Highly diastereoselective reductive coupling of 2-bromo-2,3,3,3- tetrafluoropropanamide with aldehydes promoted by triphenylphosphine- titanium(IV) isopropoxide. An efficient route to the synthesis of erythro-α-fluoro-α-(trifluoromethyl)-β-hydroxy amides

Article abstract of DOI:10.1021/jo0358729

The reductive coupling reaction of N-methoxy-N-methyl-2-bromo-2,3,3,3- tetrafluoropropanamide (Weinreb amide) with various aldehydes under the influence of the combined reagent, 1.2 equiv each of triphenylphosphine and titanium(IV) isopropoxide, took plac

Full text of DOI:10.1021/jo0358729

High ly Dia ster eoselective Red u ctive Cou p lin g of
2
-Br om o-2,3,3,3-tetr a flu or op r op a n a m id e w ith Ald eh yd es P r om oted
by Tr ip h en ylp h osp h in e-Tita n iu m (IV) Isop r op oxid e. An Efficien t
Rou te to th e Syn th esis of
er yth r o-r-F lu or o-r-(tr iflu or om eth yl)-â-h yd r oxy Am id es
Kei Sato, Takashi Sekiguchi, Takashi Ishihara,* Tsutomu Konno, and Hiroki Yamanaka
Department of Chemistry and Materials Technology, Kyoto Institute of Technology, Matsugasaki,
Sakyo-ku, Kyoto 606-8585, J apan
ishihara@ipc.kit.ac.jp
Received December 26, 2003
The reductive coupling reaction of N-methoxy-N-methyl-2-bromo-2,3,3,3-tetrafluoropropanamide
(Weinreb amide) with various aldehydes under the influence of the combined reagent, 1.2 equiv
each of triphenylphosphine and titanium(IV) isopropoxide, took place smoothly at ambient
temperature to give the corresponding R-fluoro-R-(trifluoromethyl)-â-hydroxy amides in a highly
erythro-selective manner. The high erythro selectivity was also obtained even by employing a
combination of triphenylphosphine (1.2 equiv) and a catalytic amount of titanium(IV) isopropoxide.
In tr od u ction
SCHEME 1
The introduction of fluorine atom(s) into an organic
molecule often dramatically alters its chemical properties
and pharmacological profiles, especially in the case of
biologically active compounds. As a consequence, many
efforts have been put forth toward the development of
new synthetic routes to various classes of fluorinated
compounds.1 Out of the diversity of fluorinated com-
pounds, R-fluoro-R-(trifluoromethyl)-â-hydroxy carbonyl
compounds 1 (Scheme 1), involving two successive ste-
reocenters, are recognized as one of the most important
synthetic units in organofluorine chemistry as well as in
organic synthesis because the nonfluorinated counter-
parts are frequently found in the frameworks of naturally
occurring substances. To synthesize such a compound 1
stereoselectively, as shown in Scheme 1, the aldol reac-
tion of the metal enolate 2 bearing R-fluoro and R-tri-
nonfluorinated counterpart since the electron density at
the reactive carbon decreases owing to the electron
withdrawal of the CF group.
3
Over the past two decades, intense attempts have been
made to realize the proper reaction conditions or means
which facilitate the aldol reaction in question without
fluoride ion elimination. A few modified means for the
aldol reaction have been developed successfully. For
example, treating a mixture of ethyl 2,3,3,3-tetrafluoro-
propanoate and various carbonyl compounds in THF with
LDA at -90 to -70 °C (called as the electrophile-
coexisting aldol reaction) gives the corresponding â-hy-
droxy esters 1 in 66-80% yields, but with almost no
diastereoselectivity.4 1-Substituted 1-perfluoroalkenyl
phosphates are subjected to reductive dephosphorylation
with diisobutylaluminum hydride to generate the corre-
sponding aluminum enolates, which react smoothly with
various aldehydes to give rise to the corresponding
â-hydroxy ketones 1 as diastereomeric mixtures in good
yields.5 On the other hand, the highly stereoselective
aldol reaction between 2,3,3,3-tetrafluoropropanamides
3
fluoromethyl (CF ) groups with aldehydes would be a
general and powerful tool, in view of the extensive studies
2
in the nonfluorinated enolate chemistry. However, the
literature has embraced merely quite limited studies on
such a reaction of 2, this being largely because the
generation of this type of enolate 2 is often hampered by
its decomposition through a fluoride ion elimination.3
Additionally, the enolate 2 is less reactive than the
,6
(1) (a) Biomedical Aspects of Fluorine Chemistry; Filler, R., Koba-
2 3
and aldehydes is effected by using Bu BOTf and Et N,
in which the in situ formed boron enolates react with
yashi, Y., Eds.; Kodansya & Elsevier Biomedical: Tokyo, 1982. (b)
Welch, J . T. Tetrahedron 1987, 43, 3123. (c) Welch, J . T.; Eswarakrish-
nan, S. Fluorine in Bioorganic Chemistry; J ohn Wiley & Sons: New
York, 1991. (d) Biomedical Frontiers of Fluorine Chemistry; Ojima, I.,
McCarthy, J . R., Welch, J . T., Eds.; American Chemical Society:
Washington, DC, 1996.
various aldehydes to provide the threo isomers of the
7
,8
corresponding â-hydroxy amides 1 exclusively.
(
2) (a) Mahwald, R. Chem. Rev. 1999, 99, 1095. (b) Mukaiyama, T.
(4) Qian, C.-P.; Nakai, T. Tetrahedron Lett. 1990, 31, 7043.
(5) Ishihara, T.; Yamaguchi, K.; Kuroboshi, M. Chem. Lett. 1989,
Aldrichim. Acta 1996, 29, 59. (c) Heathcock, C. H. Aldrichim. Acta
1
990, 23, 99.
3) Yokozawa, T.; Nakai, T.; Ishikawa, N. Tetrahedron Lett. 1984,
5, 3987.
1191.
(
(6) Ishihara, T.; Kuroboshi, M.; Yamaguchi, K.; Okada, Y. J . Org.
Chem. 1990, 55, 3107.
2
1
0.1021/jo0358729 CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/30/2004
J . Org. Chem. 2004, 69, 5041-5047
5041

Sato et al.
SCHEME 2
EtAlCl
2
, effected the reaction with low diastereoselectiv-
ity to give a mixture of the erythro- and threo-isomers in
4% and 66% yield, respectively (entries 4 and 5).
Changing the Lewis acid from Et AlCl to TiCl (O-i-Pr)
8
2
2
2
resulted in a substantial decrease in the yield (26%), and
the erythro selectivity was profoundly increased from 67%
to 96%, though a large amount of 5a (65% yield) was
produced (entry 7). Very interestingly, when Ti(O-i-Pr)
2
was employed as Lewis acid in place of TiCl (O-i-Pr) ,
4
2
During the course of our studies on the chemistry of
this type of enolates 2 and their aldol and related
reactions, we have successfully attained an efficient and
convenient means for the highly stereoselective aldol
reaction of the enolate derived from 2-bromo-2,3,3,3-
tetrafluoropropanamide 3a (Z ) N(OMe)Me, Scheme 2).
Herein, we wish to describe the results of the reductive
coupling reactions between the amides or ester 3 and
various aldehydes promoted by a combination of tri-
phenylphosphine and Lewis acid. The reaction of 3a is
the first example of the erythro-selective aldol reaction
of the type of enolate 2 and can be complementary to the
boron enolate-mediated aldol reaction reported previ-
ously.8
significant improvement of the yield was realized without
any decrease of the diastereoselectivity; the coupling
product 4a a was obtained in 84% yield with 97% erythro
selectivity (entry 8). Other reaction conditions were
4
further examined by using Ti(O-i-Pr) as Lewis acid. As
can be seen in entries 10-13, the amounts and the sorts
of phosphine used were found to affect both chemical
yield and diastereoselectivity. Thus, the use of tribu-
tylphosphine furnished a mixture of the erythro- and
threo-isomers in a ratio of 43:57 (entry 10). On decreasing
the amount of triphenylphosphine from 120 to 20 or 0
mol %, the chemical yield was decreased to 16% and 0%,
respectively (entries 11 and 12). It should be noted that
the reactions in polar solvents, such as acetonitrile and
THF, took place in a highly erythro-selective manner,
whereas the reaction did not in a nonpolar solvent such
as toluene (entries 14-16).
Resu lts a n d Discu ssion
P h osp h in e/Lew is Acid -Med ia ted Red u ctive Cou -
p lin g Rea ction of 3. Recent literature has disclosed the
original applications of the combination system of phos-
With the optimized reaction conditions (Table 1, entry
8
), the reductive coupling reaction of 2-bromo-2,3,3,3-
tetrafluoropropanamide 3a with various aldehydes was
examined. The results are summarized in Table 2.
As in the reaction with benzaldehyde, various aromatic
aldehydes, such as p-methylbenzaldehyde, p-methoxy-
benzaldehyde, and p-fluorobenzaldehyde, underwent the
reaction with high diastereoselectivity to afford prefer-
entially the corresponding erythro products 4a in high
yields (entries 1-4). The reaction with R,â-unsaturated
aldehydes, such as crotonaldehyde and cinnamaldehyde,
also efficiently proceeded in a highly diastereoselective
fashion (entries 5 and 6). The bulkiness of the aldehyde
was observed to affect the efficiency of the reaction. When
the substituent R in aldehyde was changed from n-propyl
to isopropyl or tert-butyl, the yield of 4a was decreased
from 72% to 65% or 53% yield, respectively, but the
diastereoselectivity was obtained with a satisfactorily
high level (entries 7-9).
9
,10
phine/Lewis acid to the aldol-type reaction.
Thus, this
combined reagent has the characteristic ability of reduc-
ing R-halo carbonyl compounds to generate the corre-
sponding enolate, which reacts smoothly with various
carbonyl compounds, affording the coupling products in
good yields with high stereoselectivity. These results
strongly stimulated us to apply the combined reagent to
fluorinated R-bromo carbonyl compounds.
We initiated our research for optimal conditions of the
reaction between 2-bromo-2,3,3,3-tetrafluoropropanamide
3
a (called as Weinreb amide), prepared readily from the
corresponding acid chloride, and the combined reagent.
Thus, to a solution of 1.2 equiv of BF ‚Et O in CH Cl
3
2
2
2
was added 1.2 equiv each of benzaldehyde and tri-
phenylphosphine in this order. After the above reaction
mixture was stirred for 10 min at 0 °C, 3a was added
and the whole was stirred for 24 h at ambient temper-
ature (Scheme 3). As a result, the desired R-fluoro-R-
To expand the scope of this reductive coupling reaction,
the reactions of related amide 3b (Z ) NBu
c (Z ) OBn) with benzaldehyde were examined simi-
larly.
In sharp contrast to the reaction of the Weinreb amide
a , Ti(O-i-Pr) did not efficiently promote the reaction of
b and 3c (entries 1 and 6). Then, a variety of Lewis
, Zn(OTf) , Et AlCl, and EtAlCl , were
reexamined for the reactions both of 3b and of 3c. As
summarized in Table 3, Et AlCl was very effective for
2
) and ester
(trifluoromethyl)-â-hydroxy amide 4a a was obtained in
3
3
7% yield as a diastereomeric mixture of the erythro- and
1
1
threo-isomers in a ratio of 67:33, respectively, together
with the debrominated product 5a in 44% (Table 1, entry
3
3
4
1
). A number of Lewis acids were examined to improve
the yield and diastereoselectivity. Neither zinc triflate
nor zinc iodide gave the desired adduct at all (entries 2
and 3), and aluminum Lewis acids, such as Et AlCl and
2
acids, such as TiCl
4
2
2
2
2
the reaction of 3b with benzaldehyde, in which the
coupling product 4ba was afforded in 90% yield, but with
low diastereoselectivity. On the other hand, TiCl was
4
found to be a good promoter for the reaction of 3c with
benzaldehyde, the desired product 4ca being produced
in 60% yield with high erythro selectivity.
From the results mentioned above, it can be assumed
that the use of the Weinreb amide 3a and the combined
(
7) Ishihara, T.; Kuroboshi, M.; Yamaguchi, K. Chem. Lett. 1990,
2
11.
(8) Kuroboshi, M.; Ishihara, T. Bull. Chem. Soc. J pn. 1990, 63, 1191.
9) (a) Hashimoto, Y.; Kikuchi, S. Chem. Lett. 2002, 126. (b)
(
Kagoshima, H.; Hashimoto, Y.; Saigo, K. Tetrahedron Lett. 1998, 39,
465.
10) For closely related reports, see: Kagoshima, H.; Hashimoto,
Y.; Oguro, D.; Kutsuna, T.; Saigo, K. Tetrahedron Lett. 1998, 39, 1203.
11) The relative stereochemical nomenclature proposed by Noyori
8
(
(
et al. is applied. See: Noyori, R.; Ishida, I.; Sakata, J . J . Am. Chem.
Soc. 1983, 105, 1598.
3 4
reagent, PPh /Ti(O-i-Pr) , is most favorable for the present
5
042 J . Org. Chem., Vol. 69, No. 15, 2004

Synthesis of erythro-R-Fluoro-R-(trifluoromethyl)-â-hydroxy Amides
SCHEME 3
TABLE 1. Red u ctive Cou p lin g Rea ction of 3a w ith Ben za ld eh yd e (R ) P h ) u n d er Va r iou s Rea ction Con d ition s
a
isomer ratioa
a
a
yield /%
yield /%
recovery /%
entry
Lewis acid
compd
solvent
of 4a a
erythro-4a a /threo-4a a
of 5a
of 3a
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
BF3‚OEt2
Zn(OTf)2
ZnI2
Et2AlCl
EtAlCl2
PPh3
PPh3
PPh3
PPh3
PPh3
PPh3
PPh3
PPh3
P(OEt)3
Bu3P
Ph3P
none
Ph3P
Ph3P
Ph3P
Ph3P
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
MeCN
37
5
0
84
66
b
26
84
0
67:33
68:32
44
80
86
7
0
5
0
0
1
0
0
0
0
45
43
91
0
0
21
44
67:33
60:40
4
TiCl4
16
65
15
29
43
6
TiCl2(O-i-Pr)2
Ti(O-i-Pr)4
Ti(O-i-Pr)4
Ti(O-i-Pr)4
Ti(O-i-Pr)4
Ti(O-i-Pr)4
none
Ti(O-i-Pr)4
Ti(O-i-Pr)4
Ti(O-i-Pr)4
96:4
97:3
1
1
1
1
1
1
1
30
16
0
43:57
95:5
c
0
3
82
25
20
34
61
45
7
98:2
96:4
THF
toluene
a
Determined by 19F NMR. b Complex mixture. c A catalytic amount of PPh3 (0.2 equiv) was used.
TABLE 2. Dia ster eoselective Red u ctive Cou p lin g Rea ction of 3a w ith Va r iou s Ald eh yd es
a
isomer ratioa
a
a
yield /%
yield /%
recovery /%
entry
R
of 4a
erythro-4a /threo-4a
of 5a
of 3a
1
2
3
4
5
6
7
8
9
Ph
84 (73)
79 (73)
77 (61)
82 (76)
74 (63)
76 (68)
72 (60)
65 (53)
53 (43)
4a a
4a b
4a c
4a d
4a e
4a f
4a g
4a h
4a i
97:3
98:2
96:4
97:3
95:5
95:5
96:4
95:5
95:5
15
12
22
13
15
17
27
22
25
0
0
0
0
0
0
0
0
0
p-MeC6H4
p-MeOC6H4
p-FC6H4
MeCHdCH
PhCHdCH
n-Pr
i-Pr
t-Bu
a
Determined by 19F NMR. Values in parentheses are of isolated yield.
coupling reaction leading to R-fluoro-R-(trifluoromethyl)-
â-hydroxy carboxylic acid moiety with high diastereose-
lectivity in good yield.
1-4, no difference between the stoichiometric and the
catalytic Lewis acid system was observed in view of the
yields as well as the diastereoselectivities. In the case of
R,â-unsaturated aldehydes and n-butyraldehyde, the
yields of the corresponding 4a were still high but the
diastereoselectivities were decreased to a slight extent
(entries 5-7). Moreover, the reaction with bulky alde-
hydes, such as isobutyraldehyde and pivalaldehyde, led
to a significant decrease in the yields of the products
P h osp h in e/Lew is Acid -Ca ta lyzed Red u ctive Cou -
p lin g Rea ction of 3a . Next, we investigated the pos-
sibility of the phosphine/Lewis acid-catalyzed reaction of
1
2
3
a . Table 4 summarizes the results of the reactions with
benzaldehyde. As shown in entries 1-3, changing the
amount of Lewis acid Ti(O-i-Pr) from 120 mol % to 10
4
(entries 8 and 9).
mol % did not cause any substantial difference in the
yields of 4a a and diastereoselectivities. However, the
Ster eoch em ica l Assign m en t a n d P la u sible Rea c-
reaction in the presence of only 5 mol % of Ti(O-i-Pr)
resulted in a slight decrease of the yield, though the
diastereoselectivity still remained excellent (entry 4).
4
tion Mech a n ism . The stereochemical assignment of the
coupling products 4a was made as follows. The X-ray
structural analysis was carried out for the major isomer
of 4a c obtained from the reaction of 3a with p-methoxy-
benzaldehyde.13 The data permitted its stereochemical
With these optimized reaction conditions in hand
(Table 4, entry 3), we carried out the phosphine/Lewis
1
9
assignment to the erythro-isomer. F NMR analysis of
a c indicated that the CF resonance due to the erythro-
acid-catalyzed reaction between 3a and various alde-
hydes. These results are tabulated in Table 5. For the
reaction with aromatic aldehydes, as shown in entries
4
3
isomer appeared at higher field than that of the threo-
isomer.14 This tendency to the relative resonances for the
two diastereoisomers is also observed for the aldol
(12) It has been disclosed that a stoichiometric combination of
triphenylphosphine and Lewis acid is crucial to facilitate the aldol-
type reaction of R-bromo carbonyl compounds, which are limited to
N,N-diphenyl amide, thioesters, and ketones. See ref 9.
(13) For details regarding the X-ray crystal structure of 4a c, see
the Supporting Information.
J . Org. Chem, Vol. 69, No. 15, 2004 5043

Sato et al.
TABLE 3. Red u ctive Cou p lin g Rea ction of Am id e 3b or Ester 3c w ith Ben za ld eh yd e
a
isomer ratioa
a
a
yield /%
yield /%
recovery /%
entry
Lewis acid
3
of 4
erythro-4/threo-4
of 5
of 3
1
2
3
4
5
6
7
8
9
Ti(O-i-Pr)4
TiCl4
3b
3b
3b
3b
3b
3c
3c
3c
3c
3c
8
51
16
90
44
14
60
6
4ba
4ba
4ba
4ba
4ba
4ca
4ca
4ca
4ca
4ca
61
32
65
9
52
31
26
40
64
61
16
0
3
0
0
13
3
1
2
2
62:38
78:22
63:37
70:30
49:51
89:11
50:50
64:36
50:50
Zn(OTf)2
Et2AlCl
EtAlCl2
Ti(O-i-Pr)4
TiCl4
Zn(OTf)2
Et2AlCl
EtAlCl2
19
15
1
0
a
Determined by 19F NMR.
TABLE 4. Effect of th e Am ou n t of Lew is Acid
Con clu sion
isomer ratiob
In summary, we have investigated the reductive cou-
pling reaction between 2-bromo-2,3,3,3-tetrafluoropro-
panamides or ester 3a -c and various aldehydes under
b
b
b
Ti(O-i-Pr)4 yield /% erythro-4a a / yield /% recovery /%
entrya
(mol %)
of 4a a
threo-4a a
of 5a
of 3a
1
2
3
4
120
20
10
5
84
91
87
67
97:3
97:3
96:4
96:4
15
8
7
0
0
0
1
3 4
the influence of the combined reagent PPh /Ti(O-i-Pr)
and found that the reaction of N-methoxy-N-methyl-2-
bromo-2,3,3,3-tetrafluoropropanamide (3a ) with various
17
aldehydes in the presence of 1.2 equiv each of PPh
3
and
a
In all reactions, 1.2 equiv of PPh3 was employed. b Determined
by 19F NMR.
Ti(O-i-Pr) furnished the corresponding R-fluoro-R-(tri-
4
fluoromethyl)-â-hydroxy amides 4a in high yields with
high erythro selectivities. Additionally, even use of a
products obtained from the boron enolates of N,N-dialkyl-
8
catalytic amount of Ti(O-i-Pr) for the combined reagent
2
,3,3,3-tetrafluoropropanamides. On the basis of such
4
1
9
was found to facilitate the reaction efficiently with high
erythro selectivities.
relative resonances for the isomers, examining the
F
NMR spectra of other products 4a made it possible to
assign their major isomers as the erythro-isomer.
The present reaction probably involves the generation
of a titanium enolate which may take part in a six-
membered transition state, as depicted in Scheme 4.
Thus, both carbonyl oxygen and phosphorus atom could
Exp er im en ta l Section
Gen er a l Meth od s. Infrared spectra (IR) were recorded on
1
a Shimadzu FTIR-8200A (PC) spectrophotometer. H (500.13
MHz) and 13CNMR (125.75 MHz) spectra were measured with
coordinate Ti(O-i-Pr)
4
simultaneously and the ensuing
a Bruker DRX 500 NMR spectrometer in a chloroform-d
3 4
(CDCl ) solution with tetramethylsilane (Me Si) as an internal
bromine transfer occurs intramolecularly through a
closed chairlike transition state. Under this consider-
ation, the E-enolate may be disfavored owing to a
nonbonded repulsive interaction between a CF
and phenyl substituent of PPh (TS-1 vs TS-2), so that
the Z-enolate may be formed preferentially via TS-2. The
subsequent coordination of titanium atom to the carbonyl
oxygen of aldehyde may lead to a six-membered chairlike
transition state TS-4, where the substituent R occupies
the equatorial position due to a 1,3-diaxial repulsion
reference. A J EOL J NM-EX90A (84.21 MHz, FT) NMR spec-
trometer was used for determining 19F NMR spectra in a
3 3
CDCl solution with CFCl as an internal standard. Mass
3
group
spectra (MS) and high-resolution mass spectra (HRMS) were
taken on a Hitachi M-80B and/or J EOL J MS-700 mass
spectrometer by an electron impact (EI), chemical ionization
(CI), or FAB (Cs+) method using m-nitrobenzyl alcohol as a
matrix. Elemental analyses were conducted with a Yanaco
CHN CORDER MT-5 instrument.
Ma t er ia ls. 2-Bromo-2,3,3,3-tetrafluoropropanoyl chloride
was purchased from Pinnacle Chemicals Ltd. and was used
2 2
without any purification. Dichloromethane (CH Cl ) was dis-
tilled over calcium hydride under argon. All chemicals were
of reagent grade and, if necessary, were purified in the usual
manner prior to use. Thin-layer chromatography (TLC) was
done with Merck silica gel 60F254 plates, and column chroma-
tography was carried out with Wakogel C-200. All reactions
were conducted under an atmosphere of argon.
3
(TS-3 vs TS-4). Consequently, the erythro-isomer of the
product 4a is produced preferentially.
(14) It was observed that the rotational isomerism in the unsym-
metrical (N-methoxy, N-methyl) amide caused broadening of the
resonances in the 1H, C, and F NMR of 3a and 4a . In particular,
the methoxy protons for 3a appeared at 3.974 and 3.978 ppm as two
peaks, corresponding to the two amide rotamers. Fortunately, the
broadening of peaks did not complicate the analysis of diastereomeric
13
19
P r ep a r a tion of N-Meth oxy-N-m eth yl-2-br om o-2,3,3,3-
t et r a flu or op r op a n a m id e (3a ). To a solution of N,O-di-
3
distributions of 4a by using the CF resonance peaks.
5
044 J . Org. Chem., Vol. 69, No. 15, 2004

Synthesis of erythro-R-Fluoro-R-(trifluoromethyl)-â-hydroxy Amides
TABLE 5. Dia ster eoselective Ca ta lytic Red u ctive Cou p lin g Rea ction of 3a w ith Va r iou s Ald eh yd es
b
isomer ratiob
b
b
yield /%
yield /%
recovery /%
entrya
RCHO
of 4a
erythro-4a /threo-4a
of 5a
of 3a
1
2
3
4
5
6
7
8
9
Ph
87 (82)
84 (66)
91 (83)
83 (81)
91 (83)
87 (62)
84 (68)
50 (46)
20 (5)
4a a
4a b
4a c
4a d
4a e
4a f
4a g
4a h
4a i
96:4
96:4
95:5
97:3
89:11
87:13
90:10
86:14
87:13
7
9
5
16
12
9
11
37
37
0
0
0
0
3
1
0
0
3
p-MeC6H4
p-MeOC6H4
p-FC6H4
MeCHdCH
PhCHdCH
n-Pr
i-Pr
t-Bu
a
In all reactions, 1.2 equiv of PPh3 and 0.1 equiv of Ti(O-i-Pr)4 were employed. b Determined by 19F NMR. Values in parentheses are
of isolated yield.
methylhydroxylamine hydrochloride (2.15 g, 11 mmol) in CH
2
-
tracted with CH
washed with brine followed by drying over anhydrous Na
2
Cl
2
(20 mL × 5), and the organic layers were
SO
4
Cl
2
(30 mL) was added pyridine (3.48 g, 22 mmol) at 0 °C under
2
an argon atmosphere. After being stirred for 30 min, to this
reaction mixture was dropwise added a solution of 2-bromo-
and filtration. The solvents were removed in vacuo, and the
residue was purified by silica gel column chromatography
(hexane/diethyl ether, 5:1) to afford analytically pure product
4a a (0.971 g, 0.329 mmol, 73% yield).
er yt h r o-N-Me t h oxy-N -m e t h yl-2-flu or o-3-h yd r oxy-3-
p h en yl-2-(tr iflu or om eth yl)p r op a n a m id e (4a a ): mp 80-
82 °C; IR (KBr) 3344, 2943, 1651, 1288, 1196, 999, 968, 810,
2
,3,3,3-tetrafluoropropanoyl chloride (4.87 g, 10 mmol) in CH
Cl (10 mL) at 0 °C. The whole mixture was then stirred at
room temperature for 2 h and the reaction was quenched with
a saturated aqueous NaHCO solution (50 mL). The resultant
mixture was extracted with CH Cl
(50 mL × 5), and the
organic layers were washed with brine, dried over anhydrous
Na SO , and filtered. The solvents were removed in vacuo, and
the residue was purified by silica gel column chromatography
hexane/diethyl ether, 5:1) to afford analytically pure amide
a (2.41 g, 9.0 mmol, 90% yield). The amide 3b (2.2 equiv of
Bu NH, rt, 3 h, THF) and ester 3c (1.1 equiv of benzyl alcohol,
.5 equiv of Et N, rt, 3 h, ether) were prepared in a similar
manner.
2
-
2
3
2
2
-
1 1
748, 698 cm ; H NMR (CDCl ) δ 3.22 (br, 3H), 3.53 (s, 3H),
3
2
4
4.35 (br, 1H), 5.29 (d, J ) 15.5 Hz, 1H), 7.32-7.45 (m, 5H);
1
3
C NMR (CDCl ) δ 33.9, 61.8, 73.7 (d, J ) 22.7 Hz), 93.5-
3
(
3
96.0 (m), 121.3 (dq, J ) 37.5, 286.8 Hz), 127.8, 128.1, 128.8,
136.4, 164.0 (m); 19F NMR (CDCl , CFCl ) δ -73.8 (d, J ) 8.8
3
3
+
2
Hz, 3F), -175.0 to -183.0 (m, 1F); HRMS (FAB ) m/z found
+
1
3
296.0914, calcd for (M + H) C H F NO 296.0910. Anal.
1
2
14
4
3
Calcd for C H F NO : C, 48.82; H, 4.44; N, 4.74. Found: C,
1
2
14
4
3
N-Meth oxy-N-m eth yl-2-br om o-2,3,3,3-tetr a flu or op r o-
p a n a m id e (3a ): IR (neat) 2947, 1693, 1461, 1292, 1222, 1188,
48.63; H, 4.38; N, 4.72.
er yth r o-N-Meth oxy-N-m eth yl-2-flu or o-3-h yd r oxy-3-(4-
m eth ylph en yl)-2-(tr iflu or om eth yl)pr opan am ide (4ab): mp
131-133 °C; IR (KBr) 3337, 2943, 1643, 1288, 1200, 1134,
103, 956, 902 cm- ; H NMR (CDCl
1
1
) δ 3.30 (br, 3H), 3.974
1
3
3
1
and 3.978 (s, 3H, rotational isomer); C NMR (CDCl
6
Hz), 159.8; F NMR (CDCl
3
) δ 33.6,
-
1 1
1.4, 90.6 (qd, J ) 32.3, 283.9 Hz), 119.9 (dq, J ) 29.8, 283.6
1003, 968 cm ; H NMR (CDCl ) δ 2.35 (s, 3H), 3.23 (br, 3H),
3
1
9
3
+
, CFCl
3
) δ -77.3 (d, J ) 8.8 Hz,
3.56 (s, 3H), 4.10 (br, 1H), 5.26 (dd, J ) 8.0, 16.5 Hz, 1H),
+
13
3
1
2
F), -139.0 (br, 1F); MS (EI ) m/z (rel intensity) 266 (M , 39),
7.12-7.20 (m, 2H), 7.27-7.34 (m, 2H); C NMR (CDCl ) δ
3
+
78 (47), 128 (12), 88 (30), 58 (100); HRMS (M ) m/z found
21.1, 33.0, 62.0, 73.8 (d, J ) 23.7 Hz), CF CF could not be
3
7
9
66.9515, calcd for C
N ,N -D i b u t y l-2-b r o m o -2,3,3,3-t e t r a flu o r o p r o p a n -
a m id e (3b): 93% yield; IR (neat) 2939, 2877, 1674, 1434, 1288,
5
H
6
BrF
4
NO
2
266.9518.
detected. 121.4 (dq, J ) 28.2, 286.9 Hz), 127.7, 128.9, 133.4,
1
9
138.7, a carbonyl carbon could not be detected; F NMR
(CDCl
(m, 1F); HRMS (FAB ) m/z found 310.1071, calcd for (M
NO 309.1066. Anal. Calcd for C13 NO : C,
3 3
, CFCl ) δ -73.9 (d, J ) 6.6 Hz, 3F), -174.0 to -182.0
-
1
1
+
+
1
(
222, 1184, 1095, 1033, 945, 902, 702, 675 cm
CDCl ) δ 0.90 (t, J ) 7.5 Hz, 3H), 0.91 (t, J ) 7.5 Hz, 3H),
.25-1.35 (m, 4H), 1.45-1.70 (m, 4H), 3.08-3.25 (m, 2H),
;
H NMR
+
3
H) C13
H
15
F
4
3
H
15
F
4
3
1
3
1
3
2
50.49; H, 4.89; N, 4.53. Found: C, 50.12; H, 4.94; N, 4.42.
er yth r o-N-Meth oxy-N-m eth yl-2-flu or o-3-h yd r oxy-3-(4-
m e t h o x y p h e n y l)-2-(t r i flu o r o m e t h y l)p r o p a n a m i d e
(4a c): mp 96-97 °C; IR (KBr) 3337, 2943, 2846, 1651, 1516,
1
3
.45-3.56 (m, 1H), 3.60-3.72 (m, 1H); C NMR (CDCl ) δ
3
3.5, 13.6, 19.8, 19.9, 28.6, 30.7, 47.2, 48.2, 48.3, 91.9 (qd, J )
3.4, 284.6 Hz), 112.0 (dq, J ) 30.3, 283.7 Hz), 159.5 (d, J )
1
9
-1 1
0.3 Hz); F NMR (CDCl
3
, CFCl
3
) δ -77.0 (d, J ) 8.7 Hz, 3F),
1288, 1238, 1188, 999, 968, 787 cm ; H NMR (CDCl ) δ 3.24
3
(br, 3H), 3.57 (s, 3H), 3.796 and 3.799 (s, 3H, rotational
isomers), 4.16 (br, 1H), 5.26 (d, J ) 17.5 Hz, 1H), 6.85-6.92
+
-
133.2 to -132.5 (m, 1F); HRMS (M + H) m/z found
79
3
36.0583, calcd for C11
Ben zyl 2-br om o-2,3,3,3-tetr aflu or opr opan oate (3c): 90%
yield; IR (neat) 1774, 1501, 1457, 1304, 1265, 1192, 1134, 1018,
4 2
H18 BrF NO 336.0586.
1
3
(m, 2H), 7.30-7.37 (m, 2H); C NMR (CDCl
61.7, 73.4 (d, J ) 23.2 Hz), CF CF could not be detected. 113.5,
121.3 (dq, J ) 28.2, 286.7 Hz), 128.4, 129.0, 159.9, a carbonyl
3
) δ 34.0, 55.2,
3
-
1
1
9
18, 741 cm ; H NMR (CDCl
3
) δ 5.16 (s, 2H), 7.20 (s, 5H);
) δ 69.7, 88.8 (dq, J ) 272.3, 57.5 Hz), 119.7
dq, J ) 29.5, 284.0 Hz), 128.3, 128.7, 129.0, 133.5, 160.5 (d,
1
3
19
C NMR (CDCl
3
carbon could not be detected; F NMR (CDCl
3 3
, CFCl ) δ -73.9
+
(
(d, J ) 6.6 Hz, 3F), -174.0 to -183.0 (m, 1F); HRMS (FAB )
1
9
+
J ) 26.5 Hz); F NMR (CDCl
3
) δ -78.0 (d, J ) 8.8 Hz, 3F),
m/z found 326.1019, calcd for (M + H) C13
16 4 4
H F NO 326.1015.
-
135.2 (q, J ) 8.8 Hz, 1F); MS (FAB) m/z (rel intensity) 314
Anal. Calcd for C13 NO C, 48.00; H, 4.65; N, 4.31.
H
15
F
4
4
:
+
+
(
M , 100); HRMS (FAB) m/z found 313.9573, calcd for (M )
BrF 313.9566. Anal. Calcd for C10 BrF ; C, 38.12;
H, 2.24. Found: C, 38.02; H, 2.28.
Typ ica l P r oced u r e for th e Red u ctive Cou p lin g Rea c-
tion of 3a w ith Ben za ld eh yd e. To a solution of Ti(O-i-Pr)
0.154 g, 0.54 mmol) in CH Cl (2 mL) were successively added
solutions of benzaldehyde (0.057 g, 0.54 mmol) in CH Cl (1
mL), PPh (0.142 g, 0.54 mmol) in CH Cl (1 mL), and 3a (0.121
g, 0.45 mmol) in CH Cl (1 mL) at room temperature under
Found: C, 47.94; H, 4.99; N, 4.23.
er yth r o-N-Meth oxy-N-m eth yl-3-(4-flu or op h en yl)-2-flu -
or o-3-h ydr oxy-2-(tr iflu or om eth yl)pr opan am ide (4ad): IR
C
10
H
7
4
O
2
H
7
4
O
2
-1
(neat) 3422, 2943, 1655, 1512, 1393, 1285, 1203, 999, 849 cm
;
1
4
H NMR (CDCl
5.31 (dd, J ) 7.0, 17.5 Hz, 1H), 7.04-7.07 (m, 2H), 7.38-7.41
(m, 2H); 13C NMR (CDCl
) δ 33.7, 62.0, 73.2 (d, J ) 23.2 Hz),
3
) δ 3.24 (br, 3H), 3.57 (s, 3H), 4.2-4.6 (br, 1H),
(
2
2
2
2
3
3
2
2
93.0-96.0 (m), 115.1 (d, J ) 21.7 Hz), 121.3 (dq, J ) 28.1,
286.9 Hz), 129.6-129.8 (m), 132.2, 163.0 (d, J ) 248.0 Hz),
2
2
an argon atmosphere. After being stirred at the same tem-
perature for 24 h, the reaction mixture was quenched with
aqueous 10% HCl (10 mL). The resultant mixture was ex-
163.0-164.2 (m); 19F NMR (CDCl
, CFCl ) δ -73.9 (d, J ) 6.6
3 3
Hz, 3F), -113.0 to -113.6 (m, 1F), -174.0 to -183.0 (m, 1F);
+
+
HRMS (FAB ) m/z found 314.0819, calcd for (M + H)
J . Org. Chem, Vol. 69, No. 15, 2004 5045

Sato et al.
SCHEME 4
Calcd for C
1.85; H, 5.33; N, 5.34.
er yt h r o-N-Me t h oxy-N -m e t h yl-2-flu or o-3-h yd r oxy-5-
p h en yl-2-(tr iflu or om eth yl)-4-(E)-p en ten a m id e (4a f): IR
9
H
13
F
4
NO
3
: C, 41.70; H, 5.06; N, 5.40. Found: C,
4
(
neat) 3422, 2943, 1663, 1427, 1389, 1285, 1200, 1142, 968,
-
1 1
6
4
1
94, 664 cm ; H NMR (CDCl
.92 (dd, J ) 7.0, 17.0 Hz, 1H), 6.26 (dd, J ) 7.0, 16.0 Hz,
3
) δ 3.30 (br, 4H), 3.71 (s, 3H),
1
3
H), 6.77 (d, J ) 16.0 Hz, 1H), 7.25∼7.41 (m, 5H); C NMR
) δ 33.9, 61.9, 73.0 (d, J ) 23.8 Hz), 93.0∼96.4 (m), 121.3
dq, J ) 27.9, 286.6 Hz), 123.30, 126.8, 128.3, 128.6, 135.1,
35.9, 164.0 (m); 19F NMR (CDCl
, CFCl ) δ -73.5 (br s, 3F),
177.0 to -184.0 (m, 1F); HRMS (FAB ) m/z found 322.1069,
(CDCl
3
(
1
-
3
3
+
+
calcd for (M + H) C14
NO : C, 52.34; H, 4.71; N, 4.36. Found: C, 52.37; H,
.07; N, 4.31.
er yth r o-N-Meth oxy-N-m eth yl-2-flu or o-3-h ydr oxy-2-(tr i-
flu or om eth yl)h exa n a m id e (4a g): IR (neat) 3345, 2966,
16 4 3
H F NO 322.1066. Anal. Calcd for
C
5
14
H
15
F
4
3
-1 1
2
878, 1663, 1466, 1389, 1285, 1200, 968 cm ; H NMR (CDCl
δ 0.97 (t, J ) 7.0 Hz, 3H), 1.39-1.49 (m, 1H), 1.52-1.75 (m,
H), 3.32 (br, 4H), 3.75 (s, 3H), 4.18 (ddt, J ) 2.0, 16.0, 10.0
3
)
3
1
3
Hz, 1H); C NMR (CDCl
3
) δ 13.7, 18.7, 32.9, 34.0 (m), 62.0
(
CF
m), 71.9 (d, J ) 24.7 Hz), 121.6 (dq, J ) 28.2, 286.7 Hz),
CF and CdO could not be detected; 19F NMR (CDCl
,
3
3
CFCl
HRMS (FAB ) m/z found 262.1060, calcd for (M + H) C
NO 262.1066.
3
) δ -74.0 (d, J ) 6.6 Hz, 3F), -174.0 to -183.0 (m, 1F);
+
+
9
H
16
F
4
-
3
er yt h r o-N-Me t h oxy-N -m e t h yl-2-flu or o-3-h yd r oxy-4-
m eth yl-2-(tr iflu or om eth yl)p en ta n a m id e (4a h ): IR (neat)
-
1 1
3
(
3
3
(
460, 2974, 2943, 1663, 1389, 1281, 1204, 964 cm ; H NMR
CDCl ) δ 0.98-1.07 (m, 6H), 2.00-2.10 (m, 1H), 3.31 (br, 4H),
.76 (s, 3H), 3.96 (br, 1H); 13C NMR (CDCl
) δ 16.6, 20.7, 30.0,
3.9 (m), 61.9 (m), 75.6 (d, J ) 22.6 Hz), 95.0-97.5 (m), 122.2
3
3
1
9
dq, J ) 28.5, 286.9 Hz), 164.0∼165.0 (m); F NMR (CDCl
3
,
CFCl ) δ -74.4 (br, 3F), -172.0 to -183.0 (m, 1F); HRMS
3
+
+
(
2
5
FAB ) m/z found 262.1068, calcd for (M + H) C
9
H
16
F
4
NO
3
62.1066. Anal. Calcd for C H F NO : C, 41.38; H, 5.79; N,
9 15 4 3
.36. Found: C, 41.02; H, 5.61; N, 5.08.
er yth r o-N-Met h oxy-N-m et h yl-2-flu or o-3-h yd r oxy-4,4-
d im eth yl-2-(tr iflu or om eth yl)p en ta n a m id e (4a i): IR (neat)
-
1 1
3
460, 2962, 1651, 1470, 1389, 1276, 1207, 1080, 961 cm ; H
NMR (CDCl ) δ 1.01 (d, J ) 2.5 Hz, 9H), 3.08-3.62 (m, 3H),
.77 (s, 3H), 3.08 (dd, J ) 8.0, 11.0 Hz, 1H), 4.00-4.62 (m,
H); 13C NMR (CDCl
) δ 27.0, 34.2 (m), 36.2, 62.2 (m), 78.5 (d,
CF and CdO
) δ -75.2 (br s,
3
3
1
3
J ) 23.9 Hz), 121.9 (dq, J ) 28.8, 288.3 Hz), CF
could not be detected; 19F NMR (CDCl
, CFCl
F), -168.0 to -176.5 (m, 1F); HRMS (FAB ) m/z found
76.1225, calcd for (M + H) C10
N,N-Dib u t yl-2-flu or o-3-h yd r oxy-3-p h en yl-2-(t r iflu o-
r om eth yl)p r op a n a m id e (4ba ): IR (neat) 3444, 2935, 2877,
3
3
3
+
3
2
+
18 4 3
H F NO 276.1223.
1
7
635, 1458, 1380, 1288, 1180, 1134, 1072, 1029, 1002, 945, 825,
-
1
+
40, 702 cm ; HRMS (FAB ) m/z found 364.1905, calcd for
+
(M
+ H) C10
Erythro isomer: H NMR (CDCl
.90 (t, J ) 7.5 Hz, 3H), 1.00-1.46 (m, 8H), 2.80-3.40 (m, 4H),
18 4 3
H F NO 364.1900.
1
3
) δ 0.78 (t, J ) 7.5 Hz, 3H),
0
5
1
3
.10-5.18 (m, 1H), 5.20-5.30 (m, 1H), 7.28-7.43 (m, 5H);
) δ 13.5, 19.6, 20.0, 29.0, 31.1, 47.3, 47.8, 49.0,
3.9 (d, J ) 25.0 Hz), 93.7 (qd, J ) 27.9, 213.5 Hz), 121.7 (dq,
J ) 28.0, 287.1 Hz), 127.6-127.8 (m), 128.0, 128.6, 137.0, 163.5
d, J ) 18.2 Hz); 19F NMR (CDCl
, CFCl ) δ -74.8 (d, J ) 8.8
Hz, 3F), -170.5 to -169.5 (m, 1F).
Threo isomer: 1H NMR (CDCl
C
NMR (CDCl
3
7
(
3
3
12 13 5 3 12 5 3
C H F NO 314.0816. Anal. Calcd for C12H F NO : C, 46.01;
H, 3.86; N, 4.47. Found: C, 45.68; H, 3.69; N, 4.32.
3
) δ 0.80 (t, J ) 6.5 Hz, 3H),
er yth r o-N-Meth oxy-N-m eth yl-2-flu or o-3-h ydr oxy-2-(tr i-
flu or om eth yl)-4-(E)-h exen a m id e (4a e): mp 85-87 °C; IR
0.87 (t, J ) 7.0 Hz, 3H), 0.88-1.45 (m, 8H), 2.80-3.40 (m, 4H),
5.20-5.30 (m), 5.59 (d, J ) 24.0 Hz, 1H), 7.28-7.43 (m, 5H);
1
3
(
9
neat) 3356, 2943, 1659, 1443, 1458, 1211, 1192, 1142, 991,
C NMR (CDCl ) δ 13.5, 19.6, 20.0, 28.8, 31.2, 47.4, 47.9, 49.0,
3
-
1
1
68 cm ; H NMR (CDCl
3
) δ 1.76 (d, J ) 7.0 Hz, 3H), 3.33
74.2 (d, J ) 18.7 Hz), 97.9 (qd, J ) 28.1, 220.2 Hz), 122.1 (dq,
(
br, 4H), 3.74 (s, 3H), 4.67 (dd, J ) 7.2, 17.5 Hz, 1H), 5.57 (dd,
J ) 30.5, 287.0 Hz), 128.0, 128.1, 128.8, 136.3, 162.5 (d, J )
1
3
19
J ) 7.2, 15.3 Hz, 1H), 5.90 (dq, J ) 6.7, 15.3 Hz, 1H); C NMR
(
3 3
18.0 Hz); F NMR (CDCl , CFCl ) δ -72.3 (d, J ) 4.4 Hz, 3F),
CDCl
3
) δ 17.8, 34.0, 61.7, 73.0 (d, J ) 23.8 Hz), 93.0∼97.0
-186.8 to -186.0 (m, 1F).
(
m), 121.3 (dq, J ) 27.9, 286.6 Hz), 125.5, 132.4, A carbonyl
Ben zyl 2-flu or o-3-h yd r oxy-2-(tr iflu or om eth yl)-3-p h e-
1
9
carbon could not be detected.; F NMR (CDCl
3
, CFCl
3
) δ -73.6
n ylp r op a n oa te (4ca ): IR (neat) 3471, 3075, 3039, 2958, 1770,
-
1
(
2
br s, 3F), -176.0 to -186.0 (m, 1F); HRMS (FAB) m/z found
1496, 1458, 1377, 1276, 1195, 1137, 1087, 1041 cm ; HRMS
60.0913, calcd for (M+ + H) C
H
F
NO
260.0913. Anal.
+
9
14
4
3
9 16 4 3
(EI) m/z found 342.0870, calcd for (M ) C H F NO 342.0878.
5
046 J . Org. Chem., Vol. 69, No. 15, 2004

Synthesis of erythro-R-Fluoro-R-(trifluoromethyl)-â-hydroxy Amides
Erythro isomer: 1H NMR (CDCl
H), 5.28-5.38 (m, 1H), 7.09-7.37 (m, 10H); C NMR (CDCl
) δ 2.82 (br, 1H), 5.31 (s,
the monoclinic space group C2/c with a ) 8.270(4) Å, b )
3
1
3
3
2
3
)
12.122(4) Å, c ) 8.071(2) Å, b ) 97.32(3) Å, V ) 744.9(4) Å ,
δ 68.7, 73.2 (d, J ) 19.2 Hz), 94.6 (qd, J ) 29.9, 210.9 Hz),
and Z ) 2. Data collection, reduction, solution, and refinement
were all carried out using the ‘teXsan’ crystallographic soft-
ware package from Molecular Structure Corp. All non-H atoms
were refined anisotropically; 3566 observations, 398 variables;
R ) 0.076 for F2 > 4σ(F2) and wR ) 0.183 for all F .
1
1
20.6 (dq, J ) 28.2, 286.1 Hz), 127.7-127.8 (m), 128.2, 128.4,
19
28.6, 128.7, 129.3, 134.1, 135.5, 163.7 (d, J ) 25.2 Hz);
F
NMR (CDCl
3
, CFCl ) δ -74.2 (d, J ) 7.6 Hz, 3F), -187.1 (dq,
3
J ) 7.6, 21.2 Hz, 1F).
1
2
2
Thero isomer: 1H NMR (CDCl
) δ 2.65 (br, 1H), 5.04 (s, 2H),
13
3
5
6
(
1
(
.28-5.38 (m, 1H), 7.09-7.37 (m, 10H); C NMR (CDCl
3
) δ
Ack n ow led gm en t. This work was financially sup-
8.4, 73.2 (d, J ) 19.2 Hz), 94.8 (qd, J ) 29.9, 210.9 Hz), 121.2
ported by a Grant-in-Aid for Scientific Research (No.
14550808) from the Ministry of Education, Science,
Sports and Culture, J apan.
dq, J ) 29.8, 286.9 Hz), 127.6-127.7 (m), 128.2, 128.4, 128.5,
28.7, 129.3, 133.6, 135.3, 162.9 (d, J ) 23.8 Hz); F NMR
, CFCl ) δ -73.5 (d, J ) 5.9 Hz, 3F), -191.6 (dq, J )
1
9
CDCl
3
3
5
.9, 24.2 Hz, 1F).
X-r a y Cr ysta l Str u ctu r e for 4a c. A colorless prismatic
Su p p or tin g In for m a tion Ava ila ble: ORTEP drawing of
erythro-4a c, NMR spectra, and F NMR chemical shift data.
This material is available free of charge via the Internet at
http://pubs.acs.org.
crystal (from hexane-diethyl ether) of approximate dimen-
sions of 0.50 × 0.20 × 0.10 mm was mounted on a glass fiber
with epoxy cement. Data were collected on a Rigaku AFC7R
diffractometer with graphite-monochromated Cu KR radiation
and a rotating anode genetator. The material crystallized in
19
J O0358729
J . Org. Chem, Vol. 69, No. 15, 2004 5047