
The classical Reformatsky reaction provides a convenient
synthesis of b-hydroxy esters by the nucleophilic addition of
the Reformatsky reagent, prepared by the direct insertion of
zinc into a-halogenated esters, to carbonyl compounds.4 It is
an excellent alternative to the base-catalysed aldol reaction
because of the high tolerance towards functional groups and
the mild reaction conditions. Until recently, stoichiometric
amounts of chiral ligands were required to promote the
enantioselective Reformatsky reaction because of the hetero-
geneous reaction conditions.5 Following the development of
homogeneous Reformatsky-type reactions,6,7 Cozzi reported
the first catalytic enantioselective Reformatsky reaction in 2006
using 20 mol% of a chiral manganese salen catalyst and
dimethylzinc to generate the zinc reagent homogeneously.8h
Since then, BINOL derivatives, chiral aminoalcohols and a
chiral bisoxazolidine have been used to promote the catalytic
enantioselective Reformatsky reactions of a-halogenated esters
with aldehydes and ketones.8
ethyl 2-fluoro-3-hydroxy-3-phenylbutanoate
1 which was
obtained as a 63 : 37 mixture of the (2S,3S)/(2R,3R)- and
(2R,3S)/(2S,3R)-diastereomers. The solid-state structure of
the major diastereomer confirmed that it was a mixture of
the (2S,3S)- and (2R,3R)-enantiomers which is in agreement
with the literature.10a The bromine–zinc exchange reaction
between ethyl bromofluoroacetate and diethylzinc was less efficient
than the iodine–zinc exchange reaction resulting in only a 29%
conversion to ethyl 2-fluoro-3-hydroxy-3-phenylbutanoate 1
(entry 2). The conversion could be improved substantially to
76% (entry 3) when the reaction with ethyl bromofluoroacetate
was catalysed by 1 mol% of [RhCl(PPh3)3].
The scope of the reaction with ethyl iodofluoroacetate was
established with a series of ketones and the a-fluoro-b-hydroxy
esters were isolated in good yields (Table 1, entries 4–11).
The (2S,3S)/(2R,3R)- and (2R,3S)/(2S,3R)-diastereomers of
products 2, 4, 5 and 7 were separated by column chromatography.
In each case the major diastereomer was confirmed as the (2S,3S)/
(2R,3R)-diastereomer by X-ray crystallography and the molecular
structures are shown in the ESI. The (2S,3S)/(2R,3R)- and
Department of Chemistry, University of Leicester, Leicester,
LE1 7RH, UK. E-mail: Alison.Stuart@le.ac.uk;
Fax: +44 (0)116 2523789; Tel: +44 (0)116 2522136
w Electronic supplementary information (ESI) available: Table 4,
experimental procedures, NMR spectra and crystallographic data
for 2a, 4a, 5a, 7a, 10b and 10d. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c2cc17985g
c
3500 Chem. Commun., 2012, 48, 3500–3502
This journal is The Royal Society of Chemistry 2012

View Article Online
Table 1 Reformatsky-type reaction with ethyl iodofluoroacetate
Table 2 Optimisation of the enantioselective Reformatsky-type reaction
with ethyl iodofluoroacetate
SS/RR :
Conva RS/SR
Product (%)
Entry Ar
R
ratiob
Yieldc (%)
1
Ph
Ph
Ph
Ph
Ph
Ph
Me
Me
Me
Et
Pr
iso-Bu
1
1
1
2
3
4
5
6
7
8
9
100
29
76
63d : 37 98
SS/RR :
Entry ICHFCO2Et Et2Zn Temp. Yielda,b RS/SR eed (%) eed (%)
2e
3e,f
4
61d : 39
60d : 40
—
—
(equiv)
(equiv) (1C) (%)
ratioc (SS/RR) (RS/SR)
100
100
97
100
100
100
100
100
55d : 45 40 : 40
1e
2
3
4
5
6
1.5
1.5
2.0
2.0
2.0
2.0
3.0
2.0
2.5
2.5
2.5
2.5
2.5
3.5
3.5
3.5
0
0
0
100 (98) 63 : 37 NA
84 (74) 79 : 21 80
100 (58) 82 : 18 73
NA
46
51
60
72
82
77
NA
5
6
7
8
9
10
11
59 : 41
57 : 20
59d : 41 46 : 22
63d : 37 58 : 20
2-MeOC6H4 Me
4-MeOC6H4 Me
4-ClC6H4
1-indanone
1-tetralone
61 : 39
98 (60 : 25)
À20 84 (76) 75 : 25 87
À40 66 (66) 75 : 25 90
À40 99 (97) 74 : 26 87
À40 100 (94) 75 : 25 84
Me
56d : 44 5 : (45) : 20g
46 : 54
60 : 40
84
96
7
8e
À40 100
60 : 40 NA
a
b
Determined by 1H NMR spectroscopy. Determined by 19F NMR
a
1
Determined by H NMR spectroscopy. Isolated yield in parenthesis.
b
c
spectroscopy for crude product. Isolated yield. Confirmed as (SS/RR)-
e
diastereomer by X-ray crystallography. BrCHFCO2Et was used instead
d
c
d
Determined by 19F NMR spectroscopy for crude product. Determined
e
by chiral HPLC. No chiral aminoalcohol.
f
g
of ICHFCO2Et. Incorporation of [RhCl(PPh3)3] (1 mol%). 5% of
(SS/RR)-diastereomer, 20% of (RS/SR)-diastereomer, 45% mixture
of both diastereomers.
On increasing the amount of ethyl iodofluoroacetate to
2.0 equivalents in entry 3, the conversion increased to 100%
and the enantiomeric excess for the minor diastereomer increased
to 51% ee, but the enantioselectivity for the major diastereomer
decreased to 73% ee. As the reaction temperature was lowered
from 0 to À40 1C (entries 3–5), the conversion decreased to 66%
and the diasteromeric ratio dropped slightly to 75 : 25 whilst the
enantioselectivity increased for both diastereomers giving 90% ee
and 72% ee for the major and minor diastereomers respectively.
In an attempt to enhance the conversion 3.5 equivalents of
diethylzinc were used in entry 6. The conversion was increased
successfully to 99% and the enantiomeric excess for the minor
diastereomer increased to 82% ee, but the enantiomeric excess
for the major diastereomer dropped slightly (87% ee). In entry 7
the amount of ethyl iodofluoroacetate was increased to
3.0 equivalents and the high conversion (100%) was maintained
but the enantiomeric excess for both diastereomers decreased.
Therefore, entry 6 gave the best result providing an excellent
isolated yield (97%) with excellent enantiomeric excess for the
major diastereomer (87% ee) and a good enantiomeric excess
for the minor diastereomer (82% ee). Finally, the uncatalysed
reaction under identical conditions in entry 8 gave 100%
conversion with a lower 60 : 40 diastereomeric ratio demonstrating
that it would not be possible to catalyse the enantioselective
reaction with a catalytic amount of the chiral aminoalcohol.
With the reaction conditions optimised, the scope of the
enantioselective Reformatsky-type reaction with ethyl iodo-
fluoroacetate was evaluated with a series of alkyl aryl ketones
(Table 3). In contrast to the achiral reaction (Table 1, entries 1, 4–6),
when the aliphatic side of the ketone was extended to an ethyl,
propyl and iso-butyl group (Table 3, entries 1–4), the diastereo-
meric ratio increased stepwise from 74 : 26 for acetophenone
to an excellent 92 : 8 for 3-methyl-1-phenylbutan-1-one. In
each reaction excellent enantioselectivity was achieved for
the major (2S,3S)/(2R,3R)-diastereomer (93–95% ee) with
(2R,3S)/(2S,3R)-diastereomers are easy to distinguish by
19F NMR spectroscopy because the chemical shift for the
(2S,3S)/(2R,3R)-diastereomer is always at a lower field than
that exhibited by the (2R,3S)/(2S,3R)-diastereomer (Table 4,
ESIw). This data was used to assign the diastereomers of products
3, 6, 8 and 9. With the exception of compounds 5 and 8, it was also
observed that the (2S,3S)/(2R,3R)-diastereomers had a typical
one bond carbon–fluorine coupling constant of 191–194 Hz,
whilst the one bond carbon–fluorine coupling constants for the
(2R,3S)/(2S,3R)-diastereomers tended to be larger (196–201 Hz).
Only low levels of diastereoselectivity were observed in all of the
reactions and a similar lack of diastereoselectivity had also been
reported for the heterogeneous Reformatsky reaction with ethyl
bromofluoroacetate and ethyl chlorofluoroacetate.10
In Table 2 acetophenone was used as the model substrate
for optimising the enantioselective Reformatsky-type reaction
of ethyl iodofluoroacetate in the presence of (1R,2S)-1-phenyl-
2-(pyrrolidin-1-yl)propan-1-ol as the chiral ligand.8c,d In entry 1
the uncatalysed reaction was repeated using 2.5 equivalents of
diethylzinc and the results were identical to the uncatalysed
reaction using a stoichiometric amount of diethylzinc in
Table 1 (entry 1). In the enantioselective reactions (entries 2–7)
an excess of diethylzinc was always used to deprotonate the
chiral aminoalcohol as well as to form the Reformatsky
reagent homogeneously and there was no trace of the product
resulting from the addition of diethylzinc to acetophenone in
any of the reactions. When 1 equivalent of (1R,2S)-1-phenyl-2-
(pyrrolidin-1-yl)propan-1-ol was used to catalyse the reaction in
entry 2, the conversion dropped to 84% and the diastereomeric
ratio increased from 63 : 37 to 79: 21. A good enantiomeric
excess (80% ee) was obtained for the major (2S,3S)/(2R,3R)-
diastereomer whilst a much lower enantiomeric excess (46% ee)
was observed for the minor (2R,3S)/(2S,3R)-diastereomer.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 3500–3502 3501

View Article Online
Table 3 Scope of the enantioselective Reformatsky-type reaction
with ethyl iodofluoroacetate
moderate diastereoselectivities and excellent enantioselectivities
for both diastereomers (79–94% ee) were obtained.
We thank the Royal Society (AMS) for financial support.
Notes and references
1 (a) D. O’Hagan, J. Fluorine Chem., 2010, 131, 1071; (b) S. Purser,
P. R. Moore, S. Swallow and V. Gouverneur, Chem. Soc. Rev.,
2008, 37, 320.
2 (a) D. Cahard, X. Xu, S. Couve-Bonnaire and X. Pannecoucke,
Chem. Soc. Rev., 2010, 39, 558; (b) S. Lectard, Y. Hamashima and
M. Sodeoka, Adv. Synth. Catal., 2010, 352, 2708; (c) Y. K. Kang
and D. Y. Kim, Curr. Org. Chem., 2010, 14, 917; (d) V. A. Bruce
and D. O’Hagan, Angew. Chem., Int. Ed., 2008, 47, 1179;
(e) J.-A. Ma and D. Cahard, Chem. Rev., 2008, 108, PR1–PR43;
(f) N. Shibata, S. Mizuta and H. Kawai, Tetrahedron: Asymmetry,
2008, 19, 2633.
SS/RR :
RS/SR eed,e (%) eed,e (%)
ratioc (SS/RR) (RS/SR)
Yielda,b
(%)
Entry Ar
R
1
2
3
4
5
6
7
8
9
Ph
Ph
Ph
Ph
Me
Et
Pr
99 (97)
74 : 26 87 (3S) 82 (3S)
100 (71 : 13) 87 : 13 95
100 (95) 89 : 11 93
92 : 8 94
100 (51 : 34) 58 : 42 84
56
65
61
85
81
79
80
94
3 G. Haufe, J. Fluorine Chem., 2004, 125, 875.
4 (a) P. G. Cozzi, A. Mignogna and L. Zoli, Pure Appl. Chem., 2008,
80, 891; (b) P. G. Cozzi, Angew. Chem., Int. Ed., 2007, 46, 2568;
(c) R. Ocampo and W. R. Dolbier, Jr., Tetrahedron, 2004, 60, 9325;
iso-Bu 96 (94)
2-MeOC6H4 Me
4-MeOC6H4 Me
4-ClC6H4
1-indanone
1-tetralone
100 (99)
100 (96)
98 (97)
81 : 19 93
71 : 29 88
38 : 62 86
40 : 60 88
(d) A. Furstner, Synthesis, 1989, 571.
¨
Me
5 (a) R. J. Kloetzing, T. Thaler and P. Knochel, Org. Lett.,
2006, 8, 1125; (b) Y. Fujiwara, T. Katagiri and K. Uneyama,
Tetrahedron Lett., 2003, 44, 6161; (c) A. Ojida, T. Yamano,
N. Taya and A. Tasaka, Org. Lett., 2002, 4, 3051; (d) J. M.
100 (84)
a
1
Determined by H NMR spectroscopy. Isolated yield in parenthesis.
b
c
d
Determined by 19F NMR spectroscopy for crude product. Determined
Andres, Y. Martın, R. Pedrosa and A. Perez-Encabo, Tetrahedron,
´ ´ ´
e
1997, 53, 3787; (e) A. Mi, Z. Wang, Z. Chen, Y. Jiang, A. S. C.
Chan and T.-K. Yang, Tetrahedron: Asymmetry, 1995, 6, 2641;
(f) D. Pini, A. Mastantuono and P. Salvadori, Tetrahedron:
Asymmetry, 1994, 5, 1875; (g) K. Soai, A. Oshio and T. Saito,
J. Chem. Soc., Chem. Commun., 1993, 811; (h) K. Soai and
Y. Kawase, Tetrahedron: Asymmetry, 1991, 2, 781.
by chiral HPLC. Esters 1a and 1b were reacted with the lithium salt of
(S)-(1-phenylethyl)amine and absolute configuration determined by X-ray
crystallographic analysis; stereochemistry of the other product esters was
tentatively assumed by analogy.
6 (a) K. Sato, A. Tarui, T. Kita, Y. Ishida, H. Tamura, M. Omote,
A. Ando and I. Kumadaki, Tetrahedron Lett., 2004, 45, 5735;
(b) K. Kanai, H. Wakabayashi and T. Honda, Org. Lett., 2000,
2, 2549; (c) T. Honda, H. Wakabayashi and K. Kanai, Chem.
Pharm. Bull., 2002, 50, 307; (d) J. C. Adrian, Jr. and M. L.
Snapper, J. Org. Chem., 2003, 68, 2143.
7 (a) K. Maruoka, N. Hirayama and H. Yamamoto, Polyhedron,
1990, 9, 223; (b) I. Sato, Y. Takizawa and K. Soai, Bull. Chem. Soc.
Jpn., 2000, 73, 2825; (c) Y. Ukaji, Y. Yoshida and K. Inomata,
Tetrahedron: Asymmetry, 2000, 11, 733; (d) Y. Ukaji, S. Takenaka,
Y. Horita and K. Inomata, Chem. Lett., 2001, 254.
8 (a) C. Wolf and M. Moskowitz, J. Org. Chem., 2011, 76, 6372;
(b) N. Lin, M.-M. Chen, R.-S. Luo, Y.-Q. Deng and G. Lu,
Tetrahedron: Asymmetry, 2010, 21, 2816; (c) P. G. Cozzi,
F. Benfatti, M. G. Capdevila and A. Mignogna, Chem. Commun.,
2008, 3317; (d) P. G. Cozzi, A. Mignogna and P. Vicennati,
moderate enantioselectivity for the minor (2R,3S)/(2S,3R)-
diastereomer (56–65% ee). On the other hand, when the
aromatic ring was substituted with either electron-donating
or electron-withdrawing substituents (entries 5–7), moderate
diastereoselectivities were obtained ranging from 58 : 42 to
81 : 19. In these reactions excellent enantioselectivities were
achieved for both the major (84–93% ee) and the minor
diastereomers (79–85% ee). Although the diastereomeric ratio
was reversed for the cyclic ketones, 1-indanone and 1-tetralone,
once again excellent enantiomeric excesses were obtained for
both diastereomers (80–94% ee).
The absolute configuration of the new chiral centre formed at
the 3-position for both the major (2S,3S)/(2R,3R)-diastereomer
1a and the minor (2R,3S)/(2S,3R)-diastereomer 1b (see ESIw)
was determined to be the (S)-enantiomer resulting from
nucleophilic addition to the Re face of the ketone. This facial
selectivity is the same as that observed in the asymmetric
Reformatsky reaction between ethyl bromodifluoroacetate
and aromatic aldehydes,5a,9 as well as in the catalytic enantio-
selective Reformatsky reaction of ethyl iodoacetate with
aldehydes and ketones using the same chiral ligand, (1R,2S)-
1-phenyl-2-(pyrrolidin-1-yl)propan-1-ol.8c,d
Adv. Synth. Catal., 2008, 350, 975; (e) M. A. Ferna
B. Macia, A. J. Minnaard and B. L. Feringa, Org. Lett., 2008,
10, 4041; (f) M. A. Fernandez-Ibanez, B. Macia, A. J. Minnaard
and B. L. Feringa, Chem. Commun., 2008, 2571; (g) M. A.
Fernandez-Ibanez, B. Macia, A. J. Minnaard and B. L. Feringa,
ndez-Ibanez,
´ ´
´
´
´
´
´
´
´
Angew. Chem., Int. Ed., 2008, 47, 1317; (h) P. G. Cozzi, Angew.
Chem., Int. Ed., 2006, 45, 2951.
9 (a) J. M. Andres, M. A. Martınez, R. Pedrosa and A. Perez-
´ ´ ´
Encabo, Synthesis, 1996, 1070; (b) M. Braun, A. Vonderhagen and
D. Waldmuller, Liebigs Ann., 1995, 1447.
¨
10 (a) R. Ocampo, W. R. Dolbier, K. A. Abboud and F. Zuluaga,
J. Org. Chem., 2002, 67, 72; (b) Y. Shen and M. Qi, J. Chem. Res.,
Synop., 1993, 222; (c) R. J. Linderman and D. M. Graves, J. Org.
In conclusion, we have reported the first example of
an enantioselective Reformatsky-type reaction with ethyl
iodofluoroacetate. High diastereoselectivies (up to 92 : 8) and
enantioselectivities for the major diastereomers (93–95% ee)
Chem., 1989, 54, 661; (d) S. Brandange, O. Dahlman and
¨
¨
L. Morch, J. Am. Chem. Soc., 1981, 103, 4452; (e) H. Machleidt
and R. Wessendorf, Justus Liebigs Ann. Chem., 1964, 674, 1;
(f) E. D. Bergamann, S. Cohen and R.-M. Hoffman, J. Chem.
Soc., 1961, 3452; (g) E. T. McBee, O. R. Pierce and D. L.
Christman, J. Am. Chem. Soc., 1955, 77, 1571.
i
were achieved with PhC(O)R derivatives (R = Et, Pr, Bu).
With 1-indanone, 1-tetralone and ArC(O)Me derivatives,
11 M. Fornalczyk, PhD Thesis, University of Leicester, 2011.
c
3502 Chem. Commun., 2012, 48, 3500–3502
This journal is The Royal Society of Chemistry 2012