Highly enantioselective fluorination reactions of β-ketoesters and β-ketophosphonates catalyzed by chiral palladium complexes

Article abstract of DOI:10.1016/j.tet.2005.12.070

Using chiral palladium enolates as key intermediates, efficient catalytic enantioselective fluorination reactions of active methine compounds have been developed. These reactions can be conducted in alcoholic solvents without any precaution to exclude wat

Full text of DOI:10.1016/j.tet.2005.12.070

Tetrahedron 62 (2006) 7168–7179
Highly enantioselective fluorination reactions of b-ketoesters and
b-ketophosphonates catalyzed by chiral palladium complexes
Yoshitaka Hamashima,a Toshiaki Suzuki,a Hisashi Takano,a Yuta Shimura,a Yasunori Tsuchiya,a
Ken-ichi Moriya,a Tomomi Gotoa and Mikiko Sodeokaa,b,
*
aInstitute of Multidisciplinary Research for Advanced Materials, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577, Japan
bRIKEN (The Institute of Physical and Chemical Research), 2-1 Hirosawa, Wako, 351-0198, Japan
Received 11 December 2005; accepted 12 December 2005
Available online 6 May 2006
Abstract—Using chiral palladium enolates as key intermediates, efficient catalytic enantioselective fluorination reactions of active methine
compounds have been developed. These reactions can be conducted in alcoholic solvents without any precaution to exclude water and mois-
ture, and various b-ketoesters and b-ketophosphonates were fluorinated in a highly enantioselective manner (up to 98% ee).
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
ketones, aldol reactions, and ene reactions, to fluorinated
starting materials is an important strategy, and thus chiral
compounds, which do not have a fluorine atom at a chiral
carbon center can be utilized for the preparation of fluori-
nated drug candidates.2,3 However, the use of optically ac-
tive compounds bearing a fluorine atom at a chiral carbon
center is restricted by the limited availability of effective
methods for enantioselective construction of fluorinated
stereogenic carbon centers. Thus, enantioselective synthesis
of organofluorine compounds via direct introduction of fluo-
rine atoms is still a highly desirable goal in synthetic organic
chemistry.4 In addition to diastereoselective fluorination
reactions,5 great efforts have been made to develop chiral
fluorinating reagents.6 Shibata et al. and Cahard et al. inde-
pendently reported an ingenious procedure that allows in situ
generation of a chiral fluorinating reagent from cinchona
alkaloids and Selectfluor (see Chart 3).7 Since the initial
reports by Togni et al., catalytic enantioselective fluorination
reactions have been attracting much attention.8,9 In 2002, on
the basis of our palladium enolate chemistry, we developed
an efficient method for enantioselective fluorination reac-
tions with high generality as regards b-ketoesters.10
Subsequent studies from other laboratories revealed that
late transition metal complexes consisting of Cu(II), Ni(II),
and Zn(II) were also excellent catalysts for catalytic enantio-
selective fluorination reactions of active methine com-
pounds.11 Several groups have reported attempts at
applying organocatalysis to asymmetric fluorination reac-
tions, and catalytic enantioselective monofluorination of al-
dehydes having two hydrogens at the a-position has been
achieved.12,13 Although the scope of available substrates
for asymmetric fluorination reactions is rapidly expanding,
continuing exploitation of novel catalysts, including metal
Acquisition of novel activity distinct from that of the parent
compound by replacing atom(s) in a biologically active com-
pound with other atom(s) is a very important aspect of
medicinal chemistry. In particular, introduction of fluorine,
which rarely occurs in natural products, into bioactive com-
pounds sometimes leads to significant improvement of their
biological activity profiles,1 probably due to the unique
properties of the fluorine atom and/or the carbon–fluorine
bond (Chart 1).
1) Steric size effect [H: 1.2 Å; F: 1.35 Å]
C
H
2) Strong bond energy
C
F
[C-F: 116 kcal/mol; C-H: 99 kcal/mol]
3) The largest electron negativity
4) Hydrogen bond acceptor
C
OH
Chart 1. Properties of the fluorine atom and carbon–fluorine bonds.
For this reason, the effect of fluorine-substitution is often
examined in the course of development of new drug candi-
dates. Most such investigations have focused on replacement
of hydrogen(s) on an aromatic ring with fluorine atom(s).
The effect of substitution at sp3 carbons has been less well
investigated. For the synthesis of chiral fluorinated com-
pounds, application of well-established asymmetric reac-
tions, including hydrogenation of olefins, reduction of
Keywords: Fluorination; Palladium; b-Ketoesters; b-Ketophosphonates;
Asymmetric catalysis.
* Corresponding author. Tel./fax: +81 22 217 5601; e-mail addresses:
sodeoka@tagen.tohoku.ac.jp; sodeoka@riken.jp
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.12.070

Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
7169
complexes and organic catalysts, is necessary to meet the
need for various chiral fluorinated compounds, many of
which are expected to find applications in the fields of me-
dicinal chemistry, chemical biology, and material sciences.
Herein, we present full details of our study on the catalytic
enantioselective fluorination reactions of b-ketoesters and
b-ketophosphonates (Eqs. 1 and 2).10 Our palladium com-
plexes were effective catalysts for such active methine com-
pounds, affording the corresponding fluorinated products in
a highly enantioselective manner (up to 98% ee). We con-
firmed the utility of our methods by stereoselective synthe-
ses of fluorinated derivatives of fundamental units found in
biologically active natural and unnatural compounds.
and Scheme 1).14 Because the palladium aqua and m-
hydroxo complexes are inherently acidic or neutral, this
enolate formation is considered to occur under nonbasic
conditions. We envisaged that this palladium enolate chem-
istry would be applicable to enantioselective electrophilic
fluorination reactions. Development of the fluorination
reactions of b-ketoesters is valuable, because optically
active a-substituted a-fluoro-b-ketoesters are regarded as
nonenolizable b-ketoesters. In addition, since ketone is
easily converted to other functional groups, a-substituted
a-fluoro-b-ketoesters would be versatile synthetic precur-
sors of various a-fluorinated carboxylic acid derivatives.
Therefore, we planned to examine the enantioselective
fluorination reactions of b-ketoesters using the palladium
complexes 1 and 2.
OH
O
∗
O
O
∗
O
R
O
3
1
OR
OR
R
4
Pd-cat.
∗
(1)
1
3
2
3
1
F
R
OR
R
OR
R
O
O
2
2
F
R
R
HN
1
O
R'
R'
+
+
F-X
(3)
H
X
∗
R
R
∗
3
R
H
F
2
acidic co-product
F
R
H O, TfOH
*
2
O
O
O
R
O
O
O
P(OH)
P
P
Pd-cat.
3
3
P(OR )
∗
∗
P(OR )
(2)
1
2
X-
2
2
1
1
1
Pd+
R
R
R
O
2
2
2
F
F
O
O
R
R
3
CO R
2
1
R
3
1
OR
R
2
R
2
R
A
2
2. Results and discussion
H O
2
Chiral palladium enolate
In principle, electrophilic fluorination reactions of carbonyl
compounds entail concomitant generation of an acidic co-
product during the reaction (Eq. 3). It seems difficult to carry
out the reaction in the presence of strongly basic catalysts to
remove a hydrogen atom at the a-position, because decom-
position or neutralization of the catalysts might occur. Thus,
activation of the nucleophiles under acidic or neutral condi-
tions would be a promising alternative approach to carry out
enantioselective electrophilic fluorination catalytically. We
have already reported that the cationic palladium complexes
1 and 2 reacted with 1,3-dicarbonyl compounds, such as
b-ketoesters, to form chiral palladium enolates A (Chart 2
Scheme 1. Formation of chiral palladium enolates.
Since a bulky ester moiety was found to be essential for high
asymmetric induction in our previous study on catalytic
asymmetric Michael reactions,14 we chose tert-butyl
2-oxo-cyclopentanecarboxylate 3a as a model substrate.
Among several electrophilic fluorinating reagents, N-fluoro-
benzenesulfonimide (NFSI) 4 was the most effective (Chart
3). While salt-type reagents, such as 5 and 6, did not give the
desired product 7a in a detectable amount, the reaction of 3a
with 4 under the influence of 1a (5 mol %) in THF proceeded
smoothly to afford 7a in 72% yield with 79% ee (Table 1, en-
try 1). It should be noted that the palladium complex retained
its catalytic activity until the completion of the reaction. In
contrast, a stoichiometric amount of a conventional base
would be required for the reaction with NFSI, because sulfo-
nimide [(PhSO2)2NH] with high acidity was formed during
the reaction. Therefore, our reaction system was considered
suitable for the development of fluorination reactions.
H
2+
Pd
+
Pd
+
Pd
P
P
P
P
O
P
OH
OH
2
2
*
*
*
O
H
P
-
-
2X
2X
1
2
O
PAr
PAr
PAr
PAr
PPh
PPh
2
2
2
2
O
O
2
2
-
SO Ph
2
O S
Cl
3
O
+
+
F
N
F
N
N
+
SO Ph
2
F
N
h
a-d
e-g
-
2( BF )
4
a: Ar = Ph: (R)-BINAP
b: Ar = 4-MeC H : ( )-Tol-BINAP
N-fluoro-4-methylpyridinium
-2-sulfonate (6)
NFSI (4)
Selectfluor (5)
R
6
4
c: Ar = 4-MeOC H : (R)-MeO-BINAP
6
4
6 3
Chart 3. Electrophilic fluorinating reagents examined in this study.
d: Ar = 3,5-Me C H : ( )-DM-BINAP
e: Ar = Ph: (R)
R
2
-SEGPHOS
f : Ar = 3,5-Me C H : (R)-DM-SEGPHOS
2
6 3
To improve the enantioselectivity, we examined a series of
chiral phosphine ligands. Among the diphosphines, bulkier
ligands bearing substituents at the meta positions of aryl
group on phosphine gave better enantioselectivity (entries
g: Ar = 3,5-(t-Bu) -4-MeOC H : (R)-DTBM-SEGPHOS
2
6 2
h: (R)-H -BINAP
8
Chart 2. Chiral palladium complexes.

7170
Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
Table 1. Optimization of the reaction conditions
reaction did not proceed when 1g was used (entry 1). The
Pd-cat. 1 or 2
O
O
less bulky 1d promoted the reaction, but the reaction in
THF was sluggish, giving rise to 7b in 63% yield with
82% ee after more than 250 h (entry 2). Fortunately, dra-
matic acceleration of the reaction was observed when
EtOH was used as a solvent. The reaction reached comple-
tion after 48 h, affording the desired product in 96% yield
with 91% ee (entries 3 and 4). Even when the concentration
was reduced from 1 M to 0.1 M, 7b was isolated in 54%
yield, maintaining the same enantioselectivity (entry 5).
1H NMR study revealed that stoichiometric reaction of 3a
with 2b (0.5 equiv to 3a) was completed within 10 min in
CD3OD, while the same reaction in THF-d8 took 2 h for
completion.14a Thus, we speculated that alcoholic solvents
played a key role for rapid formation of the chiral palladium
enolates, thereby significantly accelerating the reaction.
(X = TfO)
NFSI (4)
∗
CO t-Bu
CO t-Bu
2
2
+
(1.5 eq)
solvent, 1 M
F
3a
7a
Entry
Catalysta
(mol %)b
Solvent
Temp
(ꢀC)
Time
(h)
Yieldc
(%)
eed
(%)
1
2
3
4
5
6
7
8
1a (5)
1b (5)
1c (5)
1d (5)
1e (5)
1g (5)
1h (5)
2g (2.5)
2g (2.5)
2g (2.5)
2g (2.5)
2g (2.5)
THF
THF
THF
THF
THF
THF
THF
THF
Acetone
EtOH
i-PrOH
t-BuOH
ꢁ20
ꢁ20
ꢁ20
ꢁ20
ꢁ20
0
12
12
7.5
39
39
72
72
87
92
99
82
89
92
83
93
73
90
68
79
83
80
88
71
90
82
92
92
92
92
93
ꢁ20
7.5
10
10
48
48
18
18
18
9
10
11
12
20
20
20
Because the palladium complexes are stable to water, the
reaction could be performed in aqueous solvent systems
(entries 6–9). Interestingly, the reaction in water without
co-solvents proceeded at a synthetically useful level to
give 7b in 76% yield with 89% ee (entry 9).
a
1–2: Catalyst structure; a–h: chiral ligand.
Catalyst amount.
Isolated yield.
b
c
d
Determined by chiral HPLC analysis.
4 and 6), while substituents at the para position and semi-re-
duction of the binaphthyl scaffold did not significantly influ-
ence the reaction efficiency (entries 2, 3, and 7).15 In contrast
to the Michael reaction, the use of the Pd m-hydroxo complex
2g also promoted the reaction smoothly,16 and the best selec-
tivity of 92% ee was observed (entry 8). This difference in
reactivity may be attributed to the higher electrophilicity
of NFSI than that of the enone. Further optimization of the
reaction conditions revealed that the reaction proceeded
more rapidly in polar solvents (entries 9–12). Interestingly,
an alcoholic solvent such as EtOH or i-PrOH was the best
of those tested and the reaction time was reduced from
48 h to 18 h without any loss of enantioselectivity (entries
10 and 11). In the case of t-BuOH, however, the reaction
was retarded and the starting material was not consumed
completely after 18 h (entry 12).
With the optimized reaction conditions in hand, we next ex-
amined the generality of our fluorination reaction. As sum-
marized in Table 3, other cyclic and acyclic b-ketoesters
were converted to the desired products smoothly in the pres-
ence of 2.5 mol % of the palladium complexes. Except for
3g, the desired fluorinated compounds were obtained in
a highly enantioselective manner (up to 94% ee). As in the
case of 3a, the reaction of 3c using 2d gave the product
with 94% ee, while a slight decrease of enantioselectivity
was observed when 1d was used (entries 1 and 7). In entry
3, the isolated chemical yield was lower due to the volatility
of 7e, but the excellent ee of 91% was observed. In contrast
to 3f, the reaction of 3g, which has a different substitution
Table 3. Catalytic enantioselective fluorination reactions of b-ketoesters
Pd-cat.
O
O
(2.5 mol%)
CO t-Bu
A conspicuous solvent effect was observed when 3b was
tested as a substrate (Table 2). Probably because of steric re-
pulsion derived from the phenyl ring of the substrate, the
2
CO t-Bu
1
+
2
4
1
R
R
(1.5 eq) EtOH
1 M
2
2
R
3b-g
R
F
7b-g
O
O
O
CO t-Bu
2
CO t-Bu
2
1
R
Table 2. Solvent effect on the reaction of 3b
CO t-Bu
2
2
R
Pd-cat.
O
O
1
1
1
2
(X = TfO)
3c
3d
3e: R = Me, R = Me
CO t-Bu
2
CO
t-Bu
2
+
4
2
Ph
Ph
3f: R = Me, R = Et
solvent
1 M, 20 °C
(1.5 eq)
2
Me
F
Me
3b
3g: R = Et, R = Me
7b
Entry
3
Catalyst (X) Temp (ꢀC) Time (h) Yield (%) ee (%)
Entry
Catalyst
(mol %)
Solvent
Time
(h)
Yield
(%)
ee
(%)
1
2
3
4
3c 2d (BF4)
3d 2d (TfO)
3e 2f (TfO)
3f 2d (TfO)
3g 2d (TfO)
3c 2d (BF4)
3c 1d (TfO)
3b 1d (TfO)
ꢁ10
ꢁ20
20
20
20
20
36
72
42
72
20
20
48
91
85
49b
88
47
82
95
96
94
83a
91
87
69
91
92
91a
1
2
1g (5)
1d (5)
1d (5)
2d (2.5)
2d (2.5)
2d (2.5)
2d (2.5)
2d (2.5)
2d (2.5)
THF
THF
EtOH
EtOH
EtOH
EtOH/H2O (1/1)
EtOH/H2O (4/1)
Acetone/H2O (9/1)
H2O
—
255
48
40
40
66
120
96
75
No reaction
63
96
92
54
57
53
73
76
82
3
91
91
91
78
85
88
89
5
4a
5b
6
6c
7
0
ꢁ10
8d
20
7
8
9
a
The absolute configuration was determined to be R after conversion to the
known compound.
Lower yield due to the volatility of 7e.
Compound 2d (1 mol %).
One gram scale.
b
c
d
a
X¼BF4.
b
0.1 M.

Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
Table 4. Optimization of the reaction conditions
7171
pattern from 3f, was slower and less enantioselective. Thus,
7g was obtained in only 47% yield with 69% ee after 72 h.
Even when the amount of catalyst was reduced to
1 mol %, comparable results were obtained (entry 6). It is
also noteworthy that this reaction could be easily scaled up
using reagent-grade non-distilled EtOH as a solvent (entry
8). In these reactions, we found that 2d and 2g were effective
catalysts, and various substrates were selectively fluorinated
by employing either of these two ligands according to the
nature of the b-ketoesters. Notably, it is environmentally
advantageous that this reaction proceeds well in alcoholic
solvents and even in water. Unfortunately, the reaction of
a-nonsubstituted b-ketoester 8 was unsuccessful (Scheme
2). Because the corresponding product 9 was more suscepti-
ble to enolization than 8, no asymmetric induction was
observed and a difluorinated compound 10 was obtained in
4% yield.
Pd-cat. 1 or 2
(X = TfO)
O
O
P(OEt)
O
O
NFSI (1.5 eq.)
P(OEt)
F
∗
2
2
solvent, rt, 1 M
11a
12a
Entry Catalyst (mol %) Solvent Time (h) Yielda (%) eeb (%)
1
2
3
4
5
6
7
8
1a (5)
1b (5)
1d (5)
1f (5)
1g (5)
1d (5)
1d (5)
1d (5)
1d (5)
2d (2.5)
EtOH
EtOH
EtOH
EtOH
EtOH
Acetone
THF
DMF
CH2Cl2
EtOH
5
5
2
2
5
3
3
6
6
2
81
82
91
87
46
70
66
20
43
91
75
79
95
95
98
93
92
90
92
95
9
10
a
Isolated yield.
Determined by HPLC analysis.
b
O
Pd cat. 2d
(X = TfO)
CO t-Bu
2
54%
Ph
Ph
b-ketophosphonates, the Pdm-hydroxo complex 2d gave
results comparable to those obtained using 1d (entry 10).
0% ee
H
F
(2.5 mol%)
O
NFSI (1.5 equiv.)
9
CO t-Bu
2
Ph
°
EtOH, 18 h, 20 C
O
This catalytic system was also applicable to various sub-
strates including cyclic and acyclic b-ketophosphonates
(Table 5). All the substrates examined were fluorinated in
a highly enantioselective manner (up to 97% ee). As shown
in entries 1 and 6, the reactions smoothly proceeded in the
presence of as little as 1 mol % of the catalyst without dete-
rioration of the reaction efficiency. When the reaction was
carried out at 0 ꢀC, the ee was improved to 97% ee (entry
5). In contrast to cyclic b-ketophosphonates, the reaction of
acyclic substrates was found to be slow (entries 7 and 8).
The starting materials 11e and 11f were not consumed at
40 ꢀC even after 48 h. Although the chemical yield was mod-
est to good, the ees of the products were found to be excellent
(94 and 95% ee, respectively). For these reactions, 1f gave
8
CO t-Bu
2
4%
F
F
10
Scheme 2. Fluorination of a-nonsubstituted b-ketoester.
Encouraged by the success in the fluorination reactions of
b-ketoesters, we next turned our attention to other active
methine compounds. We envisaged that other bidentate car-
bonyl compounds would react with Pd complexes to form
similar chiral palladium enolates. Among several candidates,
we focused on b-ketophosphonates17 because difluoro- and
monofluorophosphonates have been utilized in drug design
as mimics of phosphates.18,19 Compared with non-fluori-
nated phosphonates and difluorophosphonates, a-mono-
fluorophosphonates are expected to be a better surrogate of
phosphates, because they show similar second pKa values
(~6.5) to those of biological phosphates (~6.5).18 Although
several non-enantioselective or diastereoselective syntheses
of a-monofluorophosphonates have been reported,20 there
was no example of the catalytic enantioselective synthesis
of chiral a-fluoro-b-ketophosphonates before we started
our investigation. Independently, Jørgensen and Kim re-
cently reported similar reactions using Zn(II)–Ph–DBFOX
complexes and Pd–BINAP complexes, respectively.11e,f
Table 5. Catalytic enantioselective fluorination of b-ketophosphonates
Pd cat. 1
O
O
O
O
(X = TfO)
NFSI
(1.5 eq)
P(OEt)
+
P(OEt)
2
1
2
1
R
R
EtOH, 1 M
2
2
R
F
11
12
R
O
O
O
O
O
O
P(OEt)
2
P(OEt)
P(OEt)
2
2
R
Me
n
n
11a: n = 1
11b: n = 2
11c: n = 1
11d: n = 2
11e: R = Me
11f: R = Ph
Similar optimization of the reaction conditions using 11a as
a model substrate revealed that the combination of 1d as
a catalyst and EtOH as a solvent was the most appropriate
for the fluorination of b-ketophosphonates (Table 4). Thus,
when 5 mol % of 1d was used in EtOH, the reaction of 11a
reached completion after only 2 h to give the desired fluori-
nated product 12a in 91% yield (entry 3). Gratifyingly, the ee
of the product was determined to be 95% by chiral HPLC
analysis. When a bulkier ligand such as DTBM-SEGPHOS
was used, a higher enantioselectivity of 98% was observed
(entry 5), but the reaction rate was considerably decreased,
probably due to severe steric repulsion. EtOH was found to
be superior to other solvents in terms of chemical yield
and enantioselectivity (entries 6–9). In this fluorination of
Entry
11
Catalyst
(mol %)
Temp
(ꢀC)
Time
(h)
Yielda
(%)
eeb
(%)
1
2
3
4
5
6
7
8
11a
11b
11c
11d
11d
11d
11e
11f
1d (1)
1d (5)
1d (5)
1d (5)
1d (5)
1d (1)
1f (10)
1f (10)
rt
rt
rt
rt
0
rt
40
40
12
8
3
3
24
7
82
93
84
97
90
83
57
38
95
96
95
94c
97
95
94d
95
48
48
a
Isolated yield.
b
c
d
Determined by HPLC analysis.
The absolute configuration was determined by X-ray analysis.
The ee was determined after conversion to the corresponding 2,4-dinitro-
phenylhydrazone 13.

7172
Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
a slightly better selectivity than 1d. Similarly decreased reac-
tion rates in the case of acyclic substrates were noted in Kim’s
report.11e They examined various acyclic b-ketophospho-
nates employing Pd complexes similar to our Pd aqua com-
plexes 1, and longer reaction times (58–94 h) were required
to obtain the products in yields ranging from 50–79%.
To establish the absolute configuration of these fluorinated
products, the following conversion of 7b and 12d was car-
ried out (Scheme 3). As described previously,21 the absolute
configuration of 7b was determined to be R by comparison
of optical rotation of 14 with the reported value.5a In addi-
tion, R configuration was observed in the case of 7d after
conversion to 2-fluoro-2-methyl indanone, whose optically
active form was previously synthesized by Shibata and
Takeuchi.6c,7a,21 To determine the absolute stereochemistry
of the fluorinated b-ketophosphonate 12d, we planned to
attach a chiral auxiliary in some way, because no appropriate
reference compound could be found in the literature. Thus,
stereoselective reduction with NaBH4, followed by esteri-
fication with N-(2-carboxy-4,5-dichlorobenzoyl)-(ꢁ)-10,
2-camphorsultam (15) gave 16 in good yield.22
Recrystallization of 16 from hexane/ethyl acetate (3/1)
gave a single crystal suitable for X-ray structural analysis.
As shown in Figure 1, on the basis of the known structure
of (ꢁ)-camphorsultam, the absolute configuration of 12d
was unequivocally determined to be S.
Figure 1. X-ray structure of 16.
Re face
Re face
+
Pd
+
P
P
P
P
Pd
O
O
O
O
1
1
R
R
EtO
P
O
2
2
R
R
EtO
t-Bu
Si face
Si face
17
Figure 2. Plausible transition state models.
18
1) TFA
O
O
2) CH N
2
2
CO t-Bu
CO Me
2
2
Ph
Ph
Me
F
76%
Me
F
with the 3,5-dimethylphenyl group of the ligand. Thus, the
si-face of the enolate 17 might be effectively shielded by
the tert-butyl group and the aryl group on phosphine, forcing
NFSI to approach from the re-face of the enolate. The face
discrimination in the case of b-ketophosphonates can also
be explained by postulating involvement of a similar inter-
mediate. But, the structure of the corresponding enolate
would be more complex because of the sp3-hybridization
of the phosphorus atom of the substrate. Likewise, the two
ethoxy groups of the b-ketophosphonates 11 would be posi-
tioned to cause minimum steric repulsion with the aryl group
on phosphine. Comparing 18 with 17, the si-face of the eno-
late 18 seems more crowded than that of 17, since one of the
ethoxy groups should be oriented much closer to the reactive
center. This speculation is in accord with the better enantio-
selectivity observed in the case of b-ketophosphonates than
that of b-ketoesters.
7b
14
24
[α]D = - 82.9 (C = 1.0, MeOH)
5a 25
Lit. [α]
= - 85.0 (C = 1.0, MeOH)
(98% ee, R enantiomer)
D
NaBH
4
O
O
P(OEt)
F
OH
O
P(OEt)
F
(4 equiv.)
2
2
EtOH
0 °C ~ rt, 5h
90%
12d
Cl
15 (2 equiv)
EDAC (4 equiv)
DMAP (1 equiv)
Cl
N
S
O
O
CH Cl
O
2
2
O
O
O
P(OEt)
F
0 °C ~ rt, 17 h
2
16
84%
Finally, to confirm the synthetic utility of our fluorination
reactions, transformation of the fluorinated products was
investigated (Scheme 4). Because b-hydroxy or b-amino
acids are fundamental units in various natural or unnatural
compounds, their a-fluorinated derivatives are of particular
interest. First, conventional reduction with NaBH4 in the
presence of Lewis acids was examined, but it gave only
a mixture of the reduced products 19 and 20 (~1/6).
According to the reported procedure,23 we next tried reduc-
tion with silanes. Although the use of AlCl3 as an activator of
the ketone was unsuccessful, the reaction in TFA afforded 19
in a highly diastereoselective manner. We previously re-
ported that reduction with Ph3SiH afforded 19 in 75% yield.
When this silane was switched to Et3SiH, the chemical yield
Scheme 3. Determination of the absolute configuration of the products.
Because formation of the chiral palladium enolates was ob-
served using 1H NMR in the cases of both b-ketoesters and
b-ketophosphonates,10,14 we believe that these enolates are
key intermediates in the fluorination reactions. If structurally
similar enolates are generated in the reaction mixture, the
absolute configuration observed in these reactions indicates
that the fluorinating reagent reacts from the less hindered
side of the enolates with the same sense of enantioselection
(Fig. 2). In the case of b-ketoesters, a bulky tert-butyl group
would be located at one enolate face to avoid steric repulsion

Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
7173
was improved to 92%. The relative configuration can be ex-
plained by assuming a chelation model. In addition, Lewis
base activation of PhMe2SiH was found to be effective for
stereoselective reduction (drꢂ95/5).23 This reaction is likely
to proceed according to a Felkin–Anh model, and another di-
astereomer 20 was produced in 83% isolated yield. These
compounds were subjected to azidation with inversion of
configuration. Reduction of the azide group, followed by
protection of the amino group, afforded the a-fluoro-b-
amino esters 21 and 22 in good yields. Enantioenrichment
of 21 by recrystallization from ethyl acetate was possible,
and optically pure 21 was obtained (>99% ee). In addition,
the cyclic fluorinated product 7c was also converted to 23 as
a major diastereomer in 76% yield. Relative stereochemistry
of 23 was deduced from a comparison of the coupling con-
stants of the methine proton adjacent to the hydroxyl group
against the fluorine atom, which are larger than those
between protons (23: JH–F¼10.0 Hz; minor isomer:
JH–F¼24.0 Hz).24 In the presence of benzylamine and
NaCNBH3, 7c underwent reductive amination to give 24
as the major product in 65% yield (coupling constants of
b-ketophosphonates have been developed. Our reactions
are operationally convenient because special precautions to
exclude air and moisture are unnecessary, and reagent-grade
non-distilled alcoholic solvents could be used as reaction
media. In addition, transformation of the fluorinated prod-
ucts was successfully demonstrated, and the availability of
these fluorinated fundamental units should be valuable in
medicinal chemistry. Further studies to expand the scope
of the fluorination reactions and their application for the
development of enzyme inhibitors are under way in our
laboratory.
4. Experimental
4.1. General
All asymmetric reactions were carried out without precau-
tions to exclude air and moisture. Catalysts used in this paper
were prepared according to the reported procedure.10a,26
NMR spectra were recorded on a JEOL JNM-LA400 spec-
trometer, operating at 400 MHz for 1H NMR and 100.4 MHz
for 13C NMR. Chemical shifts were reported downfield from
the methine protons; 24: JH–F¼12.7 Hz; minor: JH–F
¼
27.6 Hz).24 A further example is as follows. To evaluate
the ability of the fluorinated phosphonates to act as phos-
phate mimics, dealkylation of 12d is essential. When 12d
was treated with 6 equiv of TMSBr in CH2Cl2, removal of
two ethyl groups occurred readily, affording the desired
phosphonic acid 25 in excellent yield.25
1
TMS (¼0) for H NMR. For 13C NMR, chemical shifts
were reported relative to the solvent used as an internal ref-
erence. 19F NMR was measured at 400 or 376 MHz, and
CF3COOH was used as an external standard. FAB–LRMS
and FAB–HRMS were taken on JEOL JMS GCmate II using
m-nitrobenzyl alcohol (mNBA) as a matrix. Optical rotations
were measured on a JASCO DIP-370 polarimeter. Column
chromatography was performed with silica gel 60 (40–
100 mm) purchased from Kanto Chemical Co. The enan-
tiomeric excesses (ees) were determined by HPLC or GC
analysis. HPLC analysis was performed on Shimadzu HPLC
systems consisting of the following components: pump,
LC-10AD; detector, SPD-10A set at 254 or 280 nm; column,
DAICEL CHIRALPAK AS, AD-H, and CHIRALCEL
OJ-H; mobile phase, hexane/2-propanol (IPA). GC analysis
was performed on Shimadzu GC-17A with Tokyo Kasei
CHIRALDEX G-TA (0.25 mm i.d. ꢃ30 m ꢃ0.125 mm).
Tetrahydrofuran (THF) was distilled from sodium benzo-
phenone ketyl. EtOH was distilled from calcium hydride.
Other reagents were purified by usual methods.
Et SiH
3
PhMe SiH
2
OH
OH
TFA
TBAF
CO Me
14
CO Me
2
2
Ph
Ph
t, 2 h
DMF
r
Me
92%
(dr = >95/5)
F
Me
83%
(dr = >95/5)
F
°
0
C, 20 min
19
20
1) DEAD 2) Pd/C, H2
1) DEAD 2) Pd/C, H
2
Ph3P
DPPA
THF
(Boc)2O
MeOH
rt, 1 h
78%
Ph P
3
(Boc) O
2
DPPA
THF
MeOH
rt, 1h
80%
rt, 4 h
95%
rt, 2 h
79%
NHBoc
CO Me
NHBoc
CO Me
2
2
Ph
Ph
Me
F
Me
F
21
22
1) BnNH
2
4.2. General procedure for the catalytic enantioselective
fluorination of b-ketoesters
NHBn
CO t-Bu
OH
O
NaBH
toluene, MS4A
reflux, 4 h
4
(2 equiv)
CO t-Bu
F
CO t-Bu
2
2
2
F
F
EtOH
-78 °C
2) NaCNBH , AcOH
3
To a solution of the palladium complex 2 (0.005 mmol) in
EtOH (0.2 mL) was added a b-ketoester (0.2 mmol) at
room temperature. At the indicated temperature, NFSI
(95 mg, 0.3 mmol) was added. The resulting mixture was
stirred for the time given in Table 3. After the completion
of the reaction (TLC, hexane/ether¼10/1), saturated aque-
ous NH4Cl was added for quenching. The aqueous layer
was extracted with Et2O (3ꢃ10 mL). The combined organic
layers were washed with water and brine. After drying over
Na2SO4, the solvent was removed under reduced pressure.
Further purification was performed by flash column chroma-
tography on SiO2 (hexane/Et2O¼10/1) to give the pure
product as a colorless oil. The ees of the products were
determined by chiral HPLC or GC analysis. In this study,
decoupled 19F NMR spectra were obtained for all the
products.
0 °C ~ rt, 2 h
7c
23 76%
24 65%
dr = 15/1
dr = 8/1
TMSBr
(6 equiv)
O
O
O
O
recryst.
P(OH)
F
P(OEt)
2
F
2
CH Cl
2
2
rt, 12 h,
25
97 % (>95% ee)
then MeOH
12d (95% ee)
Scheme 4. Conversion of the fluorinated products.
3. Conclusion
Using chiral palladium enolates as key intermediates, highly
enantioselective fluorination reactions of b-ketoesters and

7174
Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
4.2.1. tert-Butyl 1-fluoro-2-oxocyclopentanecarboxylate
(7a). H NMR (400 MHz, CDCl3) d 1.49 (s, 9H), 2.07–
3H), 2.30 (d, J¼4.4 Hz, 3H); 13C NMR (100 MHz, CDCl3)
d 19.5 (d, J¼23.1 Hz), 24.9, 27.5, 83.8, 97.7 (d,
J¼192 Hz), 165.9 (d, J¼25.5 Hz), 202.5 (d, J¼28.8 Hz);
19F NMR (400 MHz, CDCl3) d ꢁ156.6; FABMS (mNBA)
m/z 191 (M+1)+; [a]D34 ꢁ44.0 (c 0.91, CHCl3) (89% ee);
GC (Tokyo Kasei CHIRALDEX G-TA; 0.25 mm i.d.
ꢃ30 m ꢃ0.125 mm; temp 70 ꢀC, inj. temp 300 ꢀC, det.
temp 250 ꢀC) tr (minor)¼21.0 min, tr (major)¼21.7 min;
FAB–HRMS (mNBA) Calcd for C9H15O3F (M)+
190.1005. Found 190.1000; IR (neat) n 2981, 2935, 1753,
1
2.15 (m, 2H), 2.21–2.34 (m, 1H), 2.44–2.57 (m, 3H); 13C
NMR (100 MHz, CDCl3) d 18.0 (d, J¼4.1 Hz), 27.9, 33.8
(d, J¼20.6 Hz), 35.7, 84.0, 94.4 (d, J¼199.1 Hz), 166.4
(d, J¼27.9 Hz), 208.1 (d, J¼16.4 Hz); 19F NMR
(400 MHz, CDCl3) d ꢁ163.2; FABMS (mNBA) m/z 202
(M)+, 146 (Mꢁt-Bu)+; [a]D31 +72.7 (c 1.27, CHCl3) (92%
ee); HPLC (DAICEL CHIRALPAK AD-H, hexane/
IPA¼99/1, 0.40 mL/min, 280 nm) tr (minor)¼20.3 min, tr
(major)¼24.7 min; FAB–HRMS (mNBA) Calcd for
C10H15O3F (M)+ 202.1005. Found 202.1002. Calcd for
C10H16O3F (M+1)+ 203.1084. Found 203.1082; IR (neat) n
1736, 1458, 1395, 1370, 1291, 1256, 1134, 1106, 840 cmꢁ1
.
4.2.6. tert-Butyl 2-ethyl-2-fluoro-3-oxobutyrate (7f).
1H NMR (400 MHz, CDCl3) d 0.95 (t, J¼7.4 Hz, 3H),
1.49 (s, 9H), 1.94–2.20 (m, 2H), 2.29 (d, J¼4.64 Hz, 3H);
13C NMR (100 MHz, CDCl3) d 7.00, 25.7, 26.9 (d,
J¼21.4 Hz), 27.8, 83.8, 100.6 (d, J¼194 Hz), 165.3 (d,
J¼26.3 Hz), 202.4 (d, J¼32.9 Hz); FABMS (mNBA) m/z
205 (M+1)+, 147 (Mꢁt-Bu)+, 148 (M+2ꢁt-Bu)+; [a]3D1
ꢁ12.6 (c 1.45, CHCl3) (87% ee); HPLC (DAICEL
CHIRALPAK AD-H, hexane/IPA¼150/1, 0.40 mL/min,
280 nm) tr (minor)¼11.6 min, tr (major)¼12.5 min; FAB–
HRMS (mNBA) Calcd for C10H18O3F (M+1)+ 205.1240.
Found 205.1244; IR (neat) n 2920, 2850, 1736, 1709,
2978, 2927, 1764, 1746, 1458, 1395, 1370, 1145, 840 cmꢁ1
.
4.2.2. (R)-tert-Butyl 2-fluoro-2-methyl-3-oxo-3-phenyl-
propionate (7b). H NMR (400 MHz, CDCl3) d 1.37 (s,
1
9H), 1.82 (d, J¼22.4 Hz, 3H), 7.43–7.48 (m, 2H), 7.56–
7.60 (m, 1H), 8.02–8.06 (m, 2H); 13C NMR (100 MHz,
CDCl3) d 20.6 (d, J¼23.9 Hz), 27.6, 84.0, 96.6 (d, J¼
193 Hz, 1H), 128.5, 129.5, 133.6, 167.5 (d, J¼25.5 Hz),
191.7 (d, J¼25.5 Hz); 19F NMR (400 MHz, CDCl3) d
ꢁ151.2; FABMS (mNBA) m/z 253 (M+1)+, 197 (M+2Hꢁt-
Bu)+; [a]D33 +74.0 (c 1.3, CHCl3) (91% ee); HPLC
(DAICEL CHIRALPAK AD-H, hexane/IPA¼200/1,
0.40 mL/min, 254 nm) tr (major)¼17.7 min, tr (minor)¼
19.1 min; FAB–HRMS (mNBA) Calcd for C14H18O3F
(M+1)+ 253.1240. Found 253.1240; IR (neat) n 2981, 2935,
1754, 1700, 1449, 1395, 1370, 1288, 1247, 1145, 1123,
1459, 1446, 1376, 1280, 1261, 1168, 1084, 797 cmꢁ1
.
4.2.7. tert-Butyl 2-fluoro-2-methyl-3-oxopentanoate (7g).
1H NMR (400 MHz, CDCl3) d 1.09 (t, J¼7.3 Hz, 3H),
1.47 (s, 9H), 1.63 (d, J¼22.2 Hz, 3H), 2.65–2.70 (m, 2H);
13C NMR (100 MHz, CDCl3) d 7.2, 19.0 (d, J¼23.0 Hz),
24.7, 27.5, 83.8, 97.7 (d, J¼195 Hz), 164.9 (d, J¼
25.2 Hz), 201.5 (d, J¼28.8 Hz); FABMS (mNBA) m/z 205
(M+1)+, 147 (Mꢁt-Bu)+, 148 (M+2ꢁt-Bu)+; [a]3D1 ꢁ55.5
(c 2.28, CH2Cl2) (69% ee); HPLC (DAICEL CHIRALPAK
AS, hexane/IPA¼150/1, 0.20 mL/min, 280 nm) tr (minor)¼
19.7 min, tr (major)¼22.3 min.
979, 696 cmꢁ1
.
4.2.3. tert-Butyl 1-fluoro-2-oxocyclohexanecarboxylate
(7c). 1H NMR (400 MHz, CDCl3) d 1.53 (s, 9H), 1.82–2.12
(m, 5H), 2.40–2.74 (m, 3H); 13C NMR (100 MHz, CDCl3)
d 21.2 (d, J¼6.6 Hz), 26.5, 27.8, 36.0 (d, J¼21.4 Hz), 39.9,
83.8, 96.3 (d, J¼195 Hz), 165.7 (d, J¼23.9 Hz), 202.2 (d,
J¼19.0 Hz); 19F NMR (400 MHz, CDCl3) d ꢁ159.6;
FABMS (mNBA) m/z 217 (M+1)+, 161 (M+2ꢁt-Bu)+;
[a]3D3 ꢁ88.6 (c 1.39, CHCl3) (94% ee); HPLC (DAICEL
CHIRALPAK AD-H, hexane/IPA¼99/1, 1.0 mL/min,
280 nm) tr (minor)¼14.0 min, tr (major)¼16.6 min; FAB–
HRMS (mNBA) Calcd for C11H18O3F (M+1)+ 217.1240.
Found 217.1243; IR (neat) n 2937, 2865, 1736, 1726, 1452,
4.2.8. tert-Butyl 2-fluoro-3-oxo-3-phenylpropionate (9).
1H NMR (400 MHz, CDCl3) d 1.42 (s, 9H), 5.75 (d, J¼
49.2 Hz, 1H), 7.45 (t, J¼7.6 Hz, 2H), 7.63 (t, J¼8.7 Hz,
1H), 8.02 (d, J¼8.5 Hz, 2H); 13C NMR (100 MHz, CDCl3)
d 27.7, 84.5, 90.1 (d, J¼195.8 Hz), 1287.7, 129.4 (d,
J¼3.3 Hz), 133.6, 134.2, 163.6, 190.0; HPLC (DAICEL
CHIRALPAK AD-H, hexane/IPA¼200/1, 0.40 mL/min,
254 nm) tr¼18.9 min, tr¼20.3 min.
1395, 1370, 1291, 1252, 1144,1095, 838 cmꢁ1
.
4.2.4. (R)-tert-Butyl 2-fluoro-1-oxoindane-2-carboxylate
(7d). H NMR (400 MHz, CDCl3) d 1.43 (s, 9H), 3.40 (dd,
1
4.3. General procedure for the catalytic enantioselective
fluorination of b-ketophosphonates
J¼17.6, 23.0 Hz, 1H), 3.73 (dd, J¼10.7, 17.6 Hz, 1H), 7.43–
7.50 (m, 2H), 7.67–7.71 (m, 1H), 7.83 (d, J¼7.6 Hz 1H); 13C
NMR (100 MHz, CDCl3) d 27.8, 38.3 (d, J¼23.9 Hz), 84.1,
95.4 (d, J¼201 Hz), 125.4, 126.4, 128.4, 133.6, 136.4, 150.9
(d, J¼3.3 Hz), 166.2 (d, J¼27.2 Hz), 195.8 (d, J¼18.1 Hz);
19F NMR (400 MHz, CDCl3) d ꢁ164.4; FABMS (mNBA)
m/z 251 (M+1)+, 195 (M+2Hꢁt-Bu)+; [a]3D4 +3.8 (c 0.86,
CHCl3) (83% ee); HPLC (DAICEL CHIRALPAK AD-H,
hexane/IPA¼150/1, 0.75 mL/min, 254 nm) tr (minor)¼
24.1 min, tr (major)¼33.7 min; FAB–HRMS (mNBA)
Calcd for C14H16O3F (M+1)+ 251.1084. Found 251.1083;
IR (neat) n 2980, 2927, 1758, 1725, 1466, 1395, 1370,
To a stirred solution of the Pd complex 1 (0.01 mmol) in
EtOH (0.2 mL), 11 (0.2 mmol) and NFSI (94.8 mg,
0.3 mmol) were added successively at ambient temperature.
The reaction was monitored by TLC (hexane/ethyl
acetate¼1/1). After completion of the reaction, saturated
aqueous NH4Cl was added for quenching. The aqueous layer
was extracted with ethyl acetate (3ꢃ10 mL). The combined
organic layers were washed with brine and dried over
Na2SO4. After evaporation of the solvent, the obtained crude
product was purified by flash column chromatography
(hexane/ethyl acetate¼1/1) to afford 12 as a colorless oil.
The ees of the products were determined by chiral HPLC
analysis. In this study, coupling constants against protons
and phosphorus atoms were recorded in 19F NMR.
1214, 1151, 1073, 922, 834, 741 cmꢁ1
.
4.2.5. tert-Butyl 2-fluoro-2-methyl-3-oxobutyrate (7e). 1H
NMR (400 MHz, CDCl3) d 1.48 (s, 9H), 1.63 (d, J¼22.2 Hz,

Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
7175
4.3.1. Diethyl 1-fluoro-2-oxocyclopentylphosphonate
(12a). H NMR (400 MHz, CDCl3) d 1.36 (t, J¼7.1 Hz,
(dd, J¼3.3, 14.4 Hz); 19F NMR (376 MHz, CDCl3) d
ꢁ93.9 (ddd, J¼6.8, 13.9, 83.1 Hz); FAB–LRMS (mNBA)
m/z 301 (M+1)+; [a]D27 +49.1 (c 1.1, CHCl3) (94% ee);
HPLC (DAICEL CHIRALPAK AD-H, n-hexane/IPA¼9/1,
1.0 mL/min, 254 nm) tr (major)¼10.2 min, tr (minor)¼
13.6 min; FAB–HRMS (mNBA) Calcd for C14H19FO4P
(M+1)+ 301.1005. Found 301.1000; IR (neat) n 1694,
1
3H), 1.39 (t, J¼7.1 Hz, 3H), 2.03–2.56 (m, 5H), 2.68–2.81
(m, 1H), 4.19–4.30 (m, 4H); 13C NMR (100 MHz, CDCl3)
d 16.2 (d, J¼5.7 Hz), 16.3 (d, J¼5.7 Hz), 16.8 (dd, J¼4.2,
5.8 Hz), 32.0 (dd, J¼2.5, 18.1 Hz), 35.4 (d, J¼2.5 Hz),
64.0 (d, J¼6.2 Hz), 64.1 (d, J¼6.2 Hz), 96.2 (dd, J¼160.0,
200.8 Hz), 209.0 (dd, J¼3.3, 13.2 Hz); 19F NMR
(376 MHz, CDCl3) d ꢁ98.1 (ddd, J¼11.7, 25.2, 85.0 Hz);
FAB–LRMS (mNBA) m/z 239 (M+1)+; [a]D28 +130.2 (c
0.75, CHCl3) (96% ee); HPLC (DAICEL CHIRALPAK
AD-H, n-hexane/IPA¼95/5, 1.0 mL/min, 280 nm) tr
(major)¼10.6 min, tr (minor)¼11.7 min; FAB–HRMS
(mNBA) Calcd for C9H17FO4P (M+1)+ 239.0849. Found
1259, 1014, cmꢁ1
.
4.3.5. Diethyl 2-fluoro-3-oxobutan-2-ylphosphonate
(12e). 1H NMR (400 MHz, CDCl3) d 1.36 (dt, J¼0.48,
7.1 Hz, 3H), 1.37 (t, J¼0.48, 7.1 Hz, 3H), 1.71 (dd,
J¼15.4, 23.7 Hz, 3H), 2.38 (dd, J¼0.72, 5.6 Hz, 3H),
4.18–4.28 (m, 4H); 13C NMR (100 MHz, CDCl3) d 16.3
(d, J¼5.8 Hz), 16.3 (d, J¼5.7 Hz), 19.3 (d, J¼21.4 Hz),
26.2, 64.0 (d, J¼7.4 Hz), 64.2 (d, J¼7.5 Hz), 99.1 (dd,
J¼159.6, 190.0 Hz), 205.3 (dd, J¼3.3, 25.5 Hz); 19F NMR
(376 MHz, CDCl3) d 93.6 (qqd, J¼4.5, 25.0, 84.5 Hz);
FAB–LRMS (mNBA) m/z 227 [M+1]+; [a]D28 ꢁ80.8
(c 0.47, CHCl3) (94% ee); IR (neat) n 1724, 1261,
1012 cmꢁ1. The ee was determined after conversion to the
corresponding 2,4-dinitrophenylhydrazone 13.27
239.0855; IR (neat) n 1756, 1259, 1047, 1022 cmꢁ1
.
4.3.2. Diethyl 1-fluoro-2-oxocyclohexylphosphonate
1
(12b). H NMR (400 MHz, CDCl3) d 1.32 (t, J¼7.1 Hz,
3H), 1.37 (t, J¼7.1 Hz, 3H), 1.64–1.72 (m, 1H), 1.85–2.23
(m, 4H), 2.58–2.70 (m, 2H), 2.85–2.93 (m, 1H), 4.11–4.30
(m, 4H); 13C NMR (100 MHz, CDCl3) d 16.3 (d,
J¼6.6 Hz), 16.4 (d, J¼5.7 Hz), 21.7 (d, J¼8.2 Hz), 26.7,
36.1 (d, J¼19.0 Hz), 40.8, 64.1 (d, J¼6.5 Hz), 64.4 (d,
J¼6.6 Hz), 98.1 (dd, J¼154.7, 196.6 Hz), 203.0 (dd,
J¼4.9, 14.0 Hz); 19F NMR (376 MHz, CDCl3) d ꢁ91.0
(dd, J¼18.4, 80.1 Hz); FAB–LRMS (mNBA) m/z 253
(M+1)+; [a]D28 +164.2 (c 1.0, CHCl3) (96% ee); HPLC
(DAICEL CHIRALPAK AD-H, n-hexane/IPA¼99/1,
1.0 mL/min, 280 nm) tr (major)¼47.7 min, tr (minor)¼
50.4 min; FAB–HRMS (mNBA) Calcd for C10H19FO4P
(M+1)+ 253.1005. Found 253.1009; IR (neat) n 1726,
1
Compound 13: yellow solid; H NMR (400 MHz, CDCl3)
d 1.33–1.38 (m, 1H), 1.90 (dd, J¼13.9, 24.6 Hz, 3H), 2.24
(dd, J¼1.2, 2.4 Hz, 3H), 4.16–4.28 (m, 4H), 7.95 (d,
J¼9.5 Hz, 1H), 8.34–8.37 (m, 1H), 9.15 (d, J¼2.7 Hz,
1H), 11.14 (s, 1H); FAB–LRMS (mNBA) m/z 407 (M+1)+;
HPLC (DAICEL CHIRALPAK AD-H, n-hexane/IPA¼9/1,
1.0 mL/min, 254 nm) tr (major)¼21.0 min, tr (minor)¼
25.6 min; FAB–HRMS (mNBA) Calcd for C14H21FN4O7P
(M+1)+ 407. 1132. Found 407.1131.
1257, 1013 cmꢁ1
.
4.3.3. Diethyl 2-fluoro-2,3-dihydro-1-oxo-1H-inden-2-yl-
2-phosphonate (12c). H NMR (400 MHz, CDCl3) d 1.22
4.3.6. Diethyl 2-fluoro-1-oxo-1-phenylpropan-2-yl-
phosphonate (12f). H NMR (400 MHz, CDCl3) d 1.27 (t,
1
1
(t, J¼7.1 Hz, 3H), 1.38 (t, J¼7.1 Hz, 3H), 3.97 (ddd,
J¼9.3, 14.6, 17.9 Hz, 1H), 3.35–3.48 (m, 1H), 4.12–4.25
(m, 2H), 4.30 (quint, J¼7.3 Hz, 2H), 7.42–7.48 (m, 2H),
7.67 (td, J¼1.3, 7.7 Hz, 1H), 7.81 (d, J¼7.6 Hz, 1H); 13C
NMR (100 MHz, CDCl3) d 16.2 (d, J¼5.8 Hz), 16.3 (d,
J¼5.7 Hz), 36.5 (dd, J¼3.3, 21.4 Hz), 64.3 (d, J¼6.6 Hz),
95.9 (dd, J¼162.9, 199.9 Hz), 125.1, 126.4, 128.5, 134.1
(d, J¼3.3 Hz), 136.4, 149.8 (t, J¼4.6 Hz), 196.4 (dd,
J¼3.3, 14.8 Hz); 19F NMR (376 MHz, CDCl3) d ꢁ101.1
(ddd, J¼9.0, 26.4, 104.9 Hz); FAB–LRMS (mNBA) m/z
287 (M+1)+; [a]D28 +71.2 (c 0.85, CHCl3) (95% ee); HPLC
(DAICEL CHIRALPAK AD-H, n-hexane/IPA¼95/5,
1.0 mL/min, 254 nm) tr (major)¼19.5 min, tr (minor)¼
24.3 min; FAB–HRMS (mNBA) Calcd for C13H17FO4P
(M+1)+ 287.0848. Found 287.0852; IR (neat) n 1725,
J¼7.1 Hz, 3H), 1.29 (t, J¼7.1 Hz, 3H), 1.85 (dd, J¼15.2,
24.2 Hz, 3H), 4.10–4.26 (m, 4H), 7.38 (d, J¼7.7 Hz, 2H),
7.48–7.52 (m, 1H), 7.99–8.02 (m, 2H); 13C NMR
(100 MHz, CDCl3) d 16.4 (d, J¼4.9 Hz), 16.4 (d,
J¼4.9 Hz), 21.5 (d, J¼22.2 Hz), 64.1 (d, J¼4.1 Hz), 64.2
(d, J¼4.1 Hz), 100.5 (dd, J¼161.7, 193.7 Hz), 128.2,
130.0 (d, J¼7.4 Hz), 133.4, 134.7 (d, J¼3.3 Hz), 197.6
(dd, J¼4.1, 23.0 Hz); 19F NMR (376 MHz, CDCl3) d
ꢁ85.3 (ddd, J¼24.1, 48.1, 83.8 Hz); FAB–LRMS (mNBA)
m/z 289 (M+1)+; [a]D25 ꢁ36.3 (c 0.37, CHCl3) (90% ee);
HPLC (DAICEL CHIRALPAK AD-H, n-hexane/IPA¼9/1,
1.0 mL/min, 254 nm) tr (minor)¼7.3 min, tr (major)¼
7.9 min; FAB–HRMS (mNBA) Calcd for C13H19FO4P
(M+1)+ 289.1005. Found 289.1011; IR (neat) n 1681,
1260, 1048, 1014 cmꢁ1
.
1260, 1018 cmꢁ1
.
4.4. Synthesis of 16
4.3.4. (S)-Diethyl 2-fluoro-1,2,3,4-tetrahydro-1-oxonaph-
thalen-2-yl-2-phosphonate (12d). 1H NMR (400 MHz,
CDCl3) d 1.12 (t, J¼7.1 Hz, 3H), 1.37 (t, J¼7.1 Hz, 3H),
2.44–2.64 (m, 1H), 2.81–2.91 (m, 1H), 3.06–3.11 (m, 1H),
3.44–3.53 (m, 1H), 4.00–4.16 (m, 2H), 4.25–4.34 (m, 2H),
7.26 (d, J¼7.8 Hz, 1H), 7.34 (d, J¼7.6 Hz, 1H), 7.53 (td,
J¼1.5, 7.6 Hz, 1H), 8.06 (dd, J¼1.2, 7.8 Hz, 1H); 13C
NMR (100 MHz, CDCl3) d 16.0 (d, J¼6.6 Hz), 16.3 (d,
J¼4.9 Hz), 26.0 (d, J¼10.7 Hz), 31.6 (d, J¼19.7 Hz), 63.8
(d, J¼6.6 Hz), 64.6 (d, J¼6.6 Hz), 95.5 (dd, J¼156.3,
192.5 Hz), 126.9, 127.9, 128.6, 131.1, 134.2, 143.1, 190.6
To a stirred solution of NaBH4 (50 mg, 1.33 mmol) in EtOH
(3 mL) was added a solution of the optically active 12d
(100 mg, 0.33 mmol, 95% ee) in EtOH (7 mL) under ice-
bath cooling. The mixture was stirred at room temperature
for 5 h, then saturated aqueous NH4Cl was added to destroy
the excess reagent. The aqueous layer was extracted with
ether several times and the combined organic layers were
washed with brine and dried over Na2SO4. Short column
chromatography on silica gel (hexane/ethyl acetate¼1/3–
1/20) afforded the desired product in 90% yield (90 mg)

7176
Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
as a colorless oil. This reaction was stereoselective and no
stereoisomer was detected in 1H NMR of the crude products.
Subsequently, the obtained alcohol (39 mg, 0.13 mmol) was
mixed with the acid 15 (110 mg, 0.26 mmol) and DMAP
(16 mg, 0.13 mmol) in CH2Cl2 (5 mL). After 5 min,
EDAC (97.4 mg, 0.52 mmol) was added under ice-bath
cooling, and the resulting mixture was stirred for 17 h at
room temperature. After completion of the reaction, 1 N
HCl was added and the aqueous layer was extracted with
ethyl acetate. The combined organic layers were washed
with saturated aqueous NaHCO3, water, and brine, then
dried over Na2SO4. Removal of the solvent, followed by
flash column chromatography (hexane/ethyl acetate¼1/1)
afforded 16 as a white solid in 84% yield. Recrystallization
from hexane/ethyl acetate (3/1) at ambient temperature gave
colorless pillars in 76% yield. 1H NMR (400 MHz, CDCl3)
d 0.98 (s, 3H), 1.21 (t, J¼7.1 Hz, 3H), 1.25 (s, 3H), 1.26 (t,
J¼7.1 Hz, 3H), 1.33–1.48 (m, 2H), 1.87–1.97 (m, 3H),
1.98–2.15 (m, 1H), 2.35–2.7 (m, 3H), 2.97–3.14 (m, 2H),
3.34 (d, J¼13.9 Hz, 1H), 3.43 (d, J¼13.9 Hz, 1H), 3.88
(m, 1H), 4.05–4.25 (m, 4H), 6.47 (t, J¼9.5 Hz, 1H), 7.16
(d, J¼7.6 Hz, 1H), 7.19 (t, J¼7.6 Hz, 1H), 7.27 (dt, J¼1.5,
7.5 Hz, 1H), 7.44 (d, J¼7.5 Hz, 1H), 7.53 (s, 1H), 8.15 (s,
1H); 13C NMR (100 MHz, CDCl3) d 16.2 (d, J¼5.7 Hz),
1.63 (d, J¼5.8 Hz), 20.0, 20.8, 24.3 (d, J¼7.5 Hz), 26.1 (d,
J¼20.6 Hz), 26.5, 33.0, 37.5, 44.7, 47.7, 48.4, 53.0, 63.3
(d, J¼6.6 Hz), 63.7 (d, J¼5.7 Hz), 65.5, 71.0 (d,
J¼31.3 Hz), 77.2, 93.7 (d, J¼175, 182 Hz), 126.6, 127.6,
128.2, 128.7, 130.0, 131.1 (d, J¼7.4 Hz), 132.4, 134.5,
135.3, 135.6, 137.0, 162.0, 164.9; [a]2D7 ꢁ134.65 (c 0.73,
CHCl3).
in THF, 0.38 mmol) at 0 ꢀC. Saturated aqueous NH4Cl
(3 mL) was added after 20 min. The aqueous layer was
extracted with ether (5ꢃ10 mL). The combined organic
layers were dried over Na2SO4. The solvent was removed
under reduced pressure and the residue was purified by flash
column chromatography (hexane/ethyl acetate¼12/1). The
desired product 20 was obtained in 83% yield as a colorless
oil. The diastereoselectivity of this reaction was found to be
1
1
more than 95% by H NMR. H NMR (400 MHz, CDCl3)
d 1.39 (d, J¼21.7 Hz, 3H), 3.83 (s, 3H), 4.94 (d,
J¼20.6 Hz, 1H), 7.32–7.39 (m, 5H); 13C NMR (100 MHz,
CDCl3) d 20.4 (d, J¼23.0 Hz), 52.8, 77.4 (d, J¼19.7 Hz),
96.6 (d, J¼192.5 Hz), 127.8, 127.8, 128.4, 128.7, 137.4,
171.6 (d, J¼24.7 Hz); FABMS (mNBA) m/z 213 (M+1)+,
212 (M)+; [a]D30 +22.9 (c 1.45, CHCl3); FAB–HRMS
(mNBA) Calcd for C11H13O3F (M)+ 212.0849. Found
212.0844; IR (solid) n 3483, 2992, 2957, 1741, 1453,
1375, 1295, 1188, 1135, 1113, 1053, 710 cmꢁ1
.
4.5.3. tert-Butyl (1S,2R)-2-(methoxycarbonyl)-2-fluoro-
1-phenylpropylcarbamate (21). The alcohol 19 (25 mg,
0.118 mmol) was dissolved in THF (0.3 mL). To this solu-
tion were added Ph3P (49 mg, 0.19 mmol), DEAD (82 mL,
40% in toluene), and DPPA (33 mL, 0.15 mmol) succes-
sively in this order. After stirring for 2 h at room tempera-
ture, saturated aqueous NH4Cl was added, and the
resulting mixture was stirred for 5 min. The aqueous layer
was extracted with ether (3ꢃ10 mL) and the combined
organic layers were washed with water and brine, and dried
over Na2SO4. Further purification was performed by flash
column chromatography (hexane/ethyl acetate¼15/1) to
1
give the azide in 95% yield (26.7 mg). Azide: H NMR
4.5. Conversion of the fluorinated products
(400 MHz, CDCl3) d 1.36 (d, J¼21.2 Hz, 3H), 3.88 (s,
3H), 4.85 (d, J¼24.7 Hz, 1H), 7.37–7.48 (m, 5H); 13C
NMR (100 MHz, CDCl3) d 20.8 (d, J¼23.0 Hz), 53.0, 69.4
(d, J¼18.9 Hz), 95.7 (d, J¼197.5 Hz), 128.8, 129.1, 129.1,
129.4, 132.9, 170.8 (d, J¼25.5 Hz).
4.5.1. (2R,3R)-Methyl 2-fluoro-3-hydroxy-2-methyl-3-
phenylpropanoate (19). To a stirred solution of 14
(125.8 mg, 0.598 mmol, 91% ee) in TFA (2 mL) was added
Et3SiH (0.3 mL, 1.79 mmol) at 0 ꢀC. The reaction mixture
was stirred at room temperature for 2 h. The mixture was
diluted with ether (5 mL) and saturated aqueous NaHCO3
was added under ice-bath cooling to neutralize the mixture.
The separated aqueous layer was extracted with ether
(2ꢃ10 mL). The combined organic layers were washed
with water and brine. Concentration, followed by flash col-
umn chromatography (hexane/ethyl acetate¼8/1) afforded
the desired product 19 as a colorless oil in 92% yield
To a solution of this azide (160 mg, 0.598 mmol) in MeOH
(8 mL) were added (Boc)2O (156.8 mg, 0.72 mmol) and
10% Pd–C (63 mg). The reaction mixture was stirred under
a hydrogen atmosphere (balloon) for 1 h at room tempera-
ture, then passed through Celite to remove Pd–C, and the
residue was washed with CH2Cl2. After the removal of
solvent, the crude product was purified by flash column
chromatography (hexane/ethyl acetate¼12/1) to give the
N-protected b-amino ester 21 in 78% yield (146 mg) as a
white solid. Recrystallization from ethyl acetate enhanced
the enantioselectivity to >99%. 1H NMR (400 MHz,
CDCl3) d 1.38 (d, J¼21.8 Hz, 3H), 1.39 (s, 9H), 3.80
(s, 3H), 5.07 (dd, J¼9.6, 25.9 Hz, 1H), 5.55 (d, J¼
9.6 Hz, 1H), 7.29–7.38 (m, 5H); 13C NMR (100 MHz,
CDCl3) d 21.5 (d, J¼23.1 Hz), 28.2, 52.8, 59.3 (d,
J¼19.0 Hz), 80.0, 96.5 (d, J¼190.1 Hz), 128.3, 128.3,
128.5, 136.4, 154.6, 171.1 (d, J¼26.3 Hz); FABMS
(mNBA) m/z 312 (M+1)+; [a]D30 +13.8 (c 0.73, CHCl3);
HPLC (DAICEL CHIRALCEl OJ-H, n-hexane/IPA¼9/1,
1.0 mL/min, 254 nm) tr (minor)¼7.2 min, tr (major)¼
13.5 min; FAB–HRMS (mNBA) Calcd for C16H23NO3F
(M+1)+ 312.1611. Found 312.1616; IR (solid) n 3382,
2972, 2935, 1754, 1693, 1513, 1453, 1388, 1367, 1324,
1311, 1264, 1242, 1159, 1113, 1042, 1015, 980, 948,
1
(117.1 mg). H NMR of the crude products indicated high
diastereoselectivity (>95/5). H NMR (400 MHz, CDCl3)
1
d 1.60 (d, J¼22.2 Hz, 3H), 3.70 (s, 3H), 4.98 (d,
J¼15.6 Hz, 1H), 7.30–7.44 (m, 5H); 13C NMR (100 MHz,
CDCl3) d 19.3 (d, J¼11.5 Hz), 52.6, 76.2 (d, J¼23.0 Hz),
96.2 (d, J¼190.1 Hz), 127.3, 127.4, 128.2, 128.6, 137.5,
171.3 (d, J¼23.9 Hz); FABMS (mNBA) m/z 213 (M+1)+,
212 (M)+; [a]D30 +2.68 (c 2.44, CHCl3); FAB–HRMS
(mNBA) Calcd for C11H14O3F (M+1)+ 213.0927. Found
213.0922. Calcd for C11H13O3F (M)+ 212.0849. Found
212.0844; IR (solid) n 3465, 3001, 2953, 1737, 1452,
1375, 1291, 1182, 1109, 1051, 1027, 727, 700 cmꢁ1
.
4.5.2. (2R,3S)-Methyl 2-fluoro-3-hydroxy-2-methyl-3-
phenylpropanoate (20). To a stirred solution of 14
(40 mg, 0.19 mmol, 91% ee) in DMF (0.2 mL) were added
PhMe2SiH (120 mL, 0.76 mmol) and TBAF (380 mL, 1 M
910 cmꢁ1
.

Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
7177
1
4.5.4. tert-Butyl (1R,2R)-2-(methoxycarbonyl)-2-fluoro-
1-phenylpropylcarbamate (22). According to the same
procedure described above, 20 was converted to 22. Azide:
1H NMR (400 MHz, CDCl3) d 1.76 (d, J¼21.5 Hz, 3H),
3.63 (s, 3H), 4.80 (d, J¼30.0 Hz, 1H), 7.32–7.43 (m, 5H);
13C NMR (100 MHz, CDCl3) d 20.7 (d, J¼23.8 Hz), 52.6,
68.7 (d, J¼20.6 Hz), 95.9 (d, J¼195.0 Hz), 128.6, 128.6,
128.6, 129.2, 133.7, 169.9 (d, J¼23.8 Hz). Compound 22:
1H NMR (400 MHz, CDCl3) d 1.43 (s, 9H), 1.70 (d,
J¼21.7 Hz, 3H), 3.54 (s, 3H), 5.11 (dd, J¼10.0, 27.1 Hz,
1H), 5.31 (d, J¼10.0 Hz, 1H), 7.26–7.39 (m, 5H); 13C
NMR (100 MHz, CDCl3) d 21.4 (d, J¼23.9 Hz), 28.3,
52.3, 58.6 (d, J¼17.3 Hz), 80.2, 97.2 (d, J¼192.5 Hz),
127.8, 128.2, 128.5, 135.8, 137.4, 155.3, 170.3 (d,
J¼23.9 Hz); FABMS (mNBA) m/z 312 (M+1)+, 311 (M)+;
[a]2D9 ꢁ30.5 (c 0.51, CHCl3); FAB–HRMS (mNBA) Calcd
for C12H15NO4F (M+2Hꢁt-Bu)+ 256.0985. Found
256.0990. Calcd for C12H15NO4F (M+2HꢁBoc)+
212.1087. Found 212.1090; IR (solid) n 3329, 2977, 2931,
1766, 1702, 1493, 1453, 1367, 1276, 1246, 1166, 1046,
minor: 8%). H NMR (400 MHz, CDCl3) d 1.25–1.40 (m,
1H), 1.45–1.78 (m, 5H), 1.51 (s, 9H), 1.80–1.88 (m, 1H),
1.99–2.11 (m, 1H), 2.95 (ddd, J¼4.6, 8.6, 12.7 Hz, 1H),
3.83 (d, J¼13.2 Hz, 1H), 3.87 (d, J¼13.2 Hz, 1H), 7.20–7.34
(m, 5H); 13C NMR (100 MHz, CDCl3) d 21.7 (d, J¼7.4 Hz),
22.1, 27.6 (d, J¼5.0 Hz), 28.1, 31.6 (d, J¼21.4 Hz), 52.0,
59.8 (d, J¼21.4 Hz), 82.0, 96.9 (d, J¼183 Hz), 126.7,
128.0, 128.3, 140.7, 169.7 (d, J¼26.3 Hz); FAB–LRMS
(mNBA) m/z 308 (M+1)+, 252 (M+2Hꢁt-Bu)+; [a]2D6
ꢁ8.92 (c 1.44, CHCl3). FAB–HRMS (mNBA) Calcd for
C18H27NO2 F (M+1)+ 308.2026. Found 308.2016; Minor:
1H NMR (400 MHz, CDCl3) d 1.51 (s, 9H), 1.55–2.10
(m, 8H), 2.90 (ddd, J¼4.4, 11.4, 27.6 Hz, 1H), 3.72 (d,
J¼13.2 Hz, 1H), 3.91 (d, J¼13.2 Hz, 1H), 7.19–7.34
(m, 5H).
4.5.7. (S)-2-Fluoro-1,2,3,4-tetrahydro-1-oxonaphthalen-
2-yl-2-phosphonic acid (25). TMSBr (354 mL, 2.68 mmol)
was added to a solution of 12d (134 mg, 0.447 mmol, 95%
ee) in CH2Cl2 (5 mL) under ice-bath cooling. The resulting
mixture was stirred for 12 h at room temperature. After evap-
oration, MeOH (2 mL) was added and the mixture was
stirred for an additional 2 h. Gradual removal of the solvent
by leaving this solution still at ambient temperature gave
a brownish solid. Washing this solid with CHCl3 afforded
pure 25 in 97% yield (106 mg). 1H NMR (400 MHz,
CD3OD) d 2.30–2.49 (m, 1H), 2.62–2.75 (m, 1H), 3.01 (dt,
J¼2.8, 17.3 Hz, 1H), 3.43 (ddd, J¼4.9, 12.7, 21.0 Hz, 1H),
7.24 (d, J¼7.5 Hz, 1H), 7.25 (t, J¼7.8 Hz, 1H), 7.46 (dt,
J¼1.2, 7.5 Hz, 1H), 7.89 (dd, J¼1.2, 8.1 Hz, 1H);
13C NMR (100 MHz, CD3OD) 27.2 (d, J¼10.7 Hz), 32.9
(d, J¼10.2 Hz), 96.8 (dd, J¼153, 191 Hz), 127.8, 128.5 (d,
J¼1.7 Hz), 130.0, 132.9, 135.4, 145.3, 193.7 (dd, J¼3.2,
13.1 Hz); [a]2D7 58.9 (c 1.01, EtOH).
1021, 982, 947, 882, 755, 703, 576 cmꢁ1
.
4.5.5. tert-Butyl 1-fluoro-2-hydroxycyclohexanecarboxy-
late (23). To a stirred solution of 7c (22 mg, 0.101 mmol,
94% ee) in EtOH (1 mL) was added NaBH4 (9 mg,
0.24 mmol) at ꢁ78 ꢀC. The reaction mixture was concen-
trated and the residue was diluted with ethyl acetate
(10 mL). The organic layer was washed with water and
brine. Further purification was performed by flash column
chromatography (hexane/ethyl acetate¼7/1) to give 23 in
1
81% yield (major: 76%, minor: 5%). H NMR (400 MHz,
CDCl3) d 1.33–1.46 (m, 1H), 1.52 (s, 9H), 1.58–1.90 (m,
6H), 2.09 (dddd, J¼5.6, 8.3, 13.9, 27.6 Hz, 1H), 3.08 (br s,
1H), 3.91–3.96 (ddd, J¼3.9, 6.6, 10.0 Hz, 1H); 13C NMR
(100 MHz, CDCl3) d 20.1, 20.9 (J¼4.1 Hz), 28.0, 28.8
(J¼2.5 Hz), 29.7 (J¼20.6 Hz), 70.1 (J¼26.3 Hz), 28.8
(J¼2.5 Hz), 83.2, 94.4 (J¼185.9 Hz), 170.7 (J¼23.8 Hz);
[a]2D7 ꢁ18.7 (c 2.1, CHCl3); FABMS (mNBA) m/z 219
(M+1)+, 163 (M+2ꢁt-Bu)+; FAB–HRMS (mNBA) Calcd
for C11H18FO3 (M)+ 218.1318. Found 218.1318. Minor iso-
4.6. X-ray structural analysis of 16
Molecular formula: C32H37Cl2FNO8PS; molecular wꢀeight:
˚
716.56; unit-cell dimensions: a¼11.015(3) A, a¼90 , b¼
ꢀ
ꢀ
˚
˚
1
14.875(3) A, b¼98.092(1) , c¼21.073(5) A, g¼90 ; U¼
mer: H NMR (400 MHz, CDCl3) d 1.30–1.94 (m, 16H),
3
3
˚
3452.6(15) A , Z¼4, d¼1.379 g/cm ; crystal system: ortho-
rhombic; space group: P 21 21 21; type of diffractometer:
Rigaku R-AXIS-CS, crystal size, 0.50ꢃ0.25ꢃ0.21 mm3;
temperature: 296 K; theta range: 3.01–24.69ꢀ; reflections
collected: 29,415; independent reflections: 5830 [R(int)¼
0.0379]; absorption coefficient: 0.350 mmꢁ1; max and min
transmission: 0.9301 and 0.7093; solution method: direct;
refinement method: full-matrix least-squares on F2; good-
ness of fit indicator: 1.206, R1¼0.0511, wR2¼0.1014.
Crystallographic data for the structure reported in this paper
have been deposited with Cambridge Crystallographic Data
Centre as supplementary publication no. CCDC 288122.
Copies of the data can be obtained free of charge on applica-
tion to CCDC, 12 Union Road, Cambridge CB21EZ, UK
[Fax: int. code +44 1223 336 033; E-mail: deposit@ccdc.
cam.ac.uk].
2.01–2.09 (m, 1H), 3.87 (ddd, J¼4.6, 9.5, 24.0 Hz, 1H);
13C NMR (100 MHz, CDCl3) d 19.8, 23.7, 28.0, 30.2, 32.2
(J¼22.3 Hz), 72.0 (J¼20.6 Hz), 82.6, 96.0 (J¼187.6 Hz),
169.9 (J¼26.3 Hz); FABMS (mNBA) m/z 219 (M+1)+,
163 (M+2ꢁt-Bu)+; [a]D27 17.9 (c 0.25, CH3CN).
4.5.6. tert-Butyl 2-benzylamino-1-fluorocyclohexanecar-
boxylate (24). To a solution of 7c (26 mg, 0.12 mmol, 94%
ee) in toluene (2 mL) was added BnNH2 (14 mL,
0.132 mmol) at room temperature. The reaction mixture
was stirred under reflux in the presence of a catalytic amount
of TsOH (5 mg, 0.026 mmol) using MS4A as a dehydrating
agent. After 4 h, the solvent was removed under reduced
pressure. To this crude mixture were added AcOH (0.5 mL)
and NaCNBH3 (25 mg, 0.4 mmol) at room temperature.
The mixture was stirred for an additional 2 h. After dilution
with ether (10 mL), 1 N NaOH was added until the reaction
mixture reached pH 9. The aqueous layer was extracted
with ether (2ꢃ10 mL) and the combined organic layers
werewashed with water and brine. Evaporation of the solvent
and flash column chromatography (hexane/ether¼20/1) of
the residue afforded the desired product 23 (major: 65%,
Acknowledgements
This work was supported in part by a Grant-in Aid for
Encouragement of Young Scientists (B) from J.S.P.S. We

7178
Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
´
10. (a) Hamashima, Y.; Yagi, K.; Takano, H.; Tamas, L.; Sodeoka,
M. J. Am. Chem. Soc. 2002, 124, 14530–14531; (b)
Hamashima, Y.; Takano, H.; Hotta, D.; Sodeoka, M. Org.
Lett. 2003, 5, 3225–3228; (c) Hamashima, Y.; Suzuki, T.;
Shimura, Y.; Shimizu, T.; Umebayashi, N.; Tamura, T.;
Sasamoto, N.; Sodeoka, M. Tetrahedron Lett. 2005, 46,
1447–1450.
thank Dr. D. Hashizume of RIKEN for X-ray analysis, and
Ms. K. Harata of RIKEN for MS measurements. We also
thank Dr. Takao Saito of Takasago International Corp. for
a generous gift of chiral phosphine ligands.
References and notes
11. For b-ketoesters: (a) Ma, J.-A.; Cahard, D. Tetrahedron:
Asymmetry 2004, 15, 1007–1011; (b) Shibata, N.; Ishimaru,
T.; Nagai, T.; Kohno, J.; Toru, T. Synlett 2004, 1703–1706;
(c) Shibata, N.; Kohno, J.; Takai, K.; Ishimaru, T.;
Nakamura, S.; Toru, T.; Kanemasa, S. Angew. Chem., Int. Ed.
2005, 44, 4204–4207; For cyanoesters: (d) Kim, H. R.; Kim,
D. Y. Tetrahedron Lett. 2005, 46, 3115–3117; For b-keto-
phosphonates: (e) Bernardi, L.; Jørgensen, K. A. Chem.
Commun. 2005, 1324–1326; (f) Kim, S. M.; Kim, H. R.;
Kim, D. Y. Org. Lett. 2005, 7, 2309–2311.
1. (a) Kirsch, P. Modern Fluoroorganic Chemistry: Synthesis,
Reactivity, Applications; Wiley-VCH: Weinheim, 2004; (b)
Hiyama, T.; Kanie, K.; Kusumoto, T.; Morizawa, Y.;
Shimizu, M. Organofluorine Compounds: Chemistry and
Applications; Springer: Berlin, 2000; (c) Biomedical
Frontiers of Fluorine Chemistry; Ojima, I., McCarthy, J. R.,
Welch, J. T., Eds.; ACS Symposium Series 639; American
Chemical Society: Washington, DC, 1996; For reviews:
(d) O’Hagan, D.; Rzepa, H. S. Chem. Commun. 1997,
645–652; (e) Smart, B. E. J. Fluorine Chem. 2001, 109,
3–11; (f) Ismail, F. M. D. J. Fluorine Chem. 2002, 118,
27–33.
2. (a) Asymmetric Fluoroorganic Chemistry: Synthesis,
Application, and Future Directions; Ramachandran, P. V.,
Ed.; ACS Symposium Series 746; American Chemical
Society: Washington, DC, 2000; (b) Enantiocontrolled
Synthesis of Fluoro-Organic Compounds: Stereochemical
Challenge and Biomedicinal Targets; Soloshonok, V. A., Ed.;
Wiley: New York, NY, 1999.
3. (a) Mikami, K.; Itoh, Y.; Yamanaka, M. Chem. Rev. 2004, 104,
1–16; (b) Iseki, K. Tetrahedron 1998, 54, 13887–13914.
4. Reviews of enantioselective fluorination reactions: (a) Ibrahim,
H.; Togni, A. Chem. Commun. 2004, 1147–1155; (b) Ma, J.-A.;
Cahard, D. Chem. Rev. 2004, 104, 6119–6146; For a review
of catalytic asymmetric a-halogenation reactions: (c) France,
S.; Weatherwax, A.; Lectka, T. Eur. J. Org. Chem. 2005,
475–479.
5. For representative examples: (a) Iwaoka, T.; Murohashi, T.;
Sato, M.; Kaneko, C. Tetrahedron: Asymmetry 1992, 3,
1025–1028; (b) Davis, F. A.; Kasu, P. V. N. Tetrahedron Lett.
1998, 39, 6135–6138; (c) Enders, D.; Faure, S.; Potthoff, M.;
Runsink, J. Synthesis 2001, 15, 2307–2319.
6. (a) Differding, E.; Lang, R. E. Tetrahedron Lett. 1988, 29,
6087–6090; (b) Davis, F. A.; Zhou, P.; Murphy, C. K.;
Sundarababu, G.; Qi, H.; Han, W.; Przeslawski, R. M.; Chen,
B.-C.; Carroll, P. J. Org. Chem. 1998, 63, 2273–2280; (c)
Liu, Z.; Shibata, N.; Takeuchi, Y. J. Org. Chem. 2000, 65,
7583–7587.
7. (a) Shibata, N.; Suzuki, E.; Asahi, T.; Shiro, M. J. Am. Chem.
Soc. 2001, 123, 7001–7009; (b) Mohar, B.; Baudoux, J.;
Plaquevent, J.-C.; Cahard, D. Angew. Chem., Int. Ed. 2001,
40, 4214–4216; (c) Shibata, N.; Ishimaru, T.; Suzuki, E.;
Kirk, K. L. J. Org. Chem. 2003, 68, 2494–2497; (d) Zoute,
L.; Audouard, C.; Plaquevent, J.-C.; Cahard, D. Org. Biomol.
Chem. 2003, 1, 1833–1834; (e) Greedy, B.; Paris, J.-M.;
Vidal, T.; Gouverneur, V. Angew. Chem., Int. Ed. 2003, 42,
3291–3294; (f) Thierry, B.; Audouard, C.; Plaquevent, J.-C.;
Cahard, D. Synlett 2004, 856–860.
12. (a) Enders, D.; Huttl, M. R. M. Synlett 2005, 991–993; (b)
¨
Marigo, M.; Fielenbach, D.; Braunton, A.; Kjærsgaard, A.;
Jørgensen, K. A. Angew. Chem., Int. Ed. 2005, 44, 3703–
3706; (c) Steiner, D. D.; Mase, N.; Barbas, C. F., III. Angew.
Chem., Int. Ed. 2005, 44, 3706–3710; (d) Beeson, T. D.;
MacMillan, D. W. C. J. Am. Chem. Soc. 2005, 127, 8826–
8828; (e) Huang, Y.; Walji, A. M.; Larsen, C. H.; MacMillan,
D. W. C. J. Am. Chem. Soc. 2005, 127, 15051–15053.
13. For catalytic enantioselective monofluorination of oxindoles:
Hamashima, Y.; Suzuki, T.; Takano, H.; Shimura, Y.;
Sodeoka, M. J. Am. Chem. Soc. 2005, 127, 10164–10165.
14. (a) Hamashima, Y.; Hotta, D.; Sodeoka, M. J. Am. Chem. Soc.
2002, 124, 11240–11241; (b) Hamashima, Y.; Hotta, D.;
Umebayashi, N.; Tsuchiya, Y.; Suzuki, T.; Sodeoka, M. Adv.
Synth. Catal. 2005, 347, 1576–1586; (c) Sodeoka, M.;
Hamashima, Y. Bull. Chem. Soc. Jpn. 2005, 78, 941–956.
15. Kumobayashi, H.; Miura, T.; Sayo, N.; Saito, T.; Zhang, X.
Synlett 2001, 1055–1064.
16. In the case of the Michael reaction, no reaction occurred with
the Pd complexes 2. The addition of a strong protic acid,
such as TfOH, which promotes the reaction cooperatively
with the Pd enolate as an activator of the enone, was
necessary.
17. For the preparation of b-ketophosphonates: Calogeropoulou,
T.; Hammond, G. B.; Wiemer, D. F. J. Org. Chem. 1987, 52,
4185–4190.
18. a-Fluorophosphonates have been used as mimics of a phos-
phate moiety. For a general review: Berkowitz, D. B.; Bose,
M. J. Fluorine Chem. 2001, 112, 13–33.
19. For example: (a) Nieschalk, J.; O’Hagan, D. J. Chem. Soc.,
Chem. Commun. 1995, 719–720; (b) Li, C.; Wu, L.; Otaka,
A.; Smyth, M. S.; Roller, P. P.; Burke, T. R., Jr.; den Hertog,
J.; Zhang, Z.-Y. Biochem. Biophys. Res. Commun. 1995, 216,
976–984; (c) Yokoyama, T.; Yamagishi, T.; Matsumoto, K.;
Shibuya, S. Tetrahedron 1996, 52, 11725–11738; (d)
Berkowitz, D. B.; Bose, M.; Pfannenstiel, T. J.; Doukov, T.
J. Org. Chem. 2000, 65, 4498–4508.
20. (a) Kotoris, C. C.; Wen, W.; Lough, A.; Taylor, S. D. J. Chem.
Soc., Perkin Trans. 1 2000, 1271–1281; (b) Ruiz, M.; Ojea, V.;
´
Quintela, J. M.; Guillın, J. J. Chem. Commun. 2002, 1600–
1601.
8. (a) Hintermann, L.; Togni, A. Angew. Chem., Int. Ed. 2000, 39,
4359–4362; (b) Piana, S.; Devillers, I.; Togni, A.;
Rothlisberger, U. Angew. Chem., Int. Ed. 2002, 41, 979–982;
(c) Frantz, R.; Hintermann, L.; Perseghini, M.; Broggini, D.;
Togni, A. Org. Lett. 2003, 5, 1709–1712.
21. For details, see Supporting Information of Ref. 10a.
22. (a) Harada, N.; Koumura, N.; Feringa, B. L. J. Am. Chem. Soc.
1997, 119, 7256–7264; (b) Watanabe, M.; Kuwahara, S.;
Harada, N.; Koizumi, M.; Ohkuma, T. Tetrahedron:
Asymmetry 1999, 10, 2075–2078.
9. Kim, D. Y.; Park, E. J. Org. Lett. 2002, 4, 545–547.

Y. Hamashima et al. / Tetrahedron 62 (2006) 7168–7179
7179
23. (a) Fujita, M.; Hiyama, T. J. Am. Chem. Soc. 1985, 107, 8294–
8296; (b) Kitazume, T.; Kobayashi, T.; Yamamoto, T.;
Yamazaki, T. J. Org. Chem. 1987, 52, 3218–3223.
24. The difference in coupling constants of the methine proton
against the fluorine atom is derived from axial–equatorial and
axial–axial relationships on the assumption that the ester group
is at the equatorial position, as seen in the case of cyclohexane
derivatives. The unambiguous stereochemistry observed in the
reduction of 12d under similar reaction conditions supports the
predominant formation of 21.
25. For phosphonate ester dealkylation: McKenna, C. E.;
Schmidhauser, J. J. Chem. Soc., Chem. Commun. 1979, 739.
26. Fujii, A.; Hagiwara, E.; Sodeoka, M. J. Am. Chem. Soc. 1999,
121, 5450–5458.
27. Gori, G.; Bartolucci, G. B.; Sturaro, A.; Pavoli, G.; Doretti, L.;
Troiano, R.; Casetta, B. J. Chromatogr., B 1995, 165–172.