Ligand-promoted asymmetric imino-reformatsky reaction of ethyl dibromofluoroacetate

Article abstract of DOI:10.1021/ol500631j

An enantioselective Reformatsky reaction has been developed for the reaction of ethyl dibromofluoroacetate (1) with an imine. This method represents the first ligand-promoted imino-Reformatsky approach to use a halofluoroacetate. The use of an amino alcohol ligand allowed for the preparation of enantioenriched α-bromo-α-fluoro-β-lactams in good yields with enantioselectivities up to 96% ee. This process also provided access to β-lactam rings bearing two stereogenic centers.

Full text of DOI:10.1021/ol500631j

Letter
pubs.acs.org/OrgLett
Ligand-Promoted Asymmetric Imino-Reformatsky Reaction of Ethyl
Dibromofluoroacetate
Atsushi Tarui,† Haruka Nishimura,† Takeshi Ikebata,† Asuka Tahira,† Kazuyuki Sato,† Masaaki Omote,†
Hideki Minami,‡ Yoshihisa Miwa,‡ and Akira Ando*,†
†Faculty of Pharmaceutical Sciences, Setsunan University, 45-1 Nagaotoge-cho, Hirakata, Osaka 573-0101, Japan
‡Faculty of Pharmaceutical Sciences, Hiroshima International University, 5-1-1 Hirokoshingai, Kure, Hiroshima 737-0112, Japan
S
* Supporting Information
ABSTRACT: An enantioselective Reformatsky reaction has
been developed for the reaction of ethyl dibromofluoroacetate
(1) with an imine. This method represents the first ligand-
promoted imino-Reformatsky approach to use a halofluoro-
acetate. The use of an amino alcohol ligand allowed for the
preparation of enantioenriched α-bromo-α-fluoro-β-lactams in
good yields with enantioselectivities up to 96% ee. This
process also provided access to β-lactam rings bearing two stereogenic centers.
reaction of a fluorinated haloacetate with an imine in the
presence of a chiral ligand.
he incorporation of fluorine into an organic compound
Toften leads to dramatic changes in the physical, chemical,
and biological properties of the parent compound, and fluorine
has been used in this way to alter the properties of compounds
in the pharmaceutical, agrochemical, and material fields.1
Although a broad range of methods have been reported for
the regioselective introduction of fluorine substituents,
asymmetric approaches toward highly enantioenriched fluorine
compounds remain scarce.2
Ethyl dibromofluoroacetate (1) has been used extensively as
a reagent in the Reformatsky reaction, where it has been used
to provide efficient access to a range of interesting compounds,
including α-bromo-α-fluoro-β-hydroxyesters, α-bromo-α-fluo-
ro-β-hydroxyamides, and α-bromo-α-fluoro-β-lactams 2.13−15
Furthermore, reagent 1 is useful as a research object due to the
commercial availability and chemical stability. We recently
reported the imino-Reformatsky reaction of 1 using diethylzinc
to give racemic 2 (Scheme 1, eq 1).15 The reaction featured the
highly diastereoselective addition of the Reformatsky reagent,
which was prepared from the reaction of 1 with diethylzinc, to a
series of imine substrates. We also demonstrated that 2 could
be functionalized at the α-carbon using a nickel-catalyzed cross-
The Reformatsky reaction is a robust and reliable trans-
formation for the formation of C−C bonds and has been used
extensively for the synthesis of a wide range of different
compounds,3 such as bioactive natural products.4 A variety of
enantioselective approaches have been reported during the past
decade for the Reformatsky reaction of α-halogenated esters
(nonfluorinated), including imino-Reformatsky reactions, as
well as reactions involving Schiff bases,5 BINOL derivatives,6
and amino alcohol ligands.7−9 These Reformatsky reactions
have been applied to the asymmetric synthesis of β-hydroxy-
α,α-difluoroesters using bromodifluoroacetate but provided low
levels of enantioselectivity.10 Of the different reaction
conditions available, good stereocontrol was achieved by
Knochel et al.10c in their synthesis of β-hydroxy-α,α-difluoro-
esters, although this transformation did require the addition of
a stoichiometric amount of a chiral amino alcohol ligand. More
recently, high levels of enantioselectivity have been achieved
through the reaction of ethyl iododifluoroacetate or ethyl
iodofluoroacetate with ketones in the presence of a stoichio-
metric amount of chiral ligand.11,12 Unfortunately, despite
providing access to the desired products in high yields and high
enantioselectivities, these approaches have been limited by the
poor chemical stability of the iodoacetate reagents. Compared
with carbonyl electrophiles, there have been no reports in the
literature pertaining to the asymmetric imino-Reformatsky
Scheme 1. Imino-Reformatsky Reaction of
Dibromofluoroacetate and the Application of the α-Bromo-
α-fluoro-β-lactam to Further Reactions
Received: December 24, 2013
Published: April 3, 2014
© 2014 American Chemical Society
2080
dx.doi.org/10.1021/ol500631j | Org. Lett. 2014, 16, 2080−2083

Organic Letters
Letter
coupling reaction or lithiation chemistry (Scheme 1, eqs 2 and
3).16 During the course of these experiments, we succeeded in
achieving the enantioselective imino-Reformatsky reaction of 1
using a chiral amino alcohol catalyst to provide optically active
2 in good yield. Although there are already known reactions
with a carbonyl substrate, this is the first example of the
asymmetric reaction with an imine substrate. In this paper, we
report the first enantioselective imino-Reformatsky reaction of
1 with a variety of different imines.
Having optimized the reaction solvent, temperature, and
catalyst loading, we investigated a variety of different chiral
ligands, as shown in Figure 1. The substitution pattern on the
The imino-Reformatsky reaction was initially carried out in
the presence of a stoichiometric amount of (1R,2S)-1-phenyl-2-
(1-pyrrolidinyl)-propan-1-ol (3a) to determine whether this
material would affect the asymmetric induction of the reaction
(Table 1, entry 1). The reaction proceeded smoothly to afford
Table 1. Initial Investigation of Stereoselective Imino-
Reformatsky Reaction of 1
Figure 1. Screening of chiral amino alcohol ligand 3.
nitrogen atom affected the chemical yield of products in the
cases of ligand 3b−d. These ligands led to significant
improvements in the enantioselectivity of imino-Reformatsky
reaction with 84, 87, and 80% ee, respectively, although the
chemical yields of these reactions were lower than that achieved
with 3a. The prolinol-based chiral ligands 3e and 3f also led to
lower yield of the desired product, with 3f providing a
particularly low enantioselectivity. The similar tendency was
observed in other amino alcohol ligands 3g and 3h.
Introduction of a phenyl group next to the hydroxyl group
was effective for the stereoselectivity of this reaction. The low
yield of aziridine was obtained as a racemate in the screening of
ligands except for 3a.19 Based on these results, most effective
and easy available 3a was selected as the optimal chiral ligand
for the transformation in dichloromethane at −40 °C.
ligand 3
temp
yield
a
ee
b
entry
condition
(equiv)
solvent
(°C)
(%)
(%)
1
2
3
4
5
Et2Zn
Et2Zn
Et2Zn
1.0
THF
0
−40
−40
−40
−40
78
82
83
0
15
91
92
1.0
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
0.75
0.75
0.75
c
Me2Zn under air
Me2Zn with 8 mol %
of NiCl2(PPh3)
trace
6
Me2Zn with
10 mol % of NiBr2
0.75
CH2Cl2
−40
0
a
b
c
Isolated yields. Determined by HPLC. Air was introduced in the
reaction mixture through a CaCl2 drying tube after the addition of
Me2Zn.
With the optimal reaction conditions in hand, we explored
the scope and limitations of this stereoselective imino-
Reformatsky reaction using a variety of different imines. The
results of these scoping experiments are summarized in Table 2.
Although the chemical yields of the reactions were dependent
on the electronic nature of the substituents on the aromatic
ring, the enantiomeric purities of the products were high with
the (3S,4R)-isomers being obtained enantioselectively in all
cases. The reaction of aromatic imines bearing an electron-
withdrawing group gave compounds 2d−i with high enantio-
selectivity (Table 2, entries 4−9). The substitution pattern of
the aromatic compounds did not have a discernible impact on
the outcome of the reaction, with ortho-, meta-, and para-
substituted imines providing similar levels of enantioselectivity.
A relatively low yield of the product was obtained, however, in
the case of ortho-substituted imine, likely because of steric
hindrance from the ortho substituent (Table 2, entry 7).
Electron-donating substituents such as Me and MeO had an
adverse impact on the reaction, with the corresponding
products 2j and 2k being formed in moderate yields (Table
2, entries 10 and 11). The poor performance of these substrates
was attributed to the weakening of the electrophilicity of the
imines as a consequence of the electron-donating groups. It is
noteworthy that the ee values of the products could be
increased to be almost enantiopure form by the recrystallization
of the solid products. We also conducted the reaction on a
larger scale, using up to 1.95 g of the imine to afford 2a in 71%
yield and 92% ee after recrystallization (Table 2, entry 12). The
opposite enantiomer of the ligand could also be used under the
the desired product 2a in 78% yield and 15% ee. Encouraged by
this result, we screened a variety of different reaction conditions
to determine the optimal conditions for the transformation.17
Solvent optimization experiments revealed that dichloro-
methane was the best solvent for this particular transformation.
Lowering the reaction temperature to −40 °C led to
considerable improvements in both the ee value and the
chemical yield (Table 1, entry 2). Furthermore, decreasing the
ligand loading of 3a to 0.75 equiv led to a small improvement in
the asymmetric induction (Table 1, entry 3). Moreover,
another promoter was investigated toward the true catalytic
system. In the reported catalytic asymmetric Reformatsky
reactions, the combination of dimethyl zinc with air6,7d,g,8b or
dimethyl zinc with Ni salts7f,8a has been used as a promoter.
When dimethyl zinc, which is less reactive than diethyl
zinc,5a,b,6a,7d and air were used as promoters, the desired
product was not obtained and 1 was recovered from the
reaction mixture (Table 1, entry 4). Ni salts were also not
effective, recovering 1 as well as entry 4 (Table 1, entries 5 and
6). These reagents were not able to promote this reaction. In all
cases, the product was obtained diastereoselectively as a single
diastereomer with the 3S,4R configuration. This result was
significantly different from those reported for the reaction of
ethyl iodofluoroacetate with ketones, which showed no
diastereoselectivity.12 In contrast to our previous reports, no
aziridine byproduct was obtained during the screening of
reaction conditions.15,18
2081
dx.doi.org/10.1021/ol500631j | Org. Lett. 2014, 16, 2080−2083

Organic Letters
Letter
Table 2. Scope and Limitations of the Enantioselective
Imino-Reformatsky Reaction of 1
Scheme 2. Proposed Stereochemical Model
time
yield
a
ee
b
entry
R1
Bn
R2
product (h)
(%)
(%)
c
1
2
3
4
Ph
Ph
2a
2b
2c
2d
1
2
3
2
83
78
79
76
92
(94)
d
PMB
Bn
91
93
96
2-naphthyl
c
Bn
4-CF3-
C6H4
(>99)
c
5
6
7
8
Bn
Bn
Bn
Bn
4-Cl-C6H4
3-Cl-C6H4
2-Cl-C6H4
2e
2f
2
2
82
90
55
90
93
95
91
96
(98)
c
c
2g
2h
20
2
(>99)
(>99)
4-MeOCO-
C6H4
any reduction in their enantiopurities (Scheme 3). Diastereo-
selective insertion of the aryl groups was also observed in all of
the other cases tested, with the (3S,4R)-isomers being formed
as sole products in excellent yields.
c
c
9
Bn
4-CN-
C6H4
2i
2
89
91
(>99)
(>99)
10
Bn
Bn
4-Me-C6H4
2j
14
6
64
53
91
85
e
11
4-MeO-
C6H4
2k
Scheme 3. Synthetic Utility of Product 2a
f
f
12
Bn
Bn
Ph
2a
2a
4
1
71
92
g
13
Ph
b
70
−90
a
c
Isolated yields. Determined by HPLC. The values in parentheses
indicate the ee after recrystallization of the products. p-Methox-
ybenzyl. Ligand 3a (1.0 equiv) was used. The yield and ee were
obtained after a recrystallization step. (1S,2R)-Ligand was used
d
e
f
g
instead of (1R,2S)-ligand.
optimized conditions to give the (3R,4S) product 2a in 70%
yield and 90% ee (Table 2, entry 13). Unfortunately, the
reaction of 1 with aliphatic imine gave the complex mixture,
and the desired product was not detected from the crude
reaction mixture in the 19F NMR spectrum. The absolute
configuration of 2h was determined by X-ray analysis to be the
3S,4R configuration. The stereochemistries of the other
products were assumed to be the same based on a comparison
of the order of their elution by chiral HPLC analysis, the sign of
their optical rotation, and their NMR spectra.
Although the precise mechanism of this reaction remains
unclear, we have proposed a possible catalytic cycle based on
the absolute configuration products 2. The stereoselectivity of
this Reformatsky reaction can be explained using Noyori’s
representative transition model (Scheme 2).20 The transition
state was also applied to the reaction of alkylzinc and aldehyde
using N,N-dialkylnorephedrines as chiral catalyst.21 Transition
model A could be used to explain the outcome of the reaction
based on the formation of a coordination network involving the
imine nitrogen, zinc ion, ligand, and Reformatsky reagent,
which would allow for the successful formation of multiple
chiral inductions. Transition model A would therefore provide
the correct product with the desired stereochemistry, which
would be the opposite to that formed from ethyl iododifluoro-
acetate11 and ethyl iodoacetate.6,7a−e
In summary, we have developed an enantioselective imino-
Reformatsky reaction of ethyl dibromofluoroacetate (1) using a
substoichiometric amount of (1R,2S)-1-phenyl-2-(1-pyrroli-
dinyl)-propan-1-ol (3a) as a chiral ligand. The imino-
Reformatsky reaction of 1 proceeded enantioselectively to
give 2 with the 3S,4R configuration in good yields, allowing for
the construction of multiple stereogenic centers in a single step.
Furthermore, compound 2 was successfully functionalized to
give a diverse range of α-aryl-α-fluoro-β-lactams without any
reduction in the optical purity. Efforts to further expand the
scope of this reaction and elucidate the mechanism are
currently underway.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures and spectroscopic data of products.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
To demonstrate the utility of the chiral α-bromo-α-fluoro-β-
lactam 2, chiral product 2 was applied to a recently reported Ni-
catalyzed cross-coupling reaction.16a Treatment of (3S,4R)-2a
with aryl Grignard reagents in the presence of NiCl2(DME)
and the Ph-Box ligand afforded the corresponding α-arylated
coupling products 4a, 4d, and 4k diastereoselectively without
Corresponding Author
*E-mail: aando@pharm.setsunan.ac.jp.
Notes
The authors declare no competing financial interest.
2082
dx.doi.org/10.1021/ol500631j | Org. Lett. 2014, 16, 2080−2083

Organic Letters
Letter
REFERENCES
■
(1) (a) Jeschke, P. ChemBioChem 2004, 5, 570−589. (b) Muller, K.;
Faeh, C.; Diederich, F. Science 2007, 317, 1881−1886. (c) O’Hagan,
D. Chem. Soc. Rev. 2008, 37, 308−319. (d) Purser, S.; Moore, P. R.;
Swallow, S.; Gouverneur, V. Chem. Soc. Rev. 2008, 37, 320−330.
(e) Zhang, W.; Cai, C. Chem. Commun. 2008, 5686−5694.
(2) (a) Lectard, S.; Hamashima, Y.; Sodeoka, M. Adv. Synth. Catal.
2010, 352, 2708−2732. (b) Cahard, D.; Xu, X.; Bonnaire, S. C.;
Pannecoucke, X. Chem. Soc. Rev. 2010, 39, 558−568. (c) Zhao, Y.;
Pan, Y.; Sim, S.-B. D.; Tan, C.-H. Org. Biomol. Chem. 2012, 10, 479−
485. (d) Liu, Y.-L.; Yu, J.-S.; Zhou, J. Asian J. Org. Chem. 2013, 2, 194−
206.
(3) Ocamp, R.; Dorbier, W. R., Jr. Tetrahedron 2004, 60, 9325−9374.
(4) (a) Sato, Y.; Fukuda, H.; Tomizawa, M.; Masaki, T.; Shibuya, M.;
Kanoh, N.; Iwabuchi, Y. Heterocycles 2010, 81, 2239−2246.
(b) Wessjohann, L. A.; Scheid, G. O.; Eichelberger, U.; Umbreen, S.
J. Org. Chem. 2013, 78, 10588−10595.
(5) (a) Cozzi, P. G. Angew. Chem., Int. Ed. 2006, 45, 2951−2954.
(b) Cozzi, P. G.; Mignogna, A.; Zoli, L. Synthesis 2007, 17, 2746−
2750. (c) Tanaka, T.; Hayashi, M. Chem. Lett. 2008, 37, 1298−1299.
́
́ ́ ́
(6) (a) Fernandez-Ibanez, M. A.; Macia, B.; Minnaard, A. J.; Feringa,
̃
B. L. Angew. Chem., Int. Ed. 2008, 47, 1317−1319. (b) Fernan
́
dez-
́
Iban
́
ez, M. A.; Macia,
́
B.; Minnaard, A. J.; Feringa, B. L. Chem.
̃
́
́ ́ ́
dez-Ibanez, M. A.; Macia, B.;
̃
Commun. 2008, 2571−2573. (c) Fernan
Minnaard, A. J.; Feringa, B. L. Org. Lett. 2008, 10, 4041−4044.
(7) With carbonyl electrophiles: (a) Fujiwara, Y.; Katagiri, T.;
Uneyama, K. Tetrahedron Lett. 2003, 44, 6161−6163. (b) Shin, E.;
Kim, H. J.; Kim, Y.; Kim, Y.; Park, Y. S. Tetrahedron Lett. 2006, 47,
1933−1935. (c) Cozzi, P. G.; Benfatti, F.; Capdevila, M. G.;
Mignogna, A. Chem. Commun. 2008, 3317−3318. (d) Cozzi, P. G.;
Mignogna, A.; Vicennati, P. Adv. Synth. Catal. 2008, 350, 975−978.
(e) Cozzi, P. G.; Benfatti, F. Tetrahedron: Asymmetry 2010, 21, 1503−
1506. (f) Lin, N.; Chen, M.-M.; Luo, R.-S.; Deng, Y.-Q.; Lu, G.
Tetrahedron: Asymmetry 2010, 21, 2816−2824. (g) Wolf, C.;
Moskowitz, M. J. Org. Chem. 2011, 76, 6372−6376.
(8) With imine electrophiles: (a) Cozzi, P. G.; Rivalta, E. Angew.
Chem., Int. Ed. 2005, 44, 3600−3603. (b) Cozzi, P. G. Adv. Synth.
Catal. 2006, 348, 2075−2079.
́
́ ́ ́
dez-Ibanez, M. A.; Macia, B.; Alonso, D. A.;
̃
(9) See review: Fernan
Pastor, I. M. Eur. J. Org. Chem. 2013, 7028−7034.
(10) (a) Braun, M.; Vonderhagen, A.; Waldmuller, D. Liebigs Ann.
̈
1995, 1447−1450. (b) Andres
́
, J. M.; Martínez, M. A.; Pedrosa, R.;
Perez-Encabo, A. Synthesis 1996, 1070−1072. (c) Knochel, P.;
́
Kloetzing, R. J.; Thaler, T. Org. Lett. 2006, 8, 1125−1128.
(11) Fornalczyk, M.; Singh, K.; Stuart, A. M. Org. Biomol. Chem.
2012, 10, 3332−3342.
(12) Fornalczyk, M.; Singh, K.; Stuart, A. M. Chem. Commun. 2012,
48, 3500−3502.
(13) See review: David, E.; Couve-Bonnaire, S.; Jubault, P.;
Pannecoucke, X. Tetrahedron 2013, 69, 11039−11055.
(14) (a) Zoute, L.; Lemonnier, G.; Nguyen, T. M.; Quirion, J.-C.;
Jubault, P. Tetrahedron Lett. 2011, 52, 2473−2475. (b) Lemonnier, G.;
Poisson, T.; Bonnaire, S. C.; Jubault, P.; Pannecoucke, X. Eur. J. Org.
Chem. 2013, 3278−3289.
(15) Tarui, A.; Kawashima, N.; Sato, K.; Omote, M.; Miwa, Y.;
Minami, H.; Ando, A. Tetrahedron Lett. 2010, 51, 2000−2003.
(16) (a) Tarui, A.; Kondo, S.; Sato, K.; Omote, M.; Minami, H.;
Miwa, Y.; Ando, A. Tetrahedron 2013, 69, 1559−1565. (b) Tarui, A.;
Kawashima, N.; Kawakita, T.; Sato, K.; Omote, M.; Ando, A. J. Org.
Chem. 2013, 78, 7903−7911.
(17) Solvents, reaction temperature, and the amounts of chiral ligand
were examined. See Supporting Information.
(18) Tarui, A.; Kawashima, N.; Sato, K.; Omote, M.; Miwa, Y.;
Minami, H.; Ando, A. Tetrahedron Lett. 2010, 51, 4246−4249.
(19) See Suporting Information for details.
(20) Kitamura, M.; Suga, S.; Niwa, M.; Noyori, R. J. Am. Chem. Soc.
1995, 117, 4832−4842.
(21) Soai, K.; Yokoyama, S.; Hayasaka, T. J. Org. Chem. 1991, 56,
4264−4268.
2083
dx.doi.org/10.1021/ol500631j | Org. Lett. 2014, 16, 2080−2083