Chemo- and regioselectivity in the reactions between highly electrophilic fluorine containing dicarbonyl compounds and amines. Improved synthesis of the corresponding imines/enamines

Article abstract of DOI:10.1016/S0040-4020(03)00138-8

Chemo- and regioselectivity in the reactions between highly electrophilic fluorine containing dicarbonyl compounds (ethyl 4,4,4-trifluoroacetoacetate, 3,3,3-trifluoropyruvate and 1,1,1,5,5,5-hexafluoropentane-2,4-dione) and various benzylamines were systematically studied. The results obtained lead to the development of a generalized and practical method for large-scale synthesis of the corresponding imines/enamines, useful starting materials for preparation fluorinated amines and amino acid.

Full text of DOI:10.1016/S0040-4020(03)00138-8

Tetrahedron 59 (2003) 1647–1656
Chemo- and regioselectivity in the reactions between highly
electrophilic fluorine containing dicarbonyl compounds and
amines. Improved synthesis of the corresponding imines/enamines
*
Hironari Ohkura, Dmitrii O. Berbasov and Vadim A. Soloshonok
Department of Chemistry and Biochemistry, University of Oklahoma, 620 Parrington Oval, Norman, OK 73019, USA
Received 19 November 2002; revised 23 January 2003; accepted 23 January 2003
Abstract—Chemo- and regioselectivity in the reactions between highly electrophilic fluorine containing dicarbonyl compounds (ethyl 4,4,4-
trifluoroacetoacetate, 3,3,3-trifluoropyruvate and 1,1,1,5,5,5-hexafluoropentane-2,4-dione) and various benzylamines were systematically
studied. The results obtained lead to the development of a generalized and practical method for large-scale synthesis of the corresponding
imines/enamines, useful starting materials for preparation fluorinated amines and amino acid. q 2003 Elsevier Science Ltd. All rights
reserved.
1. Introduction
adequately correlated by the Hammett equation.10 We were
first to demonstrate that the presence of electron-with-
drawing perfluoroalkyl or perfluoroaryl groups, in a-pos-
ition to the imine function in derivatives 2 makes their
base-catalyzed isomerization to Schiff bases 3 virtually
irreversible and thus synthetically useful.
The 1,3-proton shift reaction (PSR)1 has emerged as
conceptually different to the classical methods,2 a conven-
tional-reducing-reagent-free approach for a reductive amin-
ation of fluorocarbonyl compounds to the corresponding
fluorine-containing amines and amino acids.3 – 7 This
approach, mimicking the biological transamination,8 i.e.
the enzyme-catalyzed interconversion of a-amino and
a-keto carboxylic acids,9 represents the most ideal solution
to the reductive amination of carbonyl compounds 1
(Scheme 1). Thus, instead of application of reducing
reagents, PSR makes use of the intramolecular reduction–
oxidation process via a base-catalyzed 1,3-proton shift in
the azaallylic system of azomethines (imines) 2 and 3. It was
shown that the mechanism of this azomethine–azomethine
isomerization involves azaallylic anions as intermediates
and the equilibrium constants of the isomerization are
Previously we reported an efficient application of this base-
catalyzed azomethine–azomethine isomerization for
preparation of fluorine-containing amines,4 a- and
b-amino acids5 starting from readily available fluorinated
aldehydes and ketones, or a- and b-keto carboxylic acids,
respectively. Our most recent achievement in this area is the
development of double-PSR methodology for a direct, one-
pot conventional-reducing-reagent-free transformation of
perfluoroalkyl-carboxylic acids to the corresponding a,a-
dihydroperfluoroalkylamines.11 However, of particular
interest are the results reported by other research groups
Scheme 1.
Keywords: chemoselectivity; regioselectivity; nucleophilicity; electrophilicity; enamines; imines; fluorine and compounds.
*
Corresponding author. Tel.: þ1-405-325-8279; fax: þ1-405-325-6111; e-mail: vadim@ou.edu
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00138-8

1648
H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
Scheme 2.
on application of the PSR methodology for transamination
of fluorine-free carbonyl compounds to the corresponding
amino-derivatives6 as well as the preparation of fluorine-
containing phosphorus analogs of a- and b-amino acids.7
type 10 (Scheme 2) via DBU-catalyzed PSR of both
enamino-ester 11a and enamino-amid 12 to the correspond-
ing Schiff bases 13 and 14.5h While the isomerization of 11a
and 12 to 13 and 14 as well as sequential hydrolysis of 13
and 14 to the target compound 10 could be conducted with
high chemical yields, the method, as a whole, is compro-
mised by the low chemoselectivity and chemical yields on
the stage of preparing the starting ester 11a and amide 12.
Despite the apparent generality and synthetic efficiency of
the base-catalyzed isomerization of 2 to 3 and their further
hydrolysis to target amino compounds 4, which could be
easily separated from the aldehyde or ketone 5, the PSR
methodology,1,4–7 as a whole process, in some cases is
plagued by the relatively low chemical yields on the stage of
preparation of the corresponding imines/enamines 2 from
the starting carbonyl compounds. In particular, the issue of
chemo- and regioselectivity is a major concern in the
reactions between benzyl amine and its derivatives with
polyfunctional and/or highly electrophilic fluorine contain-
ing carbonyl compounds. In this paper we report a full
account of a systematic study of the reactions of ethyl 4,4,4-
trifluoroacetoacetate (6), ethyl 3,3,3-trifluoropyruvate (7)
and 1,1,1,5,5,5-hexafluoropentane-2,4-dione (8) with
various benzylamines 9a–g, which lead to the development
of a generalized and practical method for large-scale
synthesis of the corresponding imines/enamines, and thus
to substantial overall improvement of the PSR
methodology.
Previously5f,h we synthesized compounds 11a and 12 under
conventional reaction conditions, such as acid-catalyzed
condensation between ethyl trifluoroacetoacetate (6) and
optically pure a-phenylethylamine (9a) in benzene at reflux
using Dean–Stark device to trap the water (Dean–Stark
conditions) (Table 1, entry 1). In contrast to the conden-
sations of fluorine-free ethyl acetoacetate with benzyla-
mines which afforded virtually quantitative yield the
corresponding enamino-esters,14 the reaction under study
showed poor chemoselectivity giving rise to a mixture of
products 11a and 12 in a ratio of 74.5/25.5 (entry 1). Since
both derivatives 11a and 12 can be used for preparation of
amino acid 105h (Scheme 2), we decided to study this
reaction in detail to develop a chemoselective method for
practical synthesis of each 11a and 12.
2.1.1. Chemoselective preparation of enamino-amide 12
(Scheme 3, Table 1). First we targeted preparation of the
enamino-amid 12, as we thought it could be easily achieved
simply by using an excess of amine 9a in the p-toluene
sulfonic acid-catalyzed reaction with keto-ester 6 (Dean–
Stark conditions). Surprisingly, application of 2.1 equiv. of
9a did not effect the ratio of products 11a and 12 (entry 2).
However, changing of the solvent (toluene) noticeably
accelerated the reaction rate and increased the ratio of the
enamino-amid 12 formation (entry 3). Interestingly, further
increase in formation of 12 was observed when the reaction
was conducted without p-toluene sulfonic acid as a catalyst
(entry 4). Under these reaction conditions we tried, once
again, an application of 3.5 equiv. of amine 9a to improve
2. Results and discussion
2.1. Control of chemoselectivity in the reactions between
ethyl 4,4,4-trifluoroacetoacetate (6) and benzylamines
9a–c
b-Perfluoroalkyl-b-amino acids represent an enormously
interesting class of b-amino acids in view of their synthetic
and biological applications,3 and in particular, in the design
and synthesis of fluorinated b-peptides.12,13 Recently we
reported a highly enantioselective (.90% ee) method for
preparing b-perfluoroalkyl-containing b-amino acids of
Table 1. Synthesis of enamino-amide 12 as a major product by the reaction of b-keto ester 6 with a-phenylethylamine 9a
Entry
Solvent
Ratio 6/9a
T (h)
Ratioa 11a/12
Yieldb (%) 12
1
2
3
4
5
6
7
Benzenec
1/1.1
1/2.1
1/2.5
1/2.5
1/3.5
1/3.0
1/3.0
24
24
6
74.5/25.5
71.3/28.7
49.1/50.9
42.7/57.3
42.5/57.5
15.5/84.5
11.3/88.7
17.2d
23.6
43.1
43.1
45.0
81.1
75.4
Benzenec
Toluenec
Toluene
Toluene
Methanol/toluene
Methanol–water/toluene
6
12
3; 3
3; 3
All reactions were conducted at reflux in the indicated solvent.
a
Determined by 19F NMR (300 MHz) analysis of the crude reaction mixtures.
Isolated yield of pure product 12.
Reaction was conducted in the presence of 5 mol% of p-toluene sulfonic acid.
Enamino-ester 11a was isolated in 43% yield.
b
c
d

H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
1649
Table 2. Synthesis of enamino-esters 11a–c by the reactions of b-keto ester
6 with benzylamines a–c
Entry 9a–c
Solvent
Acida
T
Ratiob
(h) 11a–c/12 (%) 11a–c
Yieldc
1
2
3
4
5
6
7
8
a
a
a
a
a
a
a
a
a
a
b
c
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
Benzene
H2CO3
None
9
15 56.1/43.9
68.1/31.9 57.7
d
–
MeCO2H
PhCO2H
CF3CO2H
HCO2H
HCl
9
9
9
9
86.9/13.1 84.3
79.4/20.6 71.3
.99/1
88.7/11.3 87.7
40.5e
24 44.2/55.8 7.9f
n-Hexane MeCO2H
9
9
9
9
9
93.8/6.2
.99/1
95.7/4.3
.99/1
.99/1
58.4g
9
CHCl3
CHCl3
CHCl3
CHCl3
HCO2H
81.1 (83.3)h
91.5 (93.7)h
95.7
10
11
12
MeCO2H
MeCO2H
MeCO2H
96.1
All reactions were conducted at reflux in the indicated solvent using 1:1.1
ratio of 6 and 9a–c.
Scheme 3.
a
The indicated acid was used to form in situ the corresponding salt with
amine 9a–c.
the yield of 12. Unfortunately, the excess of the amine and
even prolonged reaction time did not change the ratio of
products 11 and 12 (entry 5). These results clearly suggested
that enamino-ester 11a does not react with amine 9a to give
amid 12. Indeed, the reaction between pure enamino-ester
11a with amine 9a, conducted with and without p-toluene
sulfonic acid as a catalyst, did not produce any measurable
amounts of amid 12. Based on these results we assumed that
preparation of enamino-amid 12 would require special
reaction conditions under which amine 9a would be forced
to react first with the ester function of 6 to form the amid
moiety and then interact with the keto group of the
corresponding intermediate amid to give the enamino
functionality. To realize this reaction sequence we designed
the following two-step procedure. To block the most
reactive keto function of 6 we decided to conduct the first
stage using methanol as a reaction medium. We expected
that under these conditions the keto of 6 group might react
with methanol to form a less reactive semi-ketal derivative
while the ester function would still be active to react with
amine 9a. On the second stage, the formation of the enamine
moiety, we planned to apply standard conditions, refluxing
the reagents in toluene. Thus, keto-ester 6 was treated first
with 3.0 equiv. of amine 9a in methanol at reflux for 3 h.
After that, the mixture was evaporated, to remove the excess
of methanol. The residue was taken in toluene and refluxed
for 3 h. The result of this two-step procedure was rather
satisfactory as we obtained a significantly improved yield of
the target 12 (entry 6). The best result was obtained when we
used methanol–water in a volume ratio 4 to 2, respectively,
as a reaction medium for the first stage (entry 7). However,
substantial amounts of water in the reaction mixture turned
out to be synthetically disadvantageous, complicating the
isolation of the target product. Thus, the highest and
synthetically useful isolated yield of enamino-amid 12
(81%) on a scale of over 50 g was obtained with the
designed two-step procedure using as solvents methanol and
toluene on the first and second steps, respectively (entry 6).
b
Determined by 19F NMR analysis of the crude reaction mixtures.
Isolated yield of pure 11a–c.
c
d
Only 12 was isolated and characterized.
Conversion of 6 and 9a was about 50%.
Conversion of 6 and 9a was about 15%.
Conversion 6 and 9a was about 70%.
Yield obtained on large scale.
e
f
g
h
also achieved by decreasing the nucleophilicity of the
reacting amine. Our reasoning was based on the assumption
that decreased nucleophilicity of the amine might prevent,
or significantly de-accelerate, the rate of its reaction with the
ester group while the condensation with a more electrophilic
keto function would still be possible. Therefore, to decrease
nucleophilicity of amine 9a we decided to use its salts with
relatively weak acids. First we conducted the reaction
between keto-ester 6 and carbonate of 9a. The result was
rather disappointing as we obtained 68:32 ratio of the target
enamino-ester 11a and enamino-amid 12 (entry 1).
However, we reasoned that this result could be attributed to
the relative instability of the corresponding carbonate of 9a
in boiling benzene, so the observed ratio of the products 11a
and 12 might be rather an outcome of the direct reaction
between keto-ester 6 and free amine 9a. This assumption
was supported by the corresponding reaction that gave a
comparable ratio of the products 11a and 12 (entry 2). In
contrast, the reaction of 6 with acetate of 9a afforded the
products with a substantially increased ratio of enamino-
ester 11a (entry 3). Further reactions of keto-ester 6 with
benzoic (entry 4), trifluoroacetic (entry 5), formic (entry 6),
and hydrochloric (entry 7) acid derived salts of amine 9a
revealed that while the best, virtually complete, regio-
selectivity could be obtained using the trifluoroacetate of 9a
(entry 5), for preparative purposes the corresponding acetate
(entry 3) and formate (entry 6) of 9a should be reagents of
choice. In these cases, in contrast to the conventional
method, we did not observe separation of water; therefore,
we monitored the reaction completion by 19F NMR. To
further improve the regioselectivity, we tried the reactions
in different solvents using these non-conventional reaction
conditions (no separation of water). For instance, appli-
cation of acetate of 9a in the reaction with 6 in hexane, at
reflux, showed higher regioselectivity as compared with the
result obtained in benzene (entry 8 vs 3). However, the
2.1.2. Chemoselective preparation of enamino-esters
11a–c (Scheme 3, Table 2). For chemoselective prep-
aration of enamino-ester 11a we needed to solve just the
opposite problem: to increase reactivity of the keto group in
6 and decrease reactivity of the corresponding ester
function. We envisioned that the desired result could be

1650
H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
most synthetically useful results were obtained in the
reactions conducted in chloroform. Thus, the reaction of
keto-ester 6 with formate of amine 9a, conducted in boiling
chloroform, featured virtually complete regioselectivity
(entry 9); however, the target product 11a was isolated in
81% yield. Application of acetate of 9a, under the same
reaction conditions, was also synthetically useful, affording
the products 11a and 12 with a ratio of .95:5 (entry 10).
Comparison of these two procedures on a large scale
(.100 g) showed that the acetate of 9a (entry 10) is a
reagent of choice for preparing large quantities of enamino-
ester 11a.
Using these findings we conducted the reaction between
keto-ester 6 and acetate of benzylamine 9b. The result was
almost perfect from the point of view of regioselectivity and
chemical yield of the target enamino ester (entry 11). The
same, virtually complete, regioselectivity and high chemical
yield were observed in the reaction of the corresponding
acetate of p-(methoxy)benzylamine 9c with keto-ester 6
(entry 12).
Scheme 5.
Table 3. Synthesis of imines 18a,b by the reactions of a-keto ester 7 with
benzylamines 9a,b
Entry 9a,b
Solvent
Acida
T
Ratiob
Yieldc
(h) 18a,b/20a,b (%) 18a,b
1
2
3
4
5
6
a
a
b
b
b
b
CHCl3
CHCl3
CHCl3
CHCl3
Toluene MeCO2H
CHCl3 HCO2H
MeCO2H
MeCO2H
MeCO2H
MeCO2H
9
64
9
64
9
67.8/32.2
.95/5
,5/95
,30/70
.99/1d
69.7/30.3
–
93.3
–
2.2. Control of regioselectivity in the reactions between
highly electrophilic fluorine containing carbonyl
compounds and benzyl amines
–
65.1d
63.0
6
The reactions between fluorinated carbonyl compounds and
amines, affording the corresponding intermediate imines,
could be considered as the most methodologically straight-
forward approach for preparing fluorine-containing and
biologically relevant amines and amino acids.3 However,
with the increase of fluorine substituents on carbonyl
compounds and consequently the electrophilicity of
carbonyl compounds, the chemical outcome of their
reactions with nucleophilic amines, in general, becomes
less and less synthetically useful. Thus, the intermediate
gem-amino-alcohols 16 (Scheme 4), derived from highly
electrophilic carbonyl compounds 14, tend to undergo a
rather haloform type reaction (sometimes referred to as a
haloformic decomposition or Lieben haloform reaction15),
giving rise to trifluoroacetamides 15, then the dehydration
reaction leading to the target imines 16.16 Therefore, in
general, the regioselectivity in the reactions of highly
electrophilic fluorine-containing carbonyl compound with
nucleophilic amines represents one of the unsolved
synthetic challenges.
All reactions were conducted at reflux in the indicated solvent using 1:1.1
ratio of 7 and 9a,b.
a
The indicated acid was used to form in situ the corresponding salt with
amine 9a,b.
b
Determined by 19F NMR analysis of the crude reaction mixtures.
Isolated yield of pure products.
The 21b was isolated as the major reaction product.
c
d
the realm of polyfluorinated carbonyl compounds.17 How-
ever, the application of the corresponding N-substituted
phosphazenes 17 has some synthetic disadvantages such as
the preparation of 17 from the corresponding azides and
triphenylphosphine and the laborious purification of pro-
ducts 16 from the triphenylphosphine oxide.17 Taking into
account the successful application of salts 9a–d for
chemoselective preparation of enamines 11a–d (Scheme
3), we decided to study the reactions of these mild
nucleophilic reagents with highly electrophilic ethyl 3,3,3-
trifluoropyruvate (7) and 1,1,1,5,5,5-hexafluoropentane-2,4-
dione (8).
2.2.1. Regioselective synthesis of N-(a-phenyl)ethyl- 18a
and N-benzylimines 18b (Scheme 5, Table 3). Previously
we reported that the reaction between a-keto-ester 7 and
benzylamine 9b, conducted at room temperature, afforded
N-benzylamide 19b as a major (60% yield) reaction
product.17d Lowering the reaction temperature to 2158C
allowed us to avoid the corresponding haloform type
reaction and isolate the gem-amino-alcohol 20b in 90%
chemical yield. Similar to benzylamine 9b, phenylethyl-
amine 9a reacted with keto-ester 7 at low temperatures to
furnish gem-amino-alcohol 20a (95% yield). However, in
contrast to 20b, phenylethylamine-derived 20a was found to
undergo, under the standard Dean–Stark conditions, the
corresponding dehydration affording the target imine 18a in
70% yield.17d Further optimization of the reaction
conditions allowed us to increase the yield of imine 18a to
81–83%.5g However, the necessity of using toluene as a
Previously we successfully addressed the issue of the
regioselectivity by introducing the Staudinger reaction into
Scheme 4.

H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
1651
solvent for these reactions (no reaction was observed in
benzene) caused the formation of noticeable amounts of
haloform-type decomposition of intermediate 20a, thus
complicating the purification of target imine 18a from
trifluoroacetamide 19a.5g,17d Therefore application of the
salts of amines 9a,b in the condensations with keto-ester 7
under milder conditions looked very promising as an
attempt to improve the regioselectivity and thus the
synthetic efficiency of these reactions. Since the salts of
benzylamines 9a–d with acetic acid gave the best synthetic
results in preparation of enamines 11a–d (Schemes 3 and
4), we tested these reagents in the reactions with keto-ester
7.
transformation of the gem-amino alcohol 20b directly to
Schiff base 21b (d, 271.5 ppm), which was isolated in 65%
chemical yield by column chromatography (entry 5). These
results suggested that application of acetate 9b in the
reaction with 7 is more efficient for dehydration of the
intermediate 20b as compared with the standard Dean–
Stark conditions using the solvent (toluene). Another
conclusion that could be drawn from these results is that
N-benzylimine 18b is highly unstable in boiling toluene and
undergoes fast PSR to afford the thermodynamically more
stable product 21b.18 Nevertheless, after several unsuccess-
ful attempts using various salts of benzylamine 9b, we
finally succeeded in preparing N-benzylimine 18b by the
reaction between the trifluoroacetate of benzylamine 9b and
keto-ester 7. Thus after 6 h of the reaction in chloroform at
reflux, we were able to isolate compound 18b as an
individual compound, albeit in moderate 63% yield (entry
6). Our attempts to improve the yield of 18b by increasing
the reaction time unfortunately failed. As could be
expected,4f,g imine 18b on treatment with triethylamine
underwent the corresponding PSR to afford the Schiff base
21b at a much faster rate as compared with that of
compound 18a. However, from the point of view of
synthetic efficiency and overall chemical yield in preparing
the target 3,3,3-trifluoroalanine,5g synthesis of phenylethyl-
amine-derived imine 18a, according to the procedure
developed by this study (entry 2) is a recommended method
of choice.
Fortunately, our first attempt proved our expectations. Thus
monitoring (by 19F NMR) of the reaction conducted in
boiling chloroform between acetate of phenylethylamine 9a
and keto-ester 7, showed that after 9 h of the condensation
only two products, the intermediate gem-amino alcohol 20a
(s, 280.5 ppm) and target imine 18a (s, 268.00 ppm); were
present in the reaction mixture in a ratio of 2:1 (Table 3,
entry 1). These data suggested that under the new reaction
conditions the intermediate gem-amino-alcohol 20a slowly
but regioselectively undergoes dehydration to afford imine
18a. The complete transformation of 20a to 18a (as detected
by 19F NMR) was achieved after about 3 days of refluxing
the reaction mixture in chloroform (entry 2). Analysis of the
reaction mixture by 19F NMR showed that target imine 18a
could account for at least 95% of the crude product. Among
the other compounds in the reaction mixture we could
identify gem-amino alcohol 20a (,3%) and product of
imine 18a isomerization, the Schiff base 21a (d, 272.6)
(,2%). Target imine 18a was isolated in 93% chemical
yield, by passing the crude mixture through a short silica-gel
column. According to previously published procedures,5g
imine 18a was cleanly isomerized to afford 21a (94% yield).
With these promising results in hand, we studied next the
reactions of benzyl amine salts 9b with a-keto-ester 7.
2.2.2. Regioselective synthesis of imines 22a,b,d–f
derived from 1,1,1,5,5,5-hexafluoropentane-2,4-dione
(8) and amines 9a,b,d–f (Scheme 6, Table 4). While the
reactions of hexafluoropentane-2,4-dione 8 with low-
nucleophilic and polyfunctional arylamines are widely
used for preparing various heterocyclic compounds,19 the
condensations of 8 with highly nucleophilic aliphatic
amines are of much less synthetic value as they usually
lead to haloform reaction products. For instance, previously
we failed to prepare the target enamine 22b by the reaction
between diketone 8 and benzylamine 9b using the
conventional Dean–Stark conditions. Instead, we used this
reaction to produce in situ highly volatile trifluoroacetone
for its further condensation with excess of 9b to afford the
correspondent N-benzylimine 23 (Scheme 6).4e Interest-
ingly, the yield of the expected enamine 22b in this reaction
was only about 11%. Therefore we were very interested to
study the synthetic value of the salts of benzylamines 9 as
As discussed previously, all our former attempts to
dehydrate the gem-amino alcohol 20b, under the standard
Dean–Stark conditions, failed.17d Therefore, we were not
surprised to observe (controlled by 19F NMR) the stability
of compound 20b (s, 279.8 ppm) under the conditions that
caused dehydration of phenylethylamine derivative 20a
(entries 3,4 vs 1,2). Interestingly, the reaction of acetate 9b
with 7 conducted in toluene at reflux, resulted in substantial
Scheme 6.

1652
H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
Table 4. Synthesis of enamines 22a,b,d–f by the reactions of diketone 8 with benzylamines 9a,b,d–f
Entry
9a,b,d–f
Solvent
Acida
T (h)
Ratiob 8/22a,b,d–f
Yieldc (%) 22a,b,d–f
1
2
3
4
5
6
7
8
9
10
a
a
ae
a
a
a
b
d
e
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
Toluene
Toluene
Toluene
Toluene
Toluene
MeCO2H
MeCO2H
MeCO2H
CF3CO2H
CF3CO2H
CF3CO2H
CF3CO2H
CF3CO2H
CF3CO2H
CF3CO2H
15
120
96
15
120
9
9
9
9
9
64.1/35.9
,5/95d
,2/98f
,51/49
.99/1g
.93/7
44/56
94/6
.99/1
95/5
–
55.5
–
–
57.1
65.4
77.7
65.3
62.5
62.2
f
All reactions were conducted at reflux in the indicated solvent using 1:1.1 ratio of 8 and 9a,b,d–f.
The indicated acid was used to form in situ the corresponding salt with amine 9a,b,d–f.
a
b
Determined by 19F NMR analysis of the crude reaction mixtures.
Isolated yield of pure products.
Content of product 22a in the reaction mixture was about 85%.
Compounds 8 and acetate 9a were used in a ratio 1:2.2.
Content of 22a in the reaction mixture was ,70%.
Content of 22a in the reaction mixture was .83%.
c
d
e
f
g
low-nucleophilic reagents for preparing the corresponding
enamines 22a,b,d–f, a virtually unstudied but potentially
synthetically useful class of intermediates for preparing
polyfunctional fluorinated amino-compounds. Based on the
results obtained in this study (Tables 2 and 3), we decided
first to react diketone 8 with acetate 9a in chloroform at
reflux. The reaction proceeded quite slowly as after 15 h we
observed less than 40% conversion of starting 8 (Table 4,
entry 1). Complete transformation of the starting com-
pounds was achieved after 5 days of the reaction and,
according to the 19F NMR of the crude reaction mixture, the
ratio of the target product 22a to the rest of byproducts was
85:15. The target enamine 22a was isolated with 55.5%
chemical yield (entry 2).
the method using various trifluoroacetate salts of benzyl-
amines 9b,d–f for preparing the corresponding enamines
22b,d–f under the standard conditions (entry 6). The best
result, 77.7% yield of enamine 22b, was obtained in the
reaction of diketone 8 with the trifluoroacetate of unsub-
stituted benzylamine 9b (entry 7). On the other hand, in the
reactions of the corresponding trifluoroacetates of the
benzylamines 9d,e, bearing strong electron-withdrawing
substituents (entries 8, 9), or sterically bulky moiety 9f
(entry 10), we isolated the target enamines 22d–f in a range
of 60–65% chemical yield (entries 8–11). Since the content
of the target products 22b,d–f in the reaction mixtures, as
estimated by 19F NMR of the crude reaction product, is
generally above 90%, we believe that the isolated
yields could be further improved by proper optimization
of the work-up procedure. Nevertheless, comparison of
the yield of about 11% of the target enamine 22b,
obtained under the conventional Dean–Stark reaction
conditions,4e with the generally above 60% yields of
enamines 22a,b,d–f, obtained using trifluoroacetates
9a,b,d–f, renders the method developed in this study
synthetically useful.
The increase in the ratio of the acetate of 9a, relative to
diketone 8, as an attempt to improve the yield of 22a,
resulted in increased formation of the byproducts. Accord-
ing to the 19F NMR of the crude reaction mixture, the
content of the target product was ,70% (entry 3).
Therefore, we decided to use less nucleophilic trifluoro-
acetate of amine 9a. The reaction between diketone 8 and
trifluoroacetate of 9a surprisingly proceeded at a compar-
able with the interaction of 8 with acetate of 9a reaction rate
(entry 4 vs 1). However, the important feature of this
reaction was the absence of unwanted byproducts. Con-
tinuation of this reaction, in the chloroform at reflux, for 5
days gave a mixture containing, according to the 19F NMR,
at least 83% of the target product (entry 5). After numerous
attempts, using different salts of 9b and different solvents,
we finally found that toluene, as a reaction medium, and
trifluoroacetate of 9a, as the amino derivative in the reaction
with diketone 8, are the conditions of choice to prepare the
target 22a with a synthetically useful yield (entry 6).
According to the 19F NMR of the crude reaction mixture
(entry 6), the content of the target product 22a was at least
93%. Work-up of the reaction mixture (entry 6) allowed us
to isolate the product 22a in 65% yield. To demonstrate a
practicality of the developed reaction conditions, using
trifluoroacetate of 9a as a reagent, we performed the
reaction on a relatively large scale (20 g) with a successful
reproducibility in the chemical yield of the enamine 22a.
With these results in hand, we next studied the generality of
In summary, we demonstrated that the problem of
chemoselectivity in the reaction between ethyl 4,4,4-
trifluoroacetoacetate (6) and various benzylamines 9a–c
could be successfully solved by designing the appropriate
reaction conditions (Tables 1 and 2), allowing for highly
chemoselective preparation of enamino-ester 11 and
enamino-amid 12. The acetates and trifluoroacetates of
benzylamines 9a–f, used as ‘mild nucleophiles’ for
chemoselective preparation of enamino-ester 11, were
found to be the reagents of choice in the reactions with
highly electrophilic fluorinated carbonyl compounds, such
as a-keto-ester 7 and diketone 8, allowing the preparation
of the corresponding imines 18a,b and enamines 22a,b,d–f
in synthetically useful chemical yields (Tables 3 and 4).
Simplicity of the experimental procedures as well as
the relatively high chemical yields, compared to the
conventional methods, render the procedures developed in
this study synthetically useful for preparing various
fluorine-containing and biologically relevant amino-
compounds.

H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
1653
3. Experimental
J¼7.2 Hz), 4.75 (dd, 1H, J¼8.7, 6.9 Hz), 5.12 (s, 1H),
7.25–7.34 (m, 5H), 8.63 (brd, 1H, J¼6.9 Hz). 19F NMR d
266.75 (s). 13C NMR d 14.35 (s), 25.00 (s), 53.99 (q,
JCF¼2.6 Hz), 59.77 (s), 85.60 (q, JCF¼6.0 Hz), 120.33 (q,
JCF¼277.1 Hz), 125.36 (s), 127.31 (s), 128.73 (s), 143.94
(s), 147.75 (q, JCF¼31.1 Hz), 169.97 (s). MS: 287 (M, 5.9),
258 (M2Et, 4.2), 105 (100). HRMS calcd for C14H16F3NO2
(MþNa) 310.1031. Found: 310.1087. Anal. calcd for
C14H16F3NO2: C, 58.53; H, 5.61; N, 4.88; F, 19.84.
Found: C, 58.58; H, 5.63; N, 4.84; F, 19.77.
3.1. General
Unless otherwise noted, all reagents and solvents were
obtained from commercial suppliers and used without
further purification. All the reactions were carried out
under atmosphere without any special caution to exclude
1
air. Unless indicated, H, 19F and 13C NMR spectra, were
taken in CDCl3 solutions at 299.95, 282.24 and 75.42 MHz,
respectively, on an instrument in the University of
Oklahoma NMR Spectroscopy Laboratory. Chemical shifts
refer to TMS and CFCl3 as the internal standards.
Enamino-esters 11b,c were prepared according to the
procedure described above for the synthesis of 11a, except
that the acetates of benzylamines 9b,c (Table 2, entries 11,
12) were used instead of the acetate of 9a.
Yields refer to isolated yields of products of greater than
95% purity as estimated by 1H and 19F NMR spectrometry.
All new compounds were characterized by H, 19F, 13C
NMR, by mass spectrometry and/or elemental analysis.
1
3.1.3. 3-Benzylamino-4,4,4-trifluorobut-2-enoic acid
ethyl ester (11b) (Scheme 3, Table 2, entry 11). 1H
NMR d 1.26 (3H, t, J¼7.2 Hz), 4.13 (2H, q, J¼7.2 Hz), 4.47
(2H, d, J¼6.3 Hz), 5.16 (1H, s), 7.5–7.2 (5H, m), 8.43 (1H,
brs). 19F NMR d 267.0 (CF3, s). 13C NMR d 169.6, 147.9
(q, J¼31.1 Hz), 137.5, 128.7, 127.6, 127.1, 120.2 (q,
J¼274.7 Hz), 85.2 (q, J¼5.9 Hz), 59.7, 48.1 (q,
J¼2.7 Hz), 14.4. HRMS calcd for C13H14F3NO2 (MþNa)
296.0874. Found: 296.0755.
3.1.1. 4,4,4-Trifluoro-3-(1-phenylethylamino)but-2-enoic
acid (1-phenylethyl)amide (12) (Scheme 3, Table 1, entry
6). To a solution of keto-ester 6 (50.0 g, 0.27 mol) in
methanol (300 mL) at room temperature was added
phenylethylamine (98.7 g, 0.81 mol). The resultant mixture
was refluxed for 3 h and evaporated in vacuum. To the
slurry residue toluene (500 mL) was added and the mixture
was refluxed for 3 h using Dean–Stark trap to collect the
separating water. The solvent was evaporated in vacuum to
dryness and the residue was subjected to chromatography on
silica gel using first neat n-hexane, to wash out the
corresponding enamino-ester 11a, and then n-hexane–
AcOEt in a ratio of 10:1 to isolate the target enamino-
amid 12 (79.7 g, 81.1%). Rf¼0.14 (n-hexane/ethyl acetate in
a ratio of 4/1); 1H NMR d 1.49 (d, 1.5H, J¼6.6 Hz), 1.4985
(d, 1.5H, J¼6.9 Hz), 1.509 (d, 1.5H, J¼6.6 Hz), 1.512 (d,
1.5H, J¼6.9 Hz), 4.66 (dq, 1H, J¼10.8, 6.6 Hz), 4.95 (s,
0.5H), 4.96 (s, 0.5H), 5.16 (dq, 1H, J¼7.5, 6.9 Hz), 5.50
(brd, 1H, J¼7.5 Hz), 7.25–7.35 (m, 10H), 9.19 (brd, 1H,
J¼10.8 Hz). 19F NMR d 266.64 (s). 13C NMR d 22.12 (s),
22.22 (s), 25.18 (s), 48.50 (s), 53.89 (bq, JCF¼2.1 Hz), 88.05
(q, JCF¼5.7 Hz), 88.26 (q, JCF¼5.5 Hz), 120.51 (q,
JCF¼277.1 Hz), 125.44 (s), 126.04 (s), 126.07 (s), 127.01
(s), 127.36 (s), 128.57 (s), 128.59 (s), 128.74 (s), 143.31 (s),
143.44 (s), 144.51 (s), 144.63 (s), 145.55 (q, JCF¼30.2 Hz),
145.57 (q, JCF¼30.2 Hz), 168.16 (s). MS: 362 (M, 5.6), 105
(100). HRMS calcd for C20H21F3N2O (MþNa) 385.1504.
Found: 385.1678. Anal. calcd for C20H21F3N2O: C, 66.29;
H, 5.84; N, 7.73; F, 15.73. Found: C, 66.34; H, 5.89; N,
7.76; F, 15.67.
3.1.4. 4,4,4-Trifluoro-3-(4-methoxybenzylamino)but-2-
enoic acid ethyl ester (11c) (Scheme 3, Table 2, entry
12). 1H NMR d 8.35 (1H, brs), 7.20 (2H, d, J¼8.6 Hz), 6.86
(2H, d, J¼8.6 Hz), 5.14 (1H, s), 4.38 (2H, d, J¼5.7 Hz),
4.11 (2H, q, J¼7.4 Hz), 3.76 (3H, s), 1.23 (3H, t, J¼7.4 Hz).
19F NMR d 266.4 (CF3, s). 13C NMR d 169.9, 159.3, 148.2
(q, J¼30.8 Hz), 129.8, 128.9, 120.5 (q, J¼274.9 Hz), 114.4,
85.2 (q, J¼5.9 Hz), 60.0, 55.6, 47.9 (q, J¼2.9 Hz), 14.7.
HRMS calcd for C14H16F3NO3 (MþNa) 326.0980. Found:
326.0818.
3.1.5. 3,3,3-Trifluoro-2-(1-phenylethylimino)propionic
acid ethyl ester (18a) (Scheme 5, Table 3, entry 2). The
procedure described above for the synthesis of enamino-
ester 11a was followed except that the completion of the
reaction between keto-ester 7 and acetate of 9a required
1
64 h (control by 19F NMR): H NMR d 7.4–7.3 (5H, m),
4.92 (1H, q, J¼6.5 Hz), 4.40 (2H, q, J¼7.4 Hz), 1.56 (3H, d,
J¼6.5 Hz), 1.37 (3H, t, J¼7.4 Hz). 19F NMR d 269.5 (CF3,
s). Anal. calcd for C13H14F3NO2: C, 57.14; H, 5.16; F,
20.86; N, 5.13. Found: C, 57.24; H, 5.21; F, 20.79; N, 5.09.
3.1.6. 2-Benzylimino-3,3,3-trifluoro-propionic acid ethyl
ester (18b) (Scheme 5, Table 3, entry 5). The procedure
described above for the synthesis of enamino-ester 11a was
followed except that the reaction was conducted in toluene
at reflux (control by 19F NMR); 1H NMR d 7.4–7.2 (5H, m),
4.91 (2H, d, J¼6.5 Hz), 4.42 (2H, q, J¼6.9 Hz), 1.39 (3H, t,
J¼6.9 Hz). 19F NMR d 269.5 (CF3, s). Anal. calcd for
C12H12F3NO2: C, 55.60; H, 4.67; F, 21.99; N, 5.40. Found:
C, 55.72; H, 4.73; F, 21.81; N, 5.35.
3.1.2. 4,4,4-Trifluoro-3-(1-phenylethylamino)but-2-enoic
acid ethyl ester (11a) (Scheme 3, Table 2, entry 10). To a
solution of acetic acid (166.2 g, 2.77 mol) in the chloroform
(700 mL) at room temperature was added phenylethylamine
(335.7 g, 27.7 mol), the resultant mixture was stirred for
5 min and a solution of keto-ester 6 (464.4 g, 2.52 mol) in
chloroform (800 mL) was added to the mixture. The
resultant mixture was refluxed for 3 h followed by
evaporation of the solvent in vacuum. The residue was
placed on short silica gel column and washed with n-hexane
to afford the desired product (678.9 g, 93.7%). Rf¼0.39
(n-hexane/ethyl acetate in a ratio of 4/1); 1H NMR d 1.29 (t,
3H, J¼7.2 Hz), 1.54 (d, 3H, J¼6.9 Hz), 4.18 (qm, 2H,
3.1.7. 1,1,1,5,5,5-Hexafluoro-4-(1-phenylethylamino)-
pent-3-en-2-one (22a) (Scheme 6, Table 4, entry 6). To
a solution of trifluoroacetic acid (17.3 g, 0.29 mol) in
toluene (40 mL) at room temperature was added phenyl-
ethylamine 9a (34.93 g, 0.29 mol) resulting in a formation

1654
H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
of white precipitate. To a resultant mixture a solution of
diketone 8 (50.0 g, 0.24 mol) in toluene (100 mL) was
added and the reaction vessel was sealed (we used an
autoclave with a Teflon tap). The mixture was heated at
958C (oil bath) for 9 h. After that, the reactor was cooled
down, opened and the solvent was evaporated under reduced
pressure. The crude reaction mixture was subjected to a
silica gel chromatography (n-hexane/AcOEt) to afford the
Acknowledgements
The work was supported by the start-up fund provided by
the Department of Chemistry and Biochemistry, University
of Oklahoma.
1
target compound 22a (48.9 g, 65.4%). H NMR d 11.05
References
(1H, brs), 7.5–7.2 (5H, m), 5.82 (1H, s), 4.94 (1H, dq,
J¼10.5, 7.1 Hz), 1.65 (3H, d, J¼7.1 Hz). 19F NMR d 266.0
(CF3, s), 277.0 (CF3, s). 13C NMR d 179.6 (q, J¼35.0 Hz),
152.6 (q, J¼32.6 Hz), 141.2, 128.9, 128.0, 125.2, 119.0 (q,
J¼276 Hz), 116.6 (q, J¼286 Hz), 85.8 (qm, J¼5.1 Hz),
55.6 (q, J¼7.6 Hz), 24.2. HRMS calcd for C13H11F6NO
(MþNa) 334.0643. Found: 310.1087.
1. Soloshonok, V. A. Biomimetic Reducing Agent-Free Reduc-
tive Amination of Fluoro-Carbonyl Compounds. Practical
Asymmetric Synthesis of Enantiopure Fluoro-Amines and
Amino Acids. In Asymmetric Fluoro-Organic Chemistry:
Synthesis, Applications, and Future Directions. ACS Books;
Ramachandran, P. V., Ed.; American Chemical Society:
Washington, DC, 1999; pp 74–83 Chapter 6.
Enamines 22b,d–f were prepared using the standard
conditions described for synthesis of 22a. Isolated yields
for products 22b,d–f are given in Table 4.
2. For reviews on reductive aminations see: (a) Emerson, W. S.
Organic Reactions; Wiley: New York, 1948; Vol. 14. Chapter
3. (b) Gibson, M. S. In The Chemistry of the Amino Group;
Patai, S., Ed.; Interscience: New York, 1968. (c) Dayagi, S.;
Degani, Y. In The Chemistry of the Carbon–Nitrogen Double
Bond; Patai, S., Ed.; Wiley: New York, 1970. (d) Lane, C. F.
Synthesis 1975, 135. (e) Gribble, G. W.; Jasinski, J. M.;
Pellicone, J. T.; Panetta, J. A. Synthesis 1978, 766.
(f) Hutchins, R. O.; Natale, N. Org. Prep. Proced. Int. 1979,
11(5), 20. For publications: (g) Barney, C. L.; Huber, E. W.;
McCarthy, J. R. Tetrahedron Lett. 1990, 31, 5547. (h) Love,
B. E.; Ren, J. J. Org. Chem. 1993, 58, 5556. (i) Abdel-Magid,
A.; Carson, K. G.; Harris, B. D.; Maryanoff, C. A.; Shah, R. D.
J. Org. Chem. 1996, 61, 3849.
3.1.8. 4-Benzylamino-1,1,1,5,5,5-hexafluoropent-3-en-2-
one (22b) (Scheme 6, Table 4, entry 7). 1H NMR d 10.71
(1H, brs), 7.5–7.2 (5H, m), 5.89 (1H, s), 4.66 (2H, d,
J¼6.0 Hz). 19F NMR d 266.6 (CF3, s), 277.1 (CF3, s). 13
C
NMR d 179.6 (q, J¼34.9 Hz), 153.4 (q, J¼32.2 Hz), 135.1,
129.0, 128.4, 127.3, 119.0 (q, J¼276.4 Hz), 116.5 (q,
J¼285.5 Hz), 86.1 (qm, J¼5.0 Hz), 49.1 (q, J¼2.9 Hz).
HRMS calcd for C12H6F6NO (MþNa) 320.0486. Found:
320.0667.
3.1.9. 1,1,1,5,5,5-Hexafluoro-4-(4-trifluoromethylbenzyl-
amino)pent-3-en-2-one (22d) (Scheme 6, Table 4, entry
8). 1H NMR d 10.72 (1H, brs), 7.67 (2H, d, J¼8.4 Hz), 7.42
(2H, d, J¼8.4 Hz), 5.94 (1H, s), 4.72 (2H, d, J¼6.3 Hz). 19F
3. For general reviews on synthesis and biological importance/
applications of fluorine-containing amines and amino acids see
the following monographs: (a). In Fluorine-Containing Amino
Acids. Synthesis and Properties; Kukhar, V. P., Soloshonok,
V. A., Eds.; Wiley: Chichester, 1994. (b) In Biomedical
Frontiers of Fluorine Chemistry. ACS Books; Ojima, I.,
McCarthy, J. R., Welch, J. T., Eds.; American Chemical
Society: Washington, DC, 1996. (c) In Enantiocontrolled
Synthesis of Fluoro-Organic Compounds: Stereochemical
Challenges and Biomedicinal Targets; Soloshonok, V. A.,
Ed.; Wiley: Chichester, 1999. (d) Welch, J. T.; Eswarakrischnan,
S. Fluorine in Bioorganic Chemistry; Wiley: New York, 1991.
(e) In Selective Fluorination in Organic and Bioorganic
Chemistry; Welch, J. T., Ed.; American Chemical Society:
Washington, DC, 1991. (f) In Biomedicinal Aspects of Fluorine
Chemistry; Filler, R., Kobayashi, Y., Eds.; Elsevier: Amsterdam,
1982. (g) Sieler, M.; Jung, M. J.; Koch-Waser, J. Enzyme-
Activated Irreversible Inhibitors; Elsevier: Amsterdam, 1978.
(h) In Organofluorine Chemistry: Principles and Commercial
Applications; Banks, R. E., Smart, B. E., Tatlow, J. C., Eds.;
Plenum: New York, 1994.
NMR d 262.6 (CF3, s), 266.5 (CF3, s), 277.1 (CF3, s). 13
C
NMR d 180.1 (q, J¼34.9 Hz), 153.6 (q, J¼32.2 Hz), 139.4,
130.7 (q, J¼32.6 Hz), 127.4, 126.0 (q, J¼3.8 Hz), 123.8 (q,
J¼269.8 Hz), 119.0 (q, J¼276.0 Hz), 116.4 (q,
J¼285.8 Hz), 86.7 (qm, J¼5.0 Hz), 48.4 (q, J¼2.9 Hz).
HRMS calcd for C13H8F9NO (MþNa) 388.03598. Found:
388.056.
3.1.10. 1,1,1,5,5,5-Hexafluoro-4-(4-nitrobenzylamino)-
pent-3-en-2-one (22e) (Scheme 6, Table 4, entry 9). Mp
53–548C. 1H NMR d 10.66 (1H, brs), 8.20 (2H, d,
J¼8.9 Hz), 7.40 (2H, d, J¼8.9 Hz), 5.89 (1H, s), 4.70
(2H, d, J¼6.9 Hz). 19F NMR d 266.4 (CF3, s), 277.1 (CF3,
s). 13C NMR d 180.2 (q, J¼35.6 Hz), 153.4 (q, J¼32.6 Hz),
147.7, 142.5, 127.8, 124.2, 118.9 (q, J¼276.1 Hz), 116.2 (q,
J¼285.5 Hz), 87.1 (qm, J¼5.0 Hz), 48.1 (q, J¼2.9 Hz).
HRMS calcd for C12H8F6N2O3 (MþNa) 365.03368. Found:
365.0149.
4. For preparation of fluorine-containing amines using PSR
methodology see: (a) Soloshonok, V. A.; Gerus, I. I.;
Yagupol’skii, Y. L.; Kukhar, V. P. Zh. Org. Khim. 1988, 24,
993. Chem. Abstr. 1989, 110, 134824. (b) Kukhar, V. P.;
Soloshonok, V. A.; Galushko, S. V.; Rozhenko, A. B. Dokl.
Akad. Nauk SSSR 1990, 310, 886. Chem. Abstr. 1991, 113,
78920w. (c) Khotkevich, A. B.; Soloshonok, V. A.;
Yagupol’skii, Y. L. Zh. Obshch. Khim. 1990, 60, 1005.
Chem. Abstr. 1991, 113, 171274y. (d) Soloshonok, V. A.;
Kirilenko, A. G.; Kukhar, V. P.; Resnati, G. Tetrahedron Lett.
1994, 35, 3119. (e) Ono, T.; Kukhar, V. P.; Soloshonok, V. A.
3.1.11. 4-(Benzhydrylamino)-1,1,1,5,5,5-hexafluoropent-
3-en-2-one (22f) (Scheme 6, Table 4, entry 10). 1H NMR d
11.34 (1H, brd, J¼7.8 Hz), 7.5–7.2 (10H, m), 6.00 (1H, d,
J¼10.5 Hz), 5.91 (1H, s). 19F NMR d 265.7 (CF3, s), 277.0
(CF3, s). 13C NMR d 180.0 (q, J¼34.9 Hz), 152 (q, J¼
32.6 Hz), 139.8, 137.4, 132.2, 129.9, 128.9, 128.1, 128.1,
126.6, 119.0 (q, J¼277 Hz), 116.5 (q, J¼285.9 Hz), 86.5
(qm, J¼5.0 Hz), 63.1 (q, J¼2.6 Hz). HRMS calcd for
C18H13F6NO (MþNa) 396.0799. Found: 396.0533.

H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
1655
J. Org. Chem. 1996, 61, 6563. (f) Soloshonok, V. A.; Ono, T.
Synlett 1996, 919–921. (g) Soloshonok, V. A.; Ono, T.
Tetrahedron 1996, 52, 14701. (h) Soloshonok, V. A.; Ono, T.
J. Org. Chem. 1997, 62, 3030.
3219. (b) Iverson, B. L. Nature 1997, 385, 113. (c) Seebach,
D.; Matthews, J. L. Chem. Commun. 2015, 1997. (d) Gellman,
S. H. Acc. Chem. Res. 1998, 31, 173. (e) Gademann, K.;
Hintermann, T.; Schreiber, J. V. Curr. Med. Chem. 1999, 6,
905. (f) DeGrado, W. F.; Schneider, J. P.; Hamuro, Y. J. Pept.
Res. 1999, 54, 206. (g) Liu, D.; DeGrado, W. F. J. Am. Chem.
Soc. 2001, 123, 7553.
5. For preparation of fluorine-containing a- and b-amino acids
using PSR methodology see: (a) Soloshonok, V. A.;
Yagupol’skii, Y. L.; Kukhar, V. P. Zh. Org. Khim. 1988, 24,
1638. Chem. Abstr. 1989, 110, 154827b. (b) Soloshonok,
V. A.; Kirilenko, A. G.; Kukhar, V. P.; Resnati, G.
Tetrahedron Lett. 1993, 34, 3621. (c) Soloshonok, V. A.;
Kirilenko, A. G.; Galushko, S. V.; Kukhar, V. P. Tetrahedron
Lett. 1994, 35, 5063. (d) Soloshonok, V. A.; Kirilenko, A. G.;
Fokina, N. A.; Shishkina, I. P.; Galushko, S. V.; Kukhar, V. P.;
Svedas, V. K.; Kozlova, E. V. Tetrahedron: Asymmetry 1994,
5, 1119. (e) Soloshonok, V. A.; Kirilenko, A. G.; Fokina, N. A.;
Galushko, S. V.; Kukhar, V. P.; Svedas, V. K.; Resnati, G.
Tetrahedron: Asymmetry 1994, 5, 1225. (f) Soloshonok, V. A.;
Kukhar, V. P. Tetrahedron 1996, 52, 6953. (g) Soloshonok,
V. A.; Kukhar, V. P. Tetrahedron 1997, 53, 8307.
(h) Soloshonok, V. A.; Ono, T.; Soloshonok, I. V. J. Org.
Chem. 1997, 62, 7538. (i) Soloshonok, V. A.; Soloshonok,
I. V.; Kukhar, V. P.; Svedas, V. K. J. Org. Chem. 1998, 63,
1878. See also Ref. 4f,g.
14. Helio, A. S.; Iguatemi, M. C.; Diogo De, O. S. Synthesis 2000,
1526.
15. (a) Fisher, C. H.; Snyder, H. R.; Fuson, R. C. J. Am. Chem.
Soc. 1932, 54, 3665. (b) Bull, B. A.; Fuson, R. C. J. Am. Chem.
Soc. 1933, 55, 3424. (c) Fuson, R. C.; Bull, B. A. Chem. Rev.
1934, 56, 275. (d) Bartlett, P. D. J. Am. Chem. Soc. 1934, 56,
967. (e) Fuson, R. C.; Tullock, T. W. J. Am. Chem. Soc. 1934,
56, 1638. (f) Sutherland, L. H.; Aston, J. G. J. Am. Chem. Soc.
1939, 61, 241. (g) Parkin, H. Mendel. Bull. 1936, 9, 3.
(h) Arnold, R. T.; Buckles, R.; Stoltenberg, J. J. Am. Chem.
Soc. 1944, 66, 208. (i) Henne, A. L.; Alderson, T.; Newman,
M. S. J. Am. Chem. Soc. 1945, 67, 918.
16. As stated in the text, the low regioselectivity observed in the
reactions between highly electrophilic, fluorine-containing
carbonyl compounds and nucleophilic amines have always
been a problem limiting synthetic application of this
potentially useful reaction. For classical examples see:
(a) McBee, E. T.; Burton, T. M. J. Am. Chem. Soc. 1952,
74, 3902. (b) Hauptstein, M.; Brawn, R. A. J. Am. Chem. Soc.
1955, 82, 2288. (c) Simmons, H. E.; Wiley, D. W. J. Am.
Chem. Soc. 1960, 82, 2288–2296. (d) Fokin, A. V.; Voronkov,
A. N.; Davydova, S. M.; Komarov, V. A. Izv. Akad. Nauk
SSSR, Ser. Khim. 1979, 11, 2598. For relatively recent
examples see: (e) Vilenchik, Ya. M.; Lekontseva, G. I.;
Semerikova, L. S. Zh. Vses. Khim. O-va. 1981, 26, 210.
(f) Gassen, K. R.; Bielefeldt, D.; Marhold, A.; Andres, P.
J. Fluorine Chem. 1991, 149. (g) Braish, T. F.; Fox, D. E. Org.
Prep. Proced. Int. 1991, 23, 655. (h) Gassen, K. R.; Bielefeldt,
D.; Marhold, A.; Andres, P. J. Fluorine Chem. 1991, 149.
(i) lrod, Jr. L.; Spanton, S. G.; Cirovic, M.; Shaffer, D. I.;
Golich, T. G.; Linton, C. L.; Vievia, D. R.; Kalaritis, P.;
Schmand, H. Anal. Chim. Acta 1993, 280, 85. (j) Scheuring, J.;
Kugelbrey, K.; Weinkauf, S.; Cushman, M.; Bacher, A.;
Fischer, M. J. Org. Chem. 2001, 66, 3811.
6. For application of PSR methodology for preparing fluorine-
free amino-compounds see: (a) Johannes, W. G. H.; Johannes,
V. G.; Roeland, N. J. M.; Binne, Z. Tetrahedron Lett. 1995, 36,
3917. (b) Cainelli, G.; Giacomini, D.; Trere, A.; Boyl, P. P.
J. Org. Chem. 1996, 61, 5134. (c) Hjelmencrantz, A.; Berg, U.
J. Org. Chem. 2002, 67, 3585.
7. For application of PSR methodology for preparing fluorinated
P-amino acids see: (a) Chengye, Y. Youji Huaxue 2001, 21,
862. (b) Jingbo, X.; Chengye, Y. Heteroatom Chem. 2000, 11,
541. (c) Jingbo, X.; Xiaomei, Z.; Chengye, Y. Heteroatom
Chem. 2000, 11, 536.
8. Braunstein, A. E.; Kritsmann, M. G. Biochimiya 1937, 242,
859.
9. (a) Greenstein, J. P.; Winitz, M. Chemistry of the Amino Acids;
Krieger: Malabar, FL, 1984; Vol. 1. Chapter 6, pp 569–687
(reprint edition). (b) Snell, E. E. In Chemical and Biological
Aspects of Pyridoxal Catalysis; Fasella, P. M., Braunstein,
A. E., Rossi-Fanelli, A., Eds.; Macmillan: New York, 1963; pp
1–12. (c) In Pyridoxal Catalysis: Enzymes and Model
Systems; Snell, E. E., Braunstein, A. E., Severin, E. S.,
Torchinsky, Yu. M., Eds.; Interscience: New York, 1968.
(d) Jencks, W. P. Catalysis in Chemistry and Enzymology;
McGraw-Hill: New York, 1969; pp 143–145. (e) Guirard,
B. M.; Snell, E. E. Comprehensive Biochemistry; Florkin, M.,
Stotz, E. H., Eds.; Elsevier: New York, 1964; Vol. 15. Chapter
5. (f) Bruice, T. C.; Benkovic, S. J. Bioorganic Mechanisms;
W.A. Benjamin: New York, 1966; Vol. 2. Chapter 8.
10. (a) Smith, P. A. A.; Dang, C. V. J. Org Chem. 1976, 41, 2013.
(b) Jaeger, D. A.; Broadhurst, M. D.; Cram, D. J. J. Am. Chem.
Soc. 1979, 101, 717.
17. (a) Soloshonok, V. A.; Gerus, I. I.; Yagupolskii, Y. L. Zh. Org.
Khim. 1986, 22, 1335. Chem. Abstr. 106, 195861u.
(b) Soloshonok, V. A.; Gerus, I. I.; Yagupolskii, Y. L.;
Kukhar, V. P. Zh. Org. Khim. 1987, 23, 2308. Chem. Abstr.
109,55185p. (c) Soloshonok, V. A.; Gerus, I. I.; Yagupolskii,
Y. L.; Kukhar, V. P. Zh. Org. Khim. 1988, 24, 993. Chem.
Abstr. 110, 134824v. (d) Soloshonok, V. A.; Yagupolskii,
Y. L.; Kukhar, V. P. Zh. Org. Khim. 1988, 24, 1638. Chem.
Abstr. 110, 154827b.
18. This conclusion is in a full agreement with our previous data of
the rates of PSRs; see Ref. 4f,g.
19. For some representative recent publications, see: (a) Kuzueva,
O. G.; Burgart, Ya. V.; Saloutin, V. I.; Chupakhin, O. N.
Chem. Heterocycl. Compd. (Translation of Khimiya Geter-
otsiklicheskikh Soedinenii) 2001, 37, 1130. (b) Burgart, Ya.
V.; Kuzueva, O. G.; Pryadeina, M. V.; Kappe, C. O.; Saloutin,
V. I. Russ. J. Org. Chem. (Translation of Zhurnal Organi-
cheskoi Khimii) 2001, 37, 869. (c) Srinivas, K.; Rao, P. S.;
Narsaiah, B.; Rao, J. M. Indian J. Chem., Sect. B: Org. Chem.
Incl. Med. Chem. 2001, 40B, 191. (d) Singh, S. P.; Kumar, D.;
Batra, H.; Naithani, R.; Rozas, I.; Elguero, J. Can. J. Chem.
2000, 78, 1109. (e) Yang, G.-F.; Lu, R.-J.; Fei, X.-N.; Yang,
11. Soloshonok, V. A.; Ohkura, H.; Uneyama, K. Tetrahedron
Lett. 2002, 43, 5449.
12. Soloshonok, V. A.; Ohkura, H.; Sorochinsky, A.; Voloshin,
N.; Markovsky, A.; Belik, M.; Yamazaki, T. Tetrahedron Lett.
2002, 43, 5445.
13. To the best of our knowledge fluorinated b-peptides have
never been reported in the literature. For the recent reviews
and major contributions in the field synthesis and biological
applications of non-fluorinated b-amino acids see: (a) Cheng,
R. P.; Gelman, S. H.; DeGrado, W. F. Chem. Rev. 2001, 101,

1656
H. Ohkura et al. / Tetrahedron 59 (2003) 1647–1656
H.-Z. Chin. J. Chem. 2000, 18, 435. (f) Saloutin, V. I.; Burgart,
V. V.; Buchanan, III., A. C.; Gakh, A. A. J. Combinat. Chem.
2000, 2, 24. (i) Lopez, J.; Mintz, E. A.; Hsu, F.-L.; Bu, X. R.
Tetrahedron: Asymmetry 1998, 9, 3741. (j) Lee, H.-S.; Kim,
K. Tetrahedron Lett. 1998, 39, 5781.
Y. V.; Kuzueva, O. G.; Kappe, C. O.; Chupakhin, O. N.
J. Fluorine Chem. 2000, 103, 17. (g) Yuh-Wen, H. J. Chin.
Chem. Soc. (Taipei) 1999, 46, 955. (h) Shestopalov, A. M.;
Kislyi, V. P.; Kruglova, E. Ya.; Nikishin, K. G.; Semenov,