Enantioselective organocatalytic route to trifluoromethyl-β-amino acids using chiral bases

Article abstract of DOI:10.1016/j.jfluchem.2006.12.013

Herein are studied new aspects of enantioselective (1,3) proton transfer of ethyl-4,4,4-trifluoroacetoacetate (ETFAA) amino derivatives. When catalyzed by an appropriate chiral base, ee's as high as 71% are observed. Special emphasis is given to mechanist

Full text of DOI:10.1016/j.jfluchem.2006.12.013

Journal of Fluorine Chemistry 128 (2007) 500–506
www.elsevier.com/locate/fluor
Enantioselective organocatalytic route to trifluoromethyl-b-amino
acids using chiral bases
a
Valerie Michaut , Franc¸ois Metz , Jean-Marc Paris , Jean-Christophe Plaquevent
b
b,1
a,
*
´
a
´ `
UMR-CNRS 6014, IRCOF, Universite de Rouen, rue Tesniere, F-76821 Mont-Saint-Aignan Cedex, France
`
Rhodia Recherches, Centre de Recherches de Lyon, 85 avenue des Freres Perret F-69192 Saint-Fons, France
b
Received 3 November 2006; received in revised form 22 December 2006; accepted 28 December 2006
Available online 4 January 2007
Abstract
Herein are studied new aspects of enantioselective (1,3) proton transfer of ethyl-4,4,4-trifluoroacetoacetate (ETFAA) amino derivatives. When
catalyzed by an appropriate chiral base, ee’s as high as 71% are observed. Special emphasis is given to mechanistic insights of the reaction by use of
deuterated derivative. All results converge on a deprotonation as both rate and asymmetric determining step. This study opens a new route to
trifluoromethylated chiral building blocks.
# 2007 Elsevier B.V. All rights reserved.
Keywords: Trifluoromethyl compounds; b-Amino acids and derivatives; Chiral bases; Proton shift transfer reaction; Asymmetric synthesis
1. Introduction
b-Amino acids are of great importance because of their
biological properties, their occurrence in natural products and
as convenient precursors for b-lactams [3].
Taking into account the potential of both fluorinated
compounds and b-amino acids, it is of great interest to
synthesize b-fluorinated b-amino acids in an asymmetric way
[4]. The base-catalyzed 1,3-proton shift reaction ((1,3)PSR)
allows the isomerization of imines 1 to imines 2 to take place
(Fig. 1).
Special features of the fluorine atom make it attractive for
the design of analogs of biologically important compounds. It is
well known that the introduction of fluorine atom strongly
modifies the biological and pharmacological activity in a
molecule [1]. The isosteric and isoelectronic nature of the
fluorine to the hydroxyl group renders fluorinated analogs as
useful pharmacological lead compounds. Today, it is estimated
that about 20–25% of drugs contain at least one fluorine atom
[2].
As stated by Soloshonok et al. [5], the overall process
leading to B from A consists in a biomimetic transamination, in
which there is no need for any added reducing agent (Fig. 1). By
conducting this reaction with b-keto esters as starting materials,
b-amino acids are synthesized. This reaction was intensively
studied by Soloshonok et al. firstly in its non-asymmetric
version for which high yields were obtained [5,6]. Then, the
diastereoselective version of the (1,3)PSR gave good results
when chiral amine was used (Fig. 2) [7]. However, the
enantioselective (1,3) proton transfer catalyzed by a chiral base
was almost unexplored with only three chiral bases tested [8].
In the case of ETFAA derivative, the best enantiomeric excess
was then obtained with (À)-cinchonidine but was only 16%
(Fig. 2). As our group is interested in enantioselective methods
for the access to fluorinated building blocks [9], we decided to
re-examine the enantioselective version of the (1,3)PSR, a
problem that remains unsolved. If successful, this approach
To obtain fluorinated compounds, two main strategies can be
followed: the first one consists in the synthesis of an
unfluorinated precursor of the target molecule, followed by
introduction of fluorine at a late stage of the synthesis using
fluorinating methodologies. The other strategy is to use fluorine
containing starting materials. In this respect, ethyl-4,4,4-
trifluoroacetoacetate (ETFAA) is a good candidate since it is
easy to handle and readily available.
* Corresponding author. Tel.: +33 235522464; fax: +33 235522971.
E-mail address: jean-christophe.plaquevent@univ-rouen.fr
(J.-C. Plaquevent).
1
´
Present address: CNRS-UMR 7573, Ecole Nationale Superieure de Chimie
de Paris, 11 Rue Pierre et Marie Curie, F-75231 Paris Cedex 05, France.
0022-1139/$ – see front matter # 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.jfluchem.2006.12.013

V. Michaut et al. / Journal of Fluorine Chemistry 128 (2007) 500–506
501
Fig. 1. Biomimetic transamination via (1,3) proton shift reaction.
Fig. 2. General strategies for asymmetric biomimetic transamination.
would be more ‘‘chirally economic’’ than the diastereoselective
version, in which the inherent drawback is the self-immolation
of the chiral auxiliary (Fig. 2).
The screening of chiral bases was therefore performed in
toluene at 110 8C (Table 1).
Worthy of note is that with 3a as starting material, reaction
times are quite long. Among the four cinchona alkaloids, the
best enantiomeric excess (20%) is obtained with cinchonine as
chiral base (Table 1, entry 4). The bis-alkaloid (DHQ)2PYR
leads to higher enantiomeric excess (entry 5) and finally the
best result is obtained with an ether derivative, the hydroquinine
4-methyl-2-quinolylether (MQE-DHQN) and reaches 43%
(entry 8).
2. Results and discussion
A first set of experiments was conducted with enamine 3a,
product of the condensation of ETFAA with benzylamine.
Synthesis of 3a from ETFAA was performed according to
Soloshonok’s conditions [10], and yielded directly the enamine
instead of imine 1a, because of internal stabilization by
hydrogen bond formation, but (1,3)PSR proceeds via the
tautomeric imine form 1a (Scheme 1).
Reaction conditions were optimized and several chiral bases
derived from cinchona alkaloids tested (Fig. 3). The reaction
rate is very dependent on both the temperature and the nature of
the solvent. The higher the temperature, the faster the
isomerization rate. In the same way, the reaction time is
greatly shortened in a polar solvent. However, we observed that
the enantiomeric excess evolved in the opposite way and is
lowered in those conditions (high temperature, polar solvent).
The best enantiomeric excess obtained here is much better
than the best one in the literature for the enantioselective
(1,3)PSR (43% versus 16%) [8]. However, in order to improve
this promising result, a better understanding of the mechanism
of the reaction was needed. Therefore, we decided to conduct
the reaction with deuterated enamine d-3a, prepared by
reducing benzonitrile with LiAlD4 and condensing the resulting
deuterated benzylamine with ETFAA following the usual
procedure.
With d-3a as starting material, the (1,3)PSR is performed in
three different kinds of solvents: toluene, acetonitrile and

502
V. Michaut et al. / Journal of Fluorine Chemistry 128 (2007) 500–506
Scheme 1. First set of experiments.
Fig. 3. Chiral bases tested.
butan-1-ol. In all cases, imine d-2a is the only product (Scheme
2); this observation supports a suprafacial deprotonation-
reprotonation mechanism. On the other hand, each time, the
reaction rate is about two times slower with d-3a than with the
hydrogenated starting material 3a: an isotopic effect is
observed and the deprotonation step of the imine d-1a is
considered to be the rate determining step. Thus the observation
that the reaction proceeds faster in a polar solvent is rather
obvious since polarity is expected to favour the deprotonation
step. The fact that a similar result is obtained in protic and in
aprotic solvents suggests that the asymmetric step is based on
an enantioselective discrimination of prochiral benzylic
protons.
As we assumed that the reaction rate depends on the acidity
of the benzylic proton, we decided to study the influence of the
nature of the amine moiety. Increasing the acidity of the
benzylic proton would allow the reaction to be performed in
less drastic conditions in hope of increasing the enantiomeric
excess.
Table 1
Screening of chiral bases
Soloshonok et al. studied this effect for the transamination of
fluorinated b-keto carboxylic ester but only in a non
Entry Chiral base
Reaction time Conv. (%)a ee (%)b Conf.c
1
2
3
4
QN
QD
CD
CN
46 h
46 h
46 h
46 h
92
89
10
3
S
R
R
S
100
49
12
20
5
6
7
(DHQD)2PYR)
(DHQ)2AQN)
(DHQ)2PHAL)
5 d
14
67
0
36
20
–
R
S
–
2.5 d
5 d
8
(MQE-DHQN)
5 d
20
43
S
a
Conversion was determined by 19F NMR.
b
Enantiomeric excess measured by HPLC on chiracel OD-H column with
hexane/iPrOH (99.9/0.1) as eluent.
c
Absolute configuration determined by hydrolysis into the corresponding
amino acid hydrochloride and comparison of optical rotation with literature data
(see Section 4) [11].
Scheme 2. (1,3)PSR with deuterated enamine.

V. Michaut et al. / Journal of Fluorine Chemistry 128 (2007) 500–506
503
Scheme 3. Role of the amine moiety.
Table 2
Screening of enamines 3 and chiral bases; best results obtained
Entry
Enamine
Chiral base
Reaction conditions
Conv. (%)a
ee (%)b
Conf.c
1
2
3b
3b
3b
3b
3b
3b
3c
3d
3e
3e
3e
MQE-DHQN
MQE-DHQN
MQE-DHQN
(DHQD)2PYR
(DHQ)2AQN
(DHQ)2PHAL
MQE-DHQN
CN or MQE-DHQN
MQE-DHQN
(DHQD)2PYR
(DHQ)2AQN
110 8C, 16 h
80 8C, 16 h
50 8C, 5d
80 8C, 16 h
80 8C, 16 h
80 -C, 24 h
50 8C, 48 h
50 to 110 8C, up to 4 days
80 8C, 20 h
80 8C, 48 h
100
100
100
100
87
53
56
56
33
26
71
54
–
S
S
S
R
S
S
–
–
R
S
R
3
4
5
6
7
100
40
8
–
9
44
21
24
33
10
11
15
80 8C, 20 h
61
a
Conversion was determined by 19F NMR.
b
c
Enantiomeric excess measured by HPLC on chiracel OD-H column with hexane/iPrOH (98/2) as eluent.
Absolute configuration determined by hydrolysis into the corresponding amino acid hydrochloride and comparison of optical rotation with literature data (see
Section 4) [11].
asymmetric way [12]. We decided to perform the (1,3)PSR with
enamines obtained by condensation of ETFAA and p-NO2-
benzylamine (enamine 3b), pentafluorobenzylamine (enamine
3c), benzhydrylamine (enamine 3d), and 9-aminofluorene
(enamine 3e) (Scheme 3).
We first conducted the reaction with the enamine 3b at
different temperatures. The reaction rate, as it was predicted, is
accelerated in comparison with the enamine 3a derived from
benzylamine. The screening of chiral bases revealed quite good
results concerning enantiomeric excess. With the ether
derivative of quinine (MQE-DHQN) a 56% enantiomeric
excess is obtained (Table 2, entry 2) and moreover, the bis-
alkaloid (DHQ)2PHAL leads to a 71% enantiomeric excess
(Table 2, entry 6).
It seems that the reaction rate depends linearly on the
concentration of base and this result is also in accordance with
the deprotonation being the rate determining step.
Having those promising results, we continued our study with
the other enamines at different temperatures and with different
chiral bases. Best results are summarized in Table 2.
For 3c, after 48 h, the conversion was only 40% and a 54%
enantiomeric excess was observed (Table 2, entry 7). With 3d,
the reaction did not occur (Table 2, entry 8). In comparison,
with 3e, the isomerization took place. These data can be
rationalized by the fact that in the benzhydrylamine moiety, the
two phenyl rings are allowed to rotate whereas in the
aminofluorene moiety they are blocked. Enamine 3d is
therefore more hindered than 3e and that could explain its
lack of reactivity. With 3e, best enantiomeric excesses were
obtained with (DHQ)2AQN but reached only 33% (Table 2,
entry 11). Moreover, for 2b and 2e, configuration of major
isomer, obtained with a same chiral base, is opposite (Table 2:
entry 1 versus entry 9; entry 4 versus entry 10; entry 5 versus
entry 11), whereas the same configuration is obtained for 2b
with 2a when a same chiral base is used (Table 1, entry 8 versus
Table 2, entry 1; Table 1, entry 5 versus Table 2, entry 4). To
explain this observation, we assume that two different kinds of
enantiotopic discrimination can arise. In the case of imine 1a or
1b, bearing a benzyl moiety, there are two prochiral benzylic
protons and therefore, enantiotopic atoms discrimination takes
place (Fig. 5). In the case of imine 1e there is only one acidic
proton and the chiral base leads to enantiotopic face
discrimination (Fig. 5). It should be noted that this observation
is in accordance with the deprotonation being the asymmetric
determining step as stated above.
With 3b as starting material, we also studied the effect of the
concentration in chiral base on reaction rate. (1,3)PSR is
therefore conducted with 5, 10 and 20 mol% of MQE-DHQN at
80 8C in toluene (Fig. 4).
Fig. 4. Effect of the base concentration on reaction rate.

504
V. Michaut et al. / Journal of Fluorine Chemistry 128 (2007) 500–506
double triplet, m for multiplet and b for broad. HPLC analyses
were carried out on either a HP series 1100 using a HP 3395
integrator or a Waters 600 with a Waters 486 UV detector and a
Waters 746 integrator. Chiral columns used are Chiralcel OD-H
and Chiralpak AD-H. Gas chromatography (GC) was
performed on a Hewlett Packard HP 5890A series II using a
Hewlett Packard HP 3396 series II integrator. Mass spectro-
metry was performed on a Shimadzu GCMS-QP2010 in EI or
CI mode. Elementary Analyses were realised on a CE
instruments EA 1110.
4.2. General procedure for the synthesis of enamines,
typical example with enamine 3b: (Z)-ethyl 3-(4-
nitrobenzylamino)-4,4,4-trifluorobut-2-enoate (3b)
Fig. 5. Enantiodivergent discrimination.
A solution of 4-nitrobenzylamine hydrochloride in aqueous
Na2CO3 was extracted with dichloromethane. Organic layers
were combined, dried over MgSO4, filtrated and evaporated
under vacuum. 4-Nitrobenzylamine thus obtained (1.06 g,
7 mmol) was added at room temperature to a solution of glacial
acetic acid (0.42 g, 7 mmol) in chloroform (2 mL). After
stirring for 5 min at room temperature, ETFAA (1.16 g,
6.3 mmol) diluted in chloroform (1 mL), was added. The
resultant mixture was refluxed for 2 h followed by evaporation
of the solvent in vacuum. The residue was placed on a silica gel
column and eluted with a mixture of cyclohexane/ethyl acetate
(5/1) to afford the desired product (1.80 g, 90%) as pale yellow
solid.
3. Conclusion
We studied enantioselective (1,3) proton transfer of ETFAA
derivatives catalyzed by a chiral base. Special emphasis was
placed on the mechanistic aspect of the reaction by use of a
deuterated enamine and all results support a deprotonation step
as both rate and asymmetric determining step.
Both reaction conditions and influence of the nature of the
starting material were considered; the best enantiomeric excess
was obtained with (DHQ)2PHAL as chiral base and reaches
71% which is far better than previous results reported in the
literature up to now for organocatalytic enantioselective
(1,3)PSR. This study opens a new route to trifluoromethylated
chiral building blocks. We are currently exploring transforma-
tions of the obtained b-trifluoromethylated b-amino acids for
the construction of various fluorinated targets [13].
C13H13F3N2O4, MW = 318.25, pale yellow solid,
Mp = 68 8C; IR (cmÀ1): 3275 (NH); 3109, 3083 (CHAro);
2997, 2962, 2856 (CH3, CH2); 1676 (dbCIS); 1630 (CO); 1605
(dbAro); 1516, 1351 (NO2); RMN 19F (CDCl3): À66.9 (3F, s);
1
RMN H (CDCl3): 1.28 (3H, t, J = 7.2), 4.17 (2H, q, J = 7.2),
4. Experimental
4.58 (2H, d, J = 6.7), 5.24 (1H, s), 7.45 (2H, d, J = 8.7), 8.22
(2H, d, J = 8.7), 8.59 (1H, bb); RMN 13C (CDCl3): 14.3, 47.3,
60.2, 87.0 (q, J = 5.6), 120.3 (q, J = 277.0), 124.2, 127.7, 145.6,
147.6, 147.9 (q, J = 31.0), 169.9; HPLC (Chiracel OD-H,
Hexane/i-PrOH, 98/2, 1 mL minÀ1, 254 nm): ꢀ9.03 min; GC–
MS (EI): 271(100), 245(49), 289(61), 318(M+ ); Anal. Calcd.
for C13H13F3N2O4: C, 49.06; H, 4.12; N, 8.80; Found: C, 49.05;
H, 4.02; N, 8.74.
4.1. General
Toluene was distilled from sodium and benzophenone under
argon atmosphere. Acetonitrile and butan-1-ol were distilled on
calcium hydride. Thin layer chromatography was performed
using silica TLC plates (silica gel 60 F254, Merck). Products
were visualized under UV light (254 nm) and then revealed by
potassium permanganate aqueous solution. Flash chromato-
graphies were performed on silica gel column (silica gel Si60
0.040–0.063 mm, Merck).
4.3. (Z)-Ethyl 3-(benzylamino)-4,4,4-trifluorobut-2-enoate
(3a)
Infra Red spectra were recorded on a Perkin-Elmer 16PCFT-
IR and wave numbers are given in cmÀ1. NMR Spectra were
C13H14F3NO2, MW = 273.25, yellow oil; IR (cmÀ1): 3283
(NH); 3031 (CHAro); 2982 (CH3); 2936 (CH2); 1672 (dbCIS);
1632 (CO); RMN 19F (CDCl3): À67.0 (3F, s); RMN 1H
(CDCl3): 1.18 (3H, t, J = 7.1), 4.05 (2H, q, J = 7.1), 4.39 (2H, d,
J = 6.4), 5.08 (1H, s), 7.21 (5H, m), 8.37 (1H, bb); RMN 13C
(CDCl3): 14.3, 48.1, 59.8, 85.3 (q, J = 5.9), 120.4 (q,
J = 277.0), 127.3, 127.9, 128.9, 137.8, 148.2 (q, JC-F = 31.0),
169.9; HPLC (Chiracel OD-H, Hexane/i-PrOH, 99.9/0.1,
1 mL minÀ1, 254 nm): 14.4 min; GC (BPX5, 23 m, 40 8C
1
performed on a Bruker Avance 300. Chemical shifts of H
NMR (300 MHz) were expressed in parts per million downfield
from tetramethylsilane external standard (d = 0) in CDCl3.
Chemical shifts of 13C NMR (75.5 MHz) were expressed in
parts per million downfield from CDCl3 as internal standard
(d = 77.16). Chemical shifts of 19F NMR (282 MHz) were
expressed in parts per million downfield from CFCl3 as internal
standard (d = 0). Coupling constants are reported in hertz (Hz).
Abbreviations used for signals observed are: s for singlet, d for
doublet, t for triplet, q for quartet, dd for double doublet, dt for
(5 min), 8 8C/min, 280 8C): 16.6 min; GC–MS (EI): 91(100),
ꢀ
200(52), 273(M+ ); Anal. Calcd. for C13H14F3NO2: C, 57.14;
H, 5.16; N, 5.13; Found: C, 57.06; H, 4.98; N, 5.39.

V. Michaut et al. / Journal of Fluorine Chemistry 128 (2007) 500–506
505
4.4. (Z)-Ethyl 4,4,4-trifluoro-3-(perfluorobenzylamino)but-
2-enoate (3c)
4.8. General procedure for the (1,3) enantioselective
proton shift reaction using cinchona alkaloids as chiral
base: typical example with enamine 3b
C13H9F8NO2, MW = 363.20, yellow oil; IR (cmÀ1): 3288,
3232 (NH); 2985 (CH3); 2941, 2901 (CH2); 1675 (dbCIS); 1636
(CO); RMN 19F (CDCl3): À66.9 (3F, s), À143.6 (2F, m),
In a flask were introduced, (DHQ)2PHAL (20 mg,
0.025 mmol), enamine 3b (79.6 mg, 0.25 mmol) and distilled
toluene (1 mL). The resultant mixture was stirred and heated.
Reaction was followed by 19F NMR. After 24 h at 80 8C,
solvent was evaporated in vacuum, the residue was put on a
short silica gel column and eluted with cyclohexane/ethyl
acetate (10/1) to afford desired product, imine 2b (18 mg,
90%) as yellow oil. The chiral base could be recovered
by eluting it with a mixture of dichloromethane/methanol
(97/3).
1
À153.9 (1F, t, J = 20.6), À161.5 (2F, m); RMN H (CDCl3):
1.26 (3H, t, J = 7.2), 4.14 (2H, q, J = 7.2), 4.54 (2H, d, J = 6.4),
5.19 (1H, s), 8.45 (1H, bs); RMN 13C (CDCl3): 14.3, 35.9, 60.3,
87.5, 110.0, 120.2 (q, J = 276.5), 136.1, 139.6, 143.8, 147.1 (q,
J = 31.6), 169.8; HPLC (Chiracel OD-H, Hexane/i-PrOH, 99.5/
0.5, 1 mL minÀ1, 254 nm): 7.01 min.
4.5. (Z)-Ethyl 3-[(1-2H2) benzylamino]-4,4,4-trifluorobut-
2-enoate (d-3a)
4.9. (E)-Ethyl 3-(4-nitrobenzylideneamino)-4,4,4-
trifluorobutanoate (2b)
C13H12D2F3NO2, MW = 275.26, colourless oil; IR (cmÀ1):
3277, 3232 (NH); 3064, 3030 (CHAro); 2980 (CH3); 2929
(CH2); 1672 (dbCIS); 1630 (CO); RMN 19F (CDCl3): À67.1
C13H13F3N2O4, MW = 318.25, yellow oil; IR (cmÀ1): 3083
(CHAro); 2986 (CH3); 2937 (CH2); 1637 (CO); 1650 (CN);
1526, 1348 (NO2Aro); RMN 19F (CDCl3): À75.4 (3F, d,
J = 6.8); RMN 1H (CDCl3,): 1.20 (3H, t, J = 7.2), 2.93 (2H, d,
J = 7.2), 4.11 (2H, m), 4.30 (1H, m), 7.96 (2H, d, J = 8.7), 8.28
(2H, d, J = 8.7), 8.52 (1H, s); RMN 13C (CDCl3): 14.2, 34.2,
61.4, 67.9 (q, J = 28.3), 124.0, 124.8 (q, J = 279.9), 129.7,
140.4, 149.8, 165.3, 169.8; HPLC (Chiracel OD-H, Hexane/i-
PrOH, 98/2, 1 mL minÀ1, 254 nm): (S): 10.3 min, (R):
12.7 min; Anal. Calcd. for C13H13F3N2O4: C, 49.06; H,
4.12; N, 8.80; Found: C, 49.13; H, 4.17; N, 8.80.
1
(3F, s); RMN H (CDCl3): 1.16 (3H, t, J = 7.2), 4.04 (2H, q,
J = 7.2), 5.08 (1H, s), 7.17–7.29 (5H), 8.35 (1H, bs); RMN 13
(CDCl3): 14.4, 47.6 (m), 59.8, 85.3, 120.4 (q, J = 277.0),
C
129.0, 128.0, 127.4, 137.7, 148.3 (q, J = 31.1), 170.0; GC–MS
ꢀ
(EI): 93(100); 227(26); 246(41); 275(M+ ); Anal. Calcd. for
C13H12D2F3NO2: C, 56.72; H, 5.86; N, 5.09; Found: C, 56.71;
H, 5.91; N, 5.19.
4.6. (Z)-Ethyl 3-(benzhydrylamino)-4,4,4-trifluorobut-2-
enoate (3d)
C19H18F3NO2, MW = 349.35, white solid, Mp = 76–78 8C;
IR (cmÀ1): 3198 (NH); 2985 (CHAro); 1667 (CO); 1625 (cdb);
1493–1446 (CH2-CH3); RMN 19F (CDCl3): À66.3 (3F, s);
4.10. (E)-Ethyl 3-(benzylideneamino)-4,4,4-
trifluorobutanoate (2a)
1
RMN H (CDCl3): 1.19 (3H, t, J = 7.1), 4.09 (2H, q, J = 7.1),
C13H14F3NO2, MW = 273.25, colourless oil; IR (cmÀ1):
3064 (CHAro); 2985 (CH3); 2890 (CH2); 1739 (CO); 1647
5.23 (1H, s), 5.84 (1H, d, J = 10.9), 7.35–7.37 (10H, m), 9.0
(1H, bd, J = 10.9); RMN 13C (CDCl3): 14.4, 60.0, 62.0, 86.6 (q,
J = 5.6), 120.4 (q, J = 277.7), 127.1, 127.7, 128.9, 142.0, 147.6
(q, JC-F = 31.8), 169.9; HPLC (Chiracel OD-H, Hexane/i-
PrOH, 99.9/0.1, 1 mL minÀ1, 254 nm): 6.6 min; Anal. Calcd.
for C19H18F3NO2: C, 65.32; H, 5.19; N, 4.01; Found: C, 65.54;
H, 5.13; N, 4.04.
(CN); RMN 19F (CDCl3): À75.5 (3F, d, J = 6.8); RMN H
1
(CDCl3): 1.09 (3H, t, J = 7.1), 2.81 (2H, d, J = 6.4), 3.99 (2H,
2q, J = 7.1), 4.15 (1H, m), 7.34 (3H, m), 7.68 (2H, m), 8.32 (1H,
s); RMN 13C (CDCl3): 14.1, 34.6, 61.1, 68.1 (q, J = 29.0), 124.6
(q, J = 280.6), 131.8, 130.6, 128.7, 135.2, 167.4, 169.8; HPLC
(Chiracel OD-H, Hexane/i-PrOH, 99.9/0.1, 1 mL minÀ1
,
254 nm): (S): 12.6 min, (R): 23.3 min; Anal. Calcd. for
C13H14F3NO2: C, 57.14; H, 5.16; N, 5.13; Found: C, 57.06;
H, 5.23; N, 5.27.
4.7. (Z)-Ethyl 3-(9H-fluoren-9-ylamino)-4,4,4-trifluorobut-
2-enoate (3e)
C19H16F3NO2, MW = 347.33, white solid, Mp = 100–
102 8C; IR (cmÀ1): 3226 (NH); 3064 (CHAro); 2981 (CH2-
CH3); 1673 (CO); 1628 (cdb); RMN 19F (CDCl3): À65.2 (3F,
s); RMN 1H (CDCl3): 1.23 (3H, t, J = 7.1), 4.06 (2H, q, J = 7.1),
5.27 (1H, s), 5.64 (1H, d, J = 11.3), 7.32 (2H, t, J = 7.5), 7.41
(2H, t, J = 7.5), 7.53 (2H, t, J = 7.5), 7.7 (2H, t, J = 7.5), 8.26
(1H, bd); RMN 13C (CDCl3): 14.3, 59.8, 59.9, 86.0 (q, J = 5.6),
120.8 (q, J = 277.0), 120.2, 125.1, 128.0, 129.1, 140.3, 144.0,
148.4 (q, J = 30.4), 169.6; HPLC (Chiracel OD-H, Hexane/i-
PrOH, 99.5/0.5, 1 mL minÀ1, 254 nm): 15.6 min; Anal. Calcd.
for C19H16F3NO2: C, 65.70; H, 4.64; N, 4.03; Found: C, 65.54;
H, 4.61; N, 3.96.
4.11. (E)-Ethyl 3-(3-2H)-[(1-2H) benzylideneamino]-4,4,4-
trifluorobutanoate (d-2a)
C13H12D2F3NO2, MW = 275.26, colourless oil; IR (cmÀ1):
3064 (CHAro); 2980 (CH3); 2907 (CH2); 1737 (CO); 1630
(CN); RMN 19F (CDCl3): À75.7 (3F, s); RMN 1H (CDCl3): 1.2
(3H, t, J = 7.1), 2.90 (2H, s), 4.10 (2H, q, J = 7.1), 7.39–7.49
(3H, m), 7.77–7.80 (2H, m); RMN 13C (CDCl3): 14.2, 34.6,
61.2, 125.1 (q, J = 279.9), 128.5, 128.8, 131.9, 135.1, 167.2 (t,
J = 24.8), 170.0; GC–MS (CI, CH4): 276 (M + 1); 304
(M + 2CH4); GC–MS (EI): 105 (100); 202 (26); 246 (36);
ꢀ
275 (M+ ).

506
V. Michaut et al. / Journal of Fluorine Chemistry 128 (2007) 500–506
4.12. (E)-Ethyl 4,4,4-trifluoro-3-
(perfluorobenzylideneamino)butanoate (2c)
Acknowledgement
Rhodia Recherches and CNRS are gratefully acknowledged
for a research grant to VM.
C13H9F8NO2, MW = 363.20, yellow oil; IR (cmÀ1): 2985
(CH3); 2935, 2907 (CH2); 1637 (CO), 1653 (CN); RMN 19F
(CDCl3, 282 MHz, d): À75.5 (3F, d, J = 6.8), À141.6 (2F, m),
À149.3 (1F, J = 21.8, J = 3.4), À161.8 (2F, m); RMN 1H
(CDCl3, 300 MHz, d): 1.26 (3H, t, J = 7.2), 4.14 (2H, q,
J = 7.2), 4.15 (2H, m), 4.25 (1H, m), 5.19 (1H, s), 8.55 (1H, s);
RMN 13C (CDCl3, 75.5 MHz, d): 14.2, 34.2, 61.4, 69.1 (q,
J = 29.0), 108.8, 109.1, 124.5, 124.8 (q, J = 279.9), 136.2,
139.5, 144.7, 156.3, 157.8, 169.6; HPLC (Chiracel OD-H,
References
´ ´ ´
[1] (a) J.P. Begue, D. Bonnet-Delpon, Chimie bioorganique et medicinale du
fluor EDP Sciences/CNRS Eds. (2005).
(b) B.E. Smart, J. Fluorine Chem. 109 (2001) 3–11;
(c) D. O’Hagan, H.S. Rzepa, Chem. Commun. (1997) 645–652;
(d) M. Schlosser, Angew. Chem. Int. Ed. 110 (1998) 1496–1513.
´
´
[2] (a) J.P. Begue, D. Bonnet-Delpon, J. Fluorine Chem. 127 (2006) 992–1012;
(b) C. Isanbor, D. O’Hagan, J. Fluorine Chem. 127 (2006) 303–319;
(c) K. Kirk, J. Fluorine Chem. 127 (2006) 1013–1029.
[3] (a) M. Liu, M.P. Sibi, Tetrahedron 58 (2002) 7991–8035;
(b) R.P. Cheng, S.H. Gellman, W.F. DeGrado, Chem. Rev. 101 (2001)
3219–3232;
Hexane/i-PrOH, 99.5/0.5, 1 mL minÀ1
,
254 nm): 5.4,
14.2 min.
4.13. Ethyl 3-(9H-fluoren-9-ylideneamino)-4,4,4-
trifluorobutanoate (2e)
(c) D. Seebach, T. Kimmerlin, R. Sebesta, M.A. Campo, A.K. Beck,
Tetrahedron 60 (2004) 7455–7506.
[4] (a) F. Huguenot, T. Brigaud, J. Org. Chem. 71 (2006) 2159–2162;
(b) X.L. Qiu, W.D. Meng, F.L. Qing, Tetrahedron 60 (2004) 6711–6745,
and references cited therein;
C19H16F3NO2, MW = 347.33, white solid, Mp = 88–90 8C;
IR (cmÀ1): 3053 (CHAro); 2985 (CH2-CH3); 1734 (CO); RMN
1
19F (CDCl3): À75.7 (3F, d, J = 6.8); RMN H (CDCl3): 1.12
(c) N. Lebouvier, C. Laroche, F. Huguenot, T. Brigaud, Tetrahedron 43
(2002) 2827–2830.
(3H, t, J = 7.1), 2.76 (1H, dd, J = 16.2, J = 6.03), 3.02 (1H, dd,
J = 16.2, J = 7.2), 4.06 (2H, q, J = 7.1), 5.49 (1H, m), 7.22 (2H,
m), 7.36 (2H, m), 7.46, 7.56 (2H, d, J = 7.5), 7.74, 7.98 (2H, d,
J = 7.5); RMN 13C (CDCl3): 14.1, 35.8, 59.4 (q, J = 28.9), 61.4,
119.5, 120.7, 123.6, 127.4, 128.6, 131.9, 132.4, 131.2 (q,
J = 296.1), 138.1, 141.3, 144.5, 166.7; HPLC (Chiracel OD-H,
Hexane/i-PrOH, 99.5/0.5, 1 mL minÀ1, 254 nm): (R): 12.5, (S):
18.1 min; Anal. Calcd. for C19H16F3NO2: C, 65.70; H, 4.64; N,
4.03; Found: C, 65.46; H, 4.53; N, 3.97.
[5] (a) T. Ono, P. Kukhar, V.A. Soloshonok, J. Org. Chem. 61 (1996) 6563–
6569;
(b) V.A. Soloshonok, A.G. Kirilenko, V.P. Kukhar, Tetrahedron Lett. 34
(1993) 3621–3624.
[6] V.A. Soloshonok, D.O. Berbasov, J. Fluorine Chem. 125 (2004) 1757–1763.
[7] (a) V.A. Soloshonok, H. Ohkura, M. Yasumoto, J. Fluorine Chem. 127
(2006) 889–893;
(b) V.A. Soloshonok, H. Ohkura, M. Yasumoto, J. Fluorine Chem. 127
(2006) 924–929;
(c) V.A. Soloshonok, H. Ohkura, M. Yasumoto, J. Fluorine Chem. 127
(2006) 930–935;
(d) V.A. Soloshonok, I.V. Soloshonok, V.P. Kukhar, V.K. Svedas, J. Org.
Chem. 63 (1998) 1878–1884;
4.14. General procedure for hydrolysis of imines: 3-amino-
4,4,4-trifluorobutanoic acid hydrochloride
(e) V.A. Soloshonok, T. Ono, I.V. Soloshonok, J. Org. Chem. 62 (1997)
7538–7539;
(f) V.A. Soloshonok, T. Ono, J. Org. Chem. 62 (1997) 3030–3031.
[8] (a) V.A. Soloshonok, V.P. Kukhar, Tetrahedron 52 (1996) 6953–6964;
(b) V.A. Soloshonok, A.G. Kirilenko, S.V. Galushko, V.P. Kukhar, Tetra-
hedron Lett. 35 (1994) 5063–5064.
Mixture of imine diluted in a solution of HCl 6 M was
refluxed during 3 h. Organic layer was extracted with
dichloromethane. Aqueous layer was recovered and water
was evaporated in vacuum. Hydrochloride of desired amino
acid was obtained.
[9] (a) B. Mohar, J. Baudoux, J.C. Plaquevent, D. Cahard, Angew. Chem. Int.
Ed. 40 (2001) 4214–4216;
(b) D. Cahard, C. Audouard, J.C. Plaquevent, N. Roques, Org. Lett. 2
(2000) 3699–3701.
C4H7ClF3NO2, MW = 193.55, white solid, Mp = 168–
170 8C; IR (cmÀ1): 3400 (OH); 2907 (NH); 1731 (CO);
1418 (C-O); RMN 19F (CD3OD): À74.3 (3F, d, J = 6.8); RMN
1H (CD3OD, 300 MHz, d): 2.88 (1H, dd, J = 8.7, J = 17.7), 3.05
(1H, dd, J = 4.1, J = 17.7), 4.52 (1H, m), 5.25 (4H, bb); RMN
13C (CD3OD, 75.5 MHz, d): 31.9, 50.7 (q, J = 32.5), 125.1 (q,
J = 279.9), 171.2; Anal. Calcd. for C4H7ClF3NO2: C, 24.82; H,
3.65; N, 7.24; Found: C, 24.84; H, 3.73; N, 7.18.
[10] H. Ohkura, D.O. Berbasov, V.A. Soloshonok, Tetrahedron 59 (2003)
1647–1656.
[11] V.A. Soloshonok, A.G. Kirilenko, N.A. Fokina, I.P. Shishkina, S.V.
Galushko, V.P. Kukhar, V.K. Svedas, V. Kozlova, Tetrahedron Asymm.
5 (1994) 1119–1126.
[12] V.A. Soloshonok, T. Ono, Tetrahedron 52 (1996) 14701–14712.
[13] V. Michaut, Dissertation, Rouen University (France), 2006, and results to
be published.