1,1,1-Trifluoropropan-2-ammonium triflate enantiomers: stereoselective synthesis and direct use in reaction with epoxides

Article abstract of DOI:10.1016/j.tetasy.2017.03.003

A three-step synthesis of enantiomerically enriched 1,1,1-trifluoro-2-propanamine based on the use of a chiral sulfinamide auxiliary is described. The reduction of the geometrically pure Z-sulfinimine (NOE, HOE) with NaBH₄ or L-Selectride leads to the corresponding (R)- or (S)-configured amine derivatives (X-ray crystallographic analysis) with 92–96% de. The typical models to explain the stereoselection for these reducing agents fail to rationalize the obtained stereoselectivities, and an in situ imine isomerization is proposed to occur. The direct use of the hydrochloric acid salt (with excess Et₃N) of this poorly nucleophilic amine for epoxide opening reactions is not possible due to the higher nucleophilicity of chloride. Hence, a novel triflate salt is introduced, synthesized through ready sulfinamide hydrolysis with trimethylsilyl triflate, which can be used directly, without the need of isolating the pure amine beforehand.

Full text of DOI:10.1016/j.tetasy.2017.03.003

Tetrahedron: Asymmetry xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
1,1,1-Trifluoropropan-2-ammonium triflate enantiomers:
stereoselective synthesis and direct use in reaction with epoxides
⇑
Gemma Packer, Julien Malassis, Neil Wells, Mark Light, Bruno Linclau
Chemistry, University of Southampton, Highfield, Southampton SO17 1BJ, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
A three-step synthesis of enantiomerically enriched 1,1,1-trifluoro-2-propanamine based on the use of a
Received 9 February 2017
Revised 1 March 2017
Accepted 6 March 2017
Available online xxxx
chiral sulfinamide auxiliary is described. The reduction of the geometrically pure Z-sulfinimine (NOE,
HOE) with NaBH4 or L-Selectride leads to the corresponding (R)- or (S)-configured amine derivatives
(X-ray crystallographic analysis) with 92–96% de. The typical models to explain the stereoselection for
these reducing agents fail to rationalize the obtained stereoselectivities, and an in situ imine isomeriza-
tion is proposed to occur. The direct use of the hydrochloric acid salt (with excess Et3N) of this poorly
nucleophilic amine for epoxide opening reactions is not possible due to the higher nucleophilicity of chlo-
ride. Hence, a novel triflate salt is introduced, synthesized through ready sulfinamide hydrolysis with
trimethylsilyl triflate, which can be used directly, without the need of isolating the pure amine
beforehand.
Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction
enantiopure 1 thanks to possible diastereomeric separations of
intermediates.9,10 However, these routes are quite lengthy.
1,1,1-Trifluoropropan-2-amine (trifluoroisopropylamine) 1 is a
valuable fluorinated building block. Both enantiomers and the
racemic mixture are commercially available, if expensive. The
enantiomers can be synthesized on an industrial scale by SF4-
Based on the above, we were interested in investigating the
enantioselective formation of 1 using the Ellman sulfinamide aux-
iliary.11 Key precedence includes the formation of the correspond-
ing imine 5a (Scheme 1) as an intermediate for the synthesis of
trifluoromethylated amino acids,12,13 as well as the formation
and stereoselective reduction of trifluoromethylated imines
5b–d.14–16 There are other examples of sulfinimine formation from
trifluoromethyl ketones.17–20
The N-sulfinylimine 5a was reported to be unstable/prone to
hydrolysis, but could be efficiently trapped with cyanide (Scheme 1
(a)).12 Grellepois showed that the unstable 5a could be easily iso-
lated as the hemiaminal 6, which when submitted to a Reformatski
reaction in THF, slowly released 5a before reacting with the zinc/
copper enolate (Scheme 1 (b)).13 In contrast, conjugated imines
5b proved to be stable and isolable (Scheme 1 (c)). Reduction of
sulfinimines 5b-d proceeded with excellent diastereoselectivities
(Scheme 1, (c-e)), with the relative configuration of the formed
amine center depending on the nature of the reducing agent used.
Although the geometry of 5a,10 5c,13 and 5d16 was not established,
the geometry of one conjugated unsaturated imine 5b (R = p-tolyl)
was proven via X-ray crystallographic analysis.14
mediated fluorination of D- and L
-alanine.1 The toxicity issues asso-
ciated with gaseous SF4 requires the use of specialized facilities,
resulting in the high commercial value of this building block, and
has spurred alternative synthetic methodologies compatible with
standard laboratories.
Racemic 1 can be synthesized from the corresponding oxime via
high pressure reduction,2,3 or by base-catalyzed isomerization of
the benzyl imine of 1,1,1-trifluoroacetone.4 Resolution of racemic
1 is possible via crystallization with enantiopure tartaric acid5
(up to 99% ee).6 There are several enantioselective syntheses that
employ chirality transfer. A seminal procedure introduced by
Soloshonok involving the use of enantiopure 1-phenylethylamine
in conjunction with the aforementioned base-catalyzed imine iso-
merization,6,7 which typically leads to 92–93% ee, is the most used.
An organocatalytic isomerization procedure starting from achiral
benzyl imine is also reported (90% ee).8 Bravo et al. reported
two methods based on a chiral sulfoxide, either starting from
trifluoroacetic acid or from trifluoroacetaldehyde, which led to
Herein we report our results in applying this methodology
towards the synthesis of enantioenriched 1,1,1-trifluoro-2-propa-
namine 1, including the establishment of the geometry of the
N-sulfinylimine 5a, as well as of the resulting reduction products
⇑
Corresponding author.
E-mail address: bruno.linclau@soton.ac.uk (B. Linclau).
http://dx.doi.org/10.1016/j.tetasy.2017.03.003
0957-4166/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Packer, G.; et al. Tetrahedron: Asymmetry (2017), http://dx.doi.org/10.1016/j.tetasy.2017.03.003

2
G. Packer et al. / Tetrahedron: Asymmetry xxx (2017) xxx–xxx
t-Bu
using both types of reducing agents. A model for the observed
stereoselection is also discussed. In addition, we report on the
use of salts of the volatile 1 in epoxide opening reactions.
t-Bu
t-Bu
(a)
(S)-4
H2N
S
O
S
S
O
TMSCN
N
O
HN
O
F3C
R
Ti(Oi-Pr)4
R
CN
F3C
F3C
R
hexane
2
2. Results and discussion
R = Ar, Alk
R = Me, 47% (19F NMR)
7
5a
Liu’s and Lu’s procedures for N-sulfinylimine formation were
initially evaluated with trifluoroacetone (Table 1, entries 1,2).
The reduction products were obtained with excellent stereoselec-
tivities, although in low yield. The two diastereomers could be
easily distinguished via 19F NMR (two doublet peaks at d À78.15
t-Bu
t-Bu
(b)
(c)
t-Bu
4
(R)-
S
S
O
S
H2N
O
THF
O
Refor-
matski
HN
F3C
N
O
F3C
Ti(OEt)4
hexane
F3C
2a
OEt
5a
and
d
À78.62 ppm), and the complete consumption of N-
70%
6
t-Bu
sulfinylimine 5a was evidenced by the absence of its singlet at
À75.48 ppm. Unfortunately, the mixture of diastereoisomers
proved inseparable by chromatography. Due to the volatility of
diethyl ether and the need to conduct the reaction in a sealed tube,
it was decided to continue our investigations using non-volatile
solvent, despite the better yield observed with Liu’s procedure. It
was found that increasing the excess of NaBH4 and addition at
lower temperature (entry 3) led to an improved dr of 96:4,
although a wide range of yields was obtained (11 to 77% for 5 reac-
tions, 55% on scale, entry 4), indicating issues with reproducibility.
The reaction was also investigated with an excess of Ti(OiPr)4 in a
attempt to prevent hydrolysis of the unstable imine, although this
made the work up considerably more troublesome (entry 5).
Subsequent reactions involving the use of DCM/MeOH as solvents
for the imine formation allowed for more consistent yields to be
obtained (entries 6,7,9 and 10). Reducing the excess of auxiliary
from 2 to 1.5 equiv did not impact the outcome of the reaction
(entries 7,9). However, reversing the excess of reagents led to a
lower yield (entry 8). Improved dr’s of 97:3 and 98:2 were also
found by suppressing the use of MeOH during the reduction
(entries 9, 10). This reaction could be carried out on 5 mL scale of
trifluoroacetone (entry 10), giving the product in 52% yield.
t-Bu
t-Bu
X-ray
(R)-4
H2N
S
S
S
N
O
O
HN
F3C
O
O
(E)
F3C
F3C
R
Ti(OEt)4,
THF/HMPA,
-78 °C
*
5b
8
3
R = Ar, Alk
92-98% de (R)
DIBAL-H, THF, -78 °C
81-99%
L
-Selectride,
90-98% de (S)
THF/HMPA, -78 °C
60-99%
t-Bu
(d)
t-Bu
t-Bu
4
(R)-
S
O
S
S
H2N
O
N
O
HN
F3C
O
F3C
R
9
Ti(Oi-Pr)4
Et2O
F3C
R
R
2
5c
90-98% de (R)
R = Ar, Alk
(not Me)
NaBH4, Et2O, -78 °C
66-85
L
-Selectride,
92-98% de (S)
Et2O, -78 °C
60-84%
t-Bu
(e)
(R)-4
S
t-Bu
t-Bu
H2N
Zr(OtBu)4
O
LiBH4,
THF
S
S
HN
O
2'
N
O
R = 9-anthryl
rt (71%)
CPME, 80 °C,
85%
F3C
9' 97% ee (R)
ANT
F3C
ANT
5d
The reduction with L-Selectride was briefly investigated, in
order to see whether a change in facial selectivity is also observed
(cf Scheme 1). This proved indeed to be the case (entry 11) with
high diastereoselectivity being achieved, albeit in low yield.
Scheme 1. Key precedence involving trifluoromethylated t-butanesulfinyl
ketimines.
Table 1
Optimization experiments for imine formation and reduction.
t-Bu
t-Bu
t-Bu
t-Bu
4
(R)-
S
O
S
H2N
O
S
reducing agent
S
N
O
HN
O
+
HN
O
CF3
2a
solvent
16 h
Ti(Oi-Pr)4
RT, 6-94 h
sealed tube
CF3
CF3
CF3
10
(RS,R)-
10
(RS,S)-
5a
Entry
Ti(OiPr)4 (equiv)
(R)-4 (equiv)
Solvent
Reduction Temp (°C)
Reagent (equiv)
Solvent
Yield (%)
dr (R:S)
1a
2.5
1.5
1.5
1.5
4
1.5
1.5
1.5
1.5
1.5
4
1.25
1
2
2
2
Et2O
À78 °C to rt
0 °C to rt
NaBH4 (3.0)
NaBH4 (1.0)
NaBH4 (3.0)
NaBH4 (3.0)
NaBH4 (3.0)
NaBH4 (3.0)
NaBH4 (3.0)
NaBH4 (3.0)
NaBH4 (3.0)
NaBH4 (3.0)
Et2O
36
15
11–77
55
48
54
55
35
56
52
98:2
95:5
96:4
96:4
96:4
95:5
96:4
96:4
97:3
98:2
4:96
2a
Hexane
Hexane
Hexane
Hexane
Hexane/DCM
Hexane/DCM
Hexane/DCM
Hexane/DCM
Hexane/DCM
THF
THF/MeOH
THF/MeOH
THF/MeOH
THF/MeOH
THF/MeOH
THF/MeOH
THF/MeOH
THF
3a
À30 °C to rt
À30 °C to rt
À30 °C to rt
À30 °C to rt
À30 °C to rt
À30 °C to rt
À30 °C to rt
À30 °C to rt
À78 °C
4b
5a
6a
2
7a
1.5
1c
2
1.5
1.05
8a
9a
10b
11c
THF
THF
22
L
-Selectride (1.05)
a
b
c
5 mmol scale of 2a.
55.8 mmol scale of 2a.
1.5 equiv of 2a.
Please cite this article in press as: Packer, G.; et al. Tetrahedron: Asymmetry (2017), http://dx.doi.org/10.1016/j.tetasy.2017.03.003

G. Packer et al. / Tetrahedron: Asymmetry xxx (2017) xxx–xxx
3
quenching with H2O, purification was achieved via filtration under
reduced pressure through a plug of silica. The sample was then
immediately submitted for NOE/HOE studies (Fig. 2; Supporting
Information). Irradiation of the methyl or t-butyl peaks did not lead
to any coupling (NOE), while irradiation of the trifluoromethyl
peak (HOE) led to coupling with both methyl and trifluoromethyl
groups. This is consistent with Z-geometry, opposite with what
was found for the
a,b-unsaturated N-sulfinylimines 5b (cf
Scheme 1).14 This outcome is surprising, given the larger size of
CF3 compared to CH3.
With the imine geometry and relative stereochemistry of the
amines known, rationalization of the stereoselection was under-
taken. Reactions of N-sulfinylimines with Selectride have been
rationalized via an open transition state similar to TS-1 (Fig. 3),21
Fig. 1. Crystal structures of (RS,R)-10 and (RS,S)-10.
including for the reduction of the E-configured
a,b-unsaturated
sulfinimines. However, with the Z-configured 5a as substrate, TS-
1 would lead to Si-face attack, which is not what is observed. Pos-
sible destabilization of TS-1 by repulsion between the CF3 group
and the sulfoxide oxygen could induce in situ isomerization, lead-
ing to TS-2, which now should undergo Re-face attack to give the
observed major diastereomer (RS,S)-10. Alternatively, a transition
state involving a sulfinimine in the s-trans conformation could be
considered as well (not shown).22,23 In contrast, N-sulfinylimine
reduction with NaBH4 is proposed to proceed via a chelated transi-
tion state.17 With the Z-imine 5a, this would give TS-3, featuring
Re-face attack, which was not observed. In this case, TS-3 could
be destabilized by repulsion between the sulfur lone pair of elec-
trons and the CF3-moiety.15 Hence, in situ imine isomerization is
again invoked, allowing the chelated transition state TS-4, which
predicts the observed (RS,R)-diastereomer.
N-Sulfinylimines are known to undergo rapid E/Z isomerization
(but with a high enough isomerization barrier as to be able to
observe the individual isomers by NMR at rt),24,25 and was reported
for a number of other N-sulfinylimine reactions.26–32 This was typ-
ically explained by invoking chelation, or through prior coordina-
tion with a Lewis acid before reaction. It was also implied by Liu
to account for their observed diastereoselectivities.15 In the case
Both N-trifluoroisopropylsulfinamide diastereomers 10 were
crystalline solids, but all attempts to achieve a preparatively useful
separation through recrystallization were met with failure.
Attempts with EtOAc/pentane as solvent did generate single crys-
tals of both diastereomers suitable for X-ray diffraction studies
(Fig. 1), which allowed unambiguous determination of the relative
configuration of the reduction products.
In order to determine the facial selectivity of the N-sulfinylim-
ine reduction, the determination of the configuration of the N-
sulfinylimine 5a was attempted. Following Lu’s procedure,12 its
formation was investigated in hexane. Analysis of the crude reac-
tion mixture via 19F NMR showed the formation of a single peak
at d À75.48 ppm (CDCl3), which indicated the selective formation
of a single imine geometric isomer. However, NOE analysis was
hampered by the presence of residual hexane and unreacted (R)-
tert-butane sulfinamide, which obscured key resonances in the
spectra. Imine formation was then investigated in CH2Cl2, with
fewer equivalents of auxiliary. An increased amount of Ti(OiPr)4
(4 equiv) was also added in an attempt to contain hydrolysis. Anal-
ysis via 19F NMR after 18 h (CDCl3) showed the formation of the
same geometric imine isomer previously observed and, after
no NOE
O
t-Bu
to CH3
no NOE
4M HCl (dioxane)
NH2•HX
to t-Bu
S
MeOH, rt, or
S
t-Bu
O
N
HN
CF3
TMSOTf (1.3 equiv),
MeOH,0 ºC to rt
HOESY
H3C
CF3
CF3
(R)-1.HCl (89%)
(RS,R)-10
1.HOTf
(90%)
HOESY
Fig. 2. Imine geometry determination.
(R)-
Scheme 2. Auxiliary removal.
TS-1
TS-2
t-Bu
t-Bu
L
- or LS-
t-Bu
t-Bu
(R) S
N
imine
Selectride
(R) S
N
N
O
isomerisation
HN
O
OBSERVED
O
O
H3C
CF3
(S)
H3C
CF3
open
M-H
CF3
H3C
M-H
transition
state
(Z)-5a
CH3
F3C
10
(RS,S)-
Z-configuration
(Si-face attack)
E-configuration
(Re-face attack)
NOT OBSERVED
NOT OBSERVED
TS-3
TS-4
t-Bu
t-Bu
(R)S
(R)S
imine
isomerisation
H
H
NaBH4
O
O
OBSERVED
M
M
N
O
HN
O
N
CH3
CF3
N
t-Bu
t-Bu
S
S
(R)
H3C
CF3
H3C
CF3
CF3
chelated
transition
state
CH3
(Z)-5a
(RS,R)-10
E-configuration
(Si-face attack)
Z-configuration
(Re-face attack)
Fig. 3. Explanation of the stereoinduction.
Please cite this article in press as: Packer, G.; et al. Tetrahedron: Asymmetry (2017), http://dx.doi.org/10.1016/j.tetasy.2017.03.003

4
G. Packer et al. / Tetrahedron: Asymmetry xxx (2017) xxx–xxx
Et3N, LiClO4
H
MeCN/MeOH
N
CF3
OH
NH2•HX
CF3
4:1
CF3
O
+
+
+
X
80 °C
5-10d
N
H
OH
OH
(X = Cl) 22%
12
11
13a
(R)-1.HX
13b (X = OTf)
(1.2 equiv)
X = Cl
X = OTf
1%
29% (~1:1)
not observed
Scheme 3. Epoxide opening starting from amine salts.
of 5a, it appears that isomerization is induced simply by unfavour-
able interactions in the transition state.
Auxiliary removal was achieved with the usual conditions to
give the hydrochloric acid salt 1ÁHCl (Scheme 2), and purification
via trituration with Et2O allowed its isolation as a white solid in
89% yield (see Supporting Information for X-ray crystallographic
between these reducing agents was expected based on existing
models for stereoselection, in situ sulfinimine isomerization to
the E-isomer has to be invoked in order to explain stereoselection.
It was shown that TMSOTf in MeOH was an excellent reagent com-
bination for auxiliary removal, which led to the synthesis of the
novel trifluoroisopropylammonium triflate salt, which could be
used directly as a convenient source of trifluoroisopropylamine
in a nucleophilic substitution reaction (excess Et3N) without a
competing reaction of the triflate anion. Given the importance of
fluorination in drug development38–41 and agrochemical indus-
try,42,43 this work will be of general interest in the biosciences.
analysis). The specific rotation {[
92% ee} is comparable to the value of [
a
]
23 = À1.15 (c 1.0, MeOH) for
D
a
]
D
25 = À2.9 (c 1.0, MeOH)
25 °C for 98% ee reported by Soloshonok.7 In order to access the tri-
flate salt (see below), we found that cleavage with TMSOTf in
MeOH, which to the best of our knowledge has not been used
yet for this transformation,33 was very facile, leading to the novel
triflate salt 1ÁOTf for which suitable crystals for X-ray crystallo-
graphic analysis were obtained (Supporting Information). Although
this combination must result in TfOH release, it was superior to
TfOH itself which gave a lower yield (77%, not shown), with a sig-
nificant exotherm.
4. Experimental
4.1.
(RS,Z)-N-(1,1,1-Trifluoro-2-propylidene)-t-butanesulfina-
mide 5a
The crystalline hydrochloric acid salt is a convenient source of
the volatile trifluoroisopropylamine. However, the trifluoromethyl
group strongly reduces the nucleophilicity of the amino group, to
the extent that 1 is much less nucleophilic than chloride. This
was confirmed by an experiment (Scheme 3), in which reaction
of epoxide 11 with 1ÁHCl in the presence of excess Et3N essentially
only gave the chlorohydrin 13a. Hence, when 1.HCl is used as start-
ing material for nucleophilic substitution reactions, prior neutral-
ization and isolation of 1 as free base with complete removal of
chloride species is required.34 However, the volatility of 1 intro-
duces practical difficulties, and hence it was investigated as to
whether the corresponding triflate 1ÁOTf, in the presence of excess
Et3N, would be a suitable trifluoroisopropylamine source to use
directly in epoxide opening reactions. The reaction of 11 (96:4
dr) with 1ÁHOTf led to the exclusive formation of amino alcohols
12, in 52:48 diastereomeric ratio. The obtained yield is comparable
to the result obtained by Upthagrove (22%), in which the free base
was reacted with a terminal epoxide.34 Brief optimization experi-
ments including the use of microwave irradiation (in acetonitrile
as solvent),35 and the use of hexafluoroisopropanol, a solvent that
promotes epoxide opening,36 were unsuccessful (3% and 15% yield
respectively).
Trifluoroacetone 2a (0.23 mL, 2.50 mmol) was cooled to À40 °C
in a dry ice/acetone bath and cannulated into a sealed tube con-
taining CH2Cl2 (0.5 mL). (R)-tert-butanesulfinamide (R)-4 (0.32 g,
2.63 mmol), Ti(OiPr)4 (2.96 mL, 10.0 mmol) and the remaining
CH2Cl2 (0.6 mL) added, the tube sealed and the reaction stirred at
rt for 18 h. Upon completion the reaction was transferred to a lar-
ger vessel and quenched via dropwise addition of H2O (2.5 mL)
while being rapidly stirred. The quenched reaction was allowed
to stir for 5 min, followed by filtration through a pad of MgSO4/
Celite and concentrated in vacuo to yield 5a as a yellow oil. The
resultant oil was partially purified via filtration through a plug of
silica under reduced pressure (isocratic elution of petrol/EtOAc,
5%), concentrated in vacuo and a sample submitted directly for
NMR studies. 1H NMR (500 MHz, CDCl3): d 2.52 (3H, br. d, J
0.3 Hz, CH3CCF3), 1.31 (9H, s, 3 Â CH3) ppm; 19F NMR (471 MHz,
CDCl3): d À75.23 (3F, s, CF3) ppm.
4.2. (RS)-N-((2R)-1,1,1-Trifluoroprop-2-yl)-t-butanesulfinamide
(RS,R)-10
Synthesis on 56 mmol scale (Table 1, entry 10). Trifluoroacetone
2a (5 mL, 55.8 mmol, 1 equiv) was cooled down to À40 °C in a card
ice/acetone bath and then added via syringe to a sealed flask fitted
with an argon balloon and containing dry hexane (50 mL), previ-
ously placed in an ice bath. (R)-(+)-2-Methylpropane-2-sulfi-
namide (R)-4 (dried under vacuum beforehand, 10.1 g,
121.2 mmol, 1.5 equiv) was then added at 0 °C, and the flask was
rinsed with dry CH2Cl2 (50 mL). Titanium isopropoxide (25 mL,
83.7 mmol, 1.5 equiv) was added at 0 °C, and the flask was fitted
with its corresponding screw top lid. The resulting mixture was
allowed to warm up to rt and stirred for 48 h. The reaction mixture
was then cooled down to À30 °C, and flushed with an argon bal-
loon, after which THF (70 mL) was added, followed by NaBH4
(6.2 g, 164.5 mmol, 3.0 equiv). The mixture was stirred at À30 °C
for 0.5 h, and then allowed to warm up to rt and stirred overnight.
The reaction was cooled to 0 °C with vigorous stirring. Water
(60 mL) was then added dropwise and the reaction stirred for
10 min. The reaction mixture was filtered through Celite and the
3. Conclusion
In conclusion, the enantioselective synthesis of 1,1,1-trifluoro-
propan-2-amine was achieved starting from trifluoroacetone using
the sulfinamide auxiliary. The intermediate imine proved too
unstable to isolate, but NOE and HOESY experiments established
its Z-geometry. This is only the second example of determination
of imine geometry derived from trifluoromethyl ketones, the other
case being an
a,b-unsaturated derivative having the E-configura-
tion.14 An in-situ imine formation/reduction protocol with NaBH4
proceeded in high diastereoselectivity (96% de), which represents
a slight improvement over previous methods. Reaction with L-
Selectride gave the opposite facial attack. X-ray crystallographic
analysis of the reduction products unambiguously established
their relative configuration. While the change in facial selectivity
Please cite this article in press as: Packer, G.; et al. Tetrahedron: Asymmetry (2017), http://dx.doi.org/10.1016/j.tetasy.2017.03.003

G. Packer et al. / Tetrahedron: Asymmetry xxx (2017) xxx–xxx
5
filter cake was washed with CH2Cl2. The resulting mixture was
evaporated under reduced pressure to give a white solid. The latter
was partitioned between CH2Cl2 (100 mL) and a saturated solution
of NH4Cl (60 mL). Phases were separated, and 10 mL of HCl (2 M)
were added to the aqueous phase, which was back extracted with
CH2Cl2 (3 Â 50 mL). Brine was eventually added during the extrac-
tion process to facilitate decantation. Organic phases were com-
CF3) ppm. MS (ESI+) m/z (%): 218 [M+H]+ (100). HRMS (ESI+) for C7-
15F3NOS [M+H]+ calcd. 218.0821, found 218.0821.
H
4.4. (R)-1,1,1-Trifluoro-2-propanamine hydrochloride 1ÁHCl
To a stirred solution of (RS,R)-10 (0.56 g, 2.58 mmol, 96:4 d.r.) in
MeOH (3.4 mL) was added dropwise a 4 M HCl in dioxane solution
(2.60 mL, 10.31 mmol) and the reaction stirred at RT for 30 min.
Upon completion the reaction was concentrated in vacuo to yield
a solid/residue. Et2O (15 mL) was added to the resultant solid/
reside in order to precipitate (R)-1ÁHCl, which was filtered and
washed with Et2O (2 Â 5 mL) to yield (R)-1ÁHCl as a white solid
(341.6 mg, 89%, 92% ee). Rf 0.77 (CH2Cl2/MeOH 90:10).
bined, dried over MgSO4 and concentrated to give the crude
a-
trifluorosulfinamide 10 as a mixture of diastereoisomers (dr
98:2). Column chromatography (petroleum ether/EtOAc 85:15 to
6:4) afforded the expected product (RS,R)-10 as a colorless oil
(6.3 g, 52%). The column was then eluted with acetone to recover
the excess (R)-(+)-2-methylpropane-2-sulfinamide (3.3 g). mp
35–37 °C (EtOAc/petrol, 96:4 sample). Rf 0.51 (MeOH/CH2Cl2
[
a
]
23 = À1.15 (c 1.0, MeOH, 23 °C), lit.7
[
a
]
25 = À2.9 (c 1.0, MeOH)
D
D
10:90). [
a
]
22 = À70.4 (c 0.99, CHCl3) 96:4 sample. IR (cmÀ1) 3197
for 98% ee. 1H NMR (500 MHz, MeOD): d 4.21 (1H, app. sept, J
7.0 Hz, CHCF3), 1.51 (3H, br. dd, J 6.9, 0.5 Hz, CH3CHCF3) ppm. 1H
D
(N–H), 2955 (C–H), 1365, 1273 (S = O), 1175 (C-N), 1054 (C-F). 1H
NMR (500 MHz, CDCl3): d 3.86 (1H, app. dsept, J 7.0, 6.8 Hz, CHCF3),
3.58 (1H, br. d, J 5.8 Hz, NH), 1.46 (3H, dq, J 6.9, 0.6 Hz, CH3CHCF3),
1.24 (9H, s, 3 Â CH3) ppm. 1H{19F} NMR (500 MHz, CDCl3): d 3.87
(1H, app. quint, J 6.8 Hz, CHCF3), 3.58 (1H, br. d, J 6.0 Hz, NH),
1.46 (3H, d, J 6.8 Hz, CH3CHCF3), 1.24 (9H, s, 3 Â CH3) ppm. 1H
{
19F} NMR (500 MHz, MeOD): d 4.21 (1H, q, J 6.9 Hz, CHCF3), 1.51
(3H, d, J 6.9 Hz, CH3CHCF3) ppm. 13C NMR (125 MHz, MeOD): d
125.7 (CF3, q, J 279.3 Hz), 49.9 (CHCF3, q, J 32.5 Hz), 12.6 (CH3-
CHCF3, q, J 2.1 Hz) ppm. 19F NMR (471 MHz, MeOD): d À77.97
(3F, d, J 7.0 Hz, CF3) ppm. NMR data correspond to those previously
reported.8 An analytical sample of 1ÁHCl was recrystallized from
hot MeCN by slow evaporation, giving suitable crystals for X-ray
diffraction (see Supporting Information).
{
19F} NMR (500 MHz, CDCl3 + D2O exchange): d 3.87 (1H, q, J
6.8 Hz, CHCF3), 1.45 (3H, d, J 6.8 Hz, CH3CHCF3), 1.24 (9H, s,
3 Â CH3) ppm. 13C NMR (125 MHz, CDCl3): d 125.4 (CF3, q, J
281.1 Hz), 56.4 (S(O)C(CH3)3), 52.6 (CHCF3, q, J 30.8 Hz), 22.3
(3 Â CH3), 14.3 (CH3CHCF3, q, J 2.1 Hz) ppm. 19F NMR (471 MHz,
4.5. (R)-1,1,1-Trifluoro-2-propanamine hydrogen triflate 1ÁHOTf
CDCl3):
d
À77.81 (3F, d,
J
7.0 Hz, CF3) ppm. 19F{1H} NMR
(376 MHz, CDCl3): d À78.06 (3F, s, CF3) ppm. MS (ESI+) m/z (%):
218 [M+H]+ (100), 240 [M+Na]+ (12), 435 [2 M+H]+ (15). HRMS
(ESI+) for C7H15F3NOS [M+H]+ calcd. 218.0821, found 218.0824.
To a solution of a-trifluorosulfinamide (RS,R)-10 (738 mg,
3.34 mmol, 1 equiv, dr 98:2) in dry MeOH (6.5 mL) at 0 °C was
added TMSOTf (800 mL, 4.4 mmol, 1.3 equiv) dropwise. The mix-
ture was allowed to warm up to rt and stirred for 21 h, before con-
centrating under vacuum. Purification via column chromatography
(CH2Cl2/MeOH 95:5 to 9:1) afforded a wet solid, which was then
4.3. (RS)-N-((2S)-1,1,1-Trifluoroprop-2-yl)-t-butanesulfinamide
(RS,S)-10
washed with Et2O to give the
a-trifluoroamine as a white solid
Trifluoroacetone 2a (0.23 mL, 2.50 mmol) was cooled to À40 °C
in a cardice/acetone bath and cannulated into a sealed tube con-
taining THF (1 mL). (R)-tert-Butanesulfinamide (R)-4 (0.32 g,
2.63 mmol), Ti(OiPr)4 (2.96 mL, 10.00 mmol) and the remaining
THF (1.3 mL) added, the tube sealed and the reaction stirred at rt
for 40 h. The reaction mixture was cooled to À78 °C, after which
1 M L-Selectride in THF (2.62 mL, 2.63 mmol) was added and the
reaction mixture stirred at À78 °C for 1 h. Upon completion, the
reaction was quenched via the dropwise addition of sat. NH4Cl
(3 mL). The reaction was filtered through a pad of MgSO4/Celite,
washing with CH2Cl2, and concentrated in vacuo to yield the crude
(803 mg, 90%, ee 96%, estimated from the 98:2 dr of 10). An analyt-
ical sample of 1ÁHOTf was recrystallized by slow evaporation of
MeOH, giving suitable crystals for x-ray diffraction. [a]
22 = +0.3 (c
D
0.64, MeOH), mp 167–170 °C (MeOH), Rf (CH2Cl2/MeOH 90:10)
0.06 (revealed with vanillin); IR (neat) 3449 (m, br.), 3005 (w.),
2543 (w., br.), 1275 (s), 1260 (s) cmÀ1
;
1H NMR (400 MHz, metha-
3
nol-d4) d 4.19 (1H, app. spt, J 7.0 Hz, CH3), 1.49 (3H, d, JHH 6.9 Hz,
CH) ppm; 13C NMR (101 MHz, methanol-d4) d 121.8 (q, JCF
1
1
317.9 Hz, CF3 TfOH), 125.6 (q, JCF 279.3 Hz, CF3 amine), 49.9 (dd,
2
2JCF 66.8 Hz, JCF 33.0 Hz, CH, overlapped with the solvent peak),
3
12.5 (q, JCF 2.2 Hz, CH3) ppm; 19F NMR (376 MHz, methanol-d4)
3
a-trifluorosulfinamide 10 as a mixture of diastereoisomers (dr
d À77.4 (3F, d, JFH 6.9 Hz, CF3 amine); À79.5 (3F, s, CF3 TfOH) ppm.
6:94), as a yellow oil. The resultant oil was purified by column
chromatography (isocratic elution of pentane/EtOAc 70:30) to
yield (RS,S)-10 as a clear syrup which crystallized upon standing
(0.12 g, 22%, 7:93 d.r.). mp 82–85 °C (EtOAc/petrol). Rf 0.51
4.6. Epoxide opening with 1ÁHCl
To a sealed tube fitted with a septum and an argon balloon were
successively added the a-trifluoroisopropylammonium chloride
(MeOH/CH2Cl2 10:90). [
a
]
D
22 = À68.8 (c 0.99, CHCl3). IR (cmÀ1
)
3167 (N–H), 2977 (C–H), 1364, 1279 (S = O), 1171 (C-N), 1054
(C-F). 1H NMR (500 MHz, CDCl3): d 3.85 (1H, app. dsept, J 7.5,
7.0 Hz, CHCF3), 3.14 (1H, br. d, J 7.5 Hz, NH), 1.47 (3H, br. dq, J
6.9, 0.4 Hz, CH3CHCF3), 1.24 (9H, s, 3 Â CH3) ppm. 1H NMR
(500 MHz, CDCl3 + D2O exchange): d 3.84 (1H, app. sept, J 7.0 Hz,
CHCF3), 1.47 (3H, dq, J 7.0, 0.6 Hz, CH3CHCF3), 1.24 (9H, s,
3 Â CH3) ppm. 1H{19F} NMR (500 MHz, CDCl3): d 3.85 (1H, dq, J
7.8, 7.0 Hz, CHCF3), 3.14 (1H, br. d, J 7.4 Hz, NH), 1.48 (3H, d, J
6.9 Hz, CH3CHCF3), 1.24 (9H, s, 3 Â CH3) ppm. 1H{19F} NMR
(500 MHz, CDCl3 + D2O exchange): d 3.84 (1H, q, J 6.9 Hz, CHCF3),
1.48 (3H, d, J 6.9 Hz, CH3CHCF3), 1.24 (9H, s, 3 Â CH3) ppm. 13C
NMR (125 MHz, CDCl3): d 125.3 (CF3, q, J 280.4 Hz), 56.7 (S(O)C
(CH3)3), 54.5 (CHCF3, q, J 31.1 Hz), 22.3 (3 Â CH3), 16.5 (CH3CHCF3,
q, J 2.1 Hz) ppm. 19F NMR (376 MHz, CDCl3): d À78.53 (3F, d, J
6.9 Hz, CF3) ppm. 19F{1H} NMR (471 MHz, CDCl3): d À78.28 (3F, s,
salt 1ÁHCl (151 mg, 1.0 mmol, 1.2 equiv), MeCN (2.8 mL), Et3N
(234 mL, 1.68 mmol, 2 equiv) and MeOH (0.5 mL) at rt. The mixture
was stirred for 30 min, after which cyclohexene oxide 11 (85 mL,
0.84 mmol, 1 equiv) was added, followed by LiClO4 (179 mg,
1.68 mmol, 2 equiv). The sealed tube was fitted with its corre-
sponding lid, and the reaction was stirred at 80 °C for 5 days, then
allowed to warm up to rt and stirred for 5 more days. Following
this, the reaction mixture was concentrated under vacuum. The
crude product was re-dissolved in CH2Cl2 (10 mL), washed with
H2O (1 Â 10 mL). Aqueous phase extracted with CH2Cl2
(3 Â 10 mL). The organic layers were combined, dried over MgSO4,
and concentrated under reduced pressure to yield a brown oil.
Purification by column chromatography (petroleum ether/ethyl
acetate 9:1 to 7:3) gave the chlorohydrin product 13a as a yellow
oil (25 mg, 22%), alongside trace amounts of the secondary amine
Please cite this article in press as: Packer, G.; et al. Tetrahedron: Asymmetry (2017), http://dx.doi.org/10.1016/j.tetasy.2017.03.003

6
G. Packer et al. / Tetrahedron: Asymmetry xxx (2017) xxx–xxx
(ꢀ1 mg, ꢀ1%). 1H NMR (400 MHz, CDCl3) 3.73 (1H, ddd, J 11.7, 9.2,
4.5 Hz), 3.51 (1H, td, J 9.7, 4.9 Hz), 2.75–2.38 (1H, m), 2.29–2.17
(1H, m), 2.16–2.02 (1H, m), 1.81–1.70 (2H, m), 1.70–1.56 (1H,
m), 1.44–1.21 (3H, m) ppm; 13C NMR (100 MHz, CDCl3) d ppm
75.3, 67.5, 35.1, 33.1, 25.6, 23.9 ppm. NMR data correspond to
those previously reported.37
A. Supplementary data
Supplementary data (2D spectra related to imine geometry
assignment, determination of diastereomeric ratio’s, copies of 1H,
13C, 19F NMR spectra of all novel compounds, X-ray crystallo-
graphic data of the reduction products and amine salts) associated
with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.tetasy.2017.03.003.
4.7. Epoxide opening with 1ÁOTf
To a sealed tube fitted with a septum and an argon balloon were
successively added the trifluoro isopropylammonium triflate salt
1ÁHOTf (265 mg, 1.0 mmol, 1.2 equiv), MeCN (2.8 mL), Et3N
(234 mL, 1.68 mmol, 2 equiv), and MeOH (0.5 mL) at rt. Mixture
was stirred for 30 min, then cyclohene oxide 11 (85 mL, 0.84 mmol,
1 equiv) was added, followed by LiClO4 (179 mg, 1.68 mmol,
2 equiv). Sealed tube was fitted with its corresponding lid, and
the reaction was stirred at 80 °C for 10 days. Following this, the
reaction mixture concentrated under vacuum. The crude product
was redissolved in CH2Cl2 (10 mL), washed with H2O (1 Â 10 mL).
Aqueous phase extracted with CH2Cl2 (3 Â 10 mL). The organic lay-
ers were combined, dried over MgSO4, and concentrated under
reduced pressure to yield a brown oil. Purification by column chro-
matography (petroleum ether/EtOAc 9:1 to 7:3) gave the expected
secondary amine 12 as a yellow oil and as a 52:48 mixture of
diastereoisomers (51 mg, 29%). Purification by preparative HPLC
(CH2Cl2/MeOH 98:2) enabled separation and isolation of two pure
fractions of each diastereoisomer, for characterisation purpose.
References
1. Raasch, M. S. J. Org. Chem. 1962, 27, 1406.
2. Dickey, J. B.; Towne, E. B.; Bloom, M. S.; Taylor, G. J.; Wallace, D. J.; Sagal, J.;
McCall, M. A.; Hedberg, D. G. Ind. Eng. Chem. 1956, 48, 209.
3. Shepard, R. A.; Sciaraffa, P. L. J. Org. Chem. 1966, 31, 964.
4. Ono, T.; Kukhar, V. P.; Soloshonok, V. A. J. Org. Chem. 1996, 61, 6563.
5. Pfrengle, W.; Pees, K.-J.; Albert, G.; Carter, P.; Rehnig, A.; Van Tuyl Cotter, H.;
US5986135, 1999.
6. Zhang, X.; Zhang, N.; Chen, G.; Turpoff, A.; Ren, H.; Takasugi, J.; Morrill, C.; Zhu,
J.; Li, C.; Lennox, W.; Paget, S.; Liu, Y.; Almstead, N.; Njoroge, F. G.; Gu, Z.;
Komatsu, T.; Clausen, V.; Espiritu, C.; Graci, J.; Colacino, J.; Lahser, F.; Risher, N.;
Weetall, M.; Nomeir, A.; Karp, G. M. Bioorg. Med. Chem. Lett. 2013, 23, 3947.
7. Soloshonok, V. A.; Ono, T. J. Org. Chem. 1997, 62, 3030.
8. Wu, Y.; Deng, L. J. Am. Chem. Soc. 2012, 134, 14334.
9. Arnone, A.; Bravo, P.; Capelli, S.; Fronza, G.; Meille, S. V.; Zanda, M.; Cavicchio,
G.; Crucianelli, M. J. Org. Chem. 1996, 61, 3375.
10. Bravo, P.; Guidetti, M.; Viani, F.; Zanda, M.; Markovsky, A. L.; Sorochinsky, A. E.;
Soloshonok, I. V.; Soloshonok, V. A. Tetrahedron 1998, 54, 12789.
11. Robak, M. T.; Herbage, M. A.; Ellman, J. A. Chem. Rev. 2010, 110, 3600.
12. Wang, H.; Zhao, X.; Li, Y.; Lu, L. Org. Lett. 2006, 8, 1379.
13. Grellepois, F. J. Org. Chem. 2013, 78, 1127.
14. Liu, Z.-J.; Liu, J.-T. Chem. Commun. 2008, 5233.
15. Xu, J.; Liu, Z.-J.; Yang, X.-J.; Wang, L.-M.; Chen, G.-L.; Liu, J.-T. Tetrahedron 2010,
66, 8933.
4.8. Data for diastereomer 1
16. Kawanami, T.; Karns, A. S.; Adams, C. M.; Serrano-Wu, M. Tetrahedron Lett.
2013, 54, 7202.
17. Review: Mei, H. B.; Xie, C.; Han, J. L.; Soloshonok, V. A. Eur. J. Org. Chem. 2016,
5917.
18. Li, P.; Jiang, M.; Liu, J.-T. Chin. J. Chem. 2014, 32, 1003.
19. Yang, Y.; Huang, Y.; Qing, F.-L. Tetrahedron Lett. 2013, 54, 3826.
20. Wang, L.; Shen, Q.; Lu, L. Chin. J. Chem. 2013, 31, 892.
21. Colyer, J. T.; Andersen, N. G.; Tedrow, J. S.; Soukup, T. S.; Faul, M. M. J. Org. Chem.
2006, 71, 6859.
22. Hua, D. H.; Miao, S. W.; Chen, J. S.; Iguchi, S. J. Org. Chem. 1991, 56, 4.
23. Zhang, W.; Sha, W.; Zhu, Y.; Han, J.; Soloshonok, V. A.; Pan, Y. Eur. J. Org. Chem.
2017. http://dx.doi.org/10.1002/ejoc.201601645.
24. Davis, F. A.; Kluger, E. W. J. Am. Chem. Soc. 1976, 98, 302.
25. Liu, G. C.; Cogan, D. A.; Owens, T. D.; Tang, T. P.; Ellman, J. A. J. Org. Chem. 1999,
64, 1278.
26. Cogan, D. A.; Liu, G. C.; Ellman, J. A. Tetrahedron 1999, 55, 8883.
27. Cogan, D. A.; Ellman, J. A. J. Am. Chem. Soc. 1999, 121, 268.
28. Barrow, J. C.; Ngo, P. L.; Pellicore, J. M.; Selnick, H. G.; Nantermet, P. G.
Tetrahedron Lett. 2001, 42, 2051.
[a]
22 = +28.2 (c 0.59, CHCl3); Rf (CH2Cl2/MeOH 98:2) 0.37; IR
D
(neat) 3395 (m, br.), 2933 (m), 1451 (w), 1266 (m), 1146 (s)
cmÀ1 1H NMR (400 MHz, CDCl3) d 3.23 (1H, app. spt, J 7.0 Hz),
;
3.16–3.07 (2H, m, simplified as 1H, td, J 9.8, 4.7 Hz upon D2O
shake), 2.36 (1H, ddd, J 11.5, 9.1, 3.7 Hz), 2.11–2.00 (2H, m)
1.81–1.62 (2H, m), 1.35–1.16 (6H, m), 1.04–0.88 (1H, m) ppm;
13C NMR (100 MHz, CDCl3) d 126.7 (q, J 281.7 Hz), 74.3, 63.8,
53.5 (q, J 28.9 Hz), 32.9, 32.0, 25.2, 24.2, 16.4 (q, J 2.20 Hz) ppm;
19F NMR (376 MHz, CDCl3) d À78.09 (3F, d, J 8.7 Hz, simplified as
s upon proton decoupling). MS [ESI+] 212 [M+H]+, HRMS [ESI+]
for C9H16F3NO [M+H]+ calcd. 212.1257, found 212.1261.
4.9. Data for diastereomer 2
29. Kochi, T.; Tang, T. P.; Ellman, J. A. J. Am. Chem. Soc. 2002, 124, 6518.
30. Kochi, T.; Tang, T. P.; Ellman, J. A. J. Am. Chem. Soc. 2003, 125, 11276.
31. McMahon, J. P.; Ellman, J. A. Org. Lett. 2005, 7, 5393.
32. Denolf, B.; Leemans, E.; De Kimpe, N. J. Org. Chem. 2007, 72, 3211.
33. Use of TMSCl in MeOH: (a) Qian, X.; Tang, E.; Fan, J.; Zhu, K.; Zhu, J.; Shabaker,
J.; Lo, E.; Malley, M.; Deshpande, R. Tetrahedron 2015, 71, 9408; (b) Guerin, D.;
Dez, I.; Gaumont, A.-C.; Pannecoucke, X.; Couve-Bonnaire, S. Org. Lett. 2016, 18,
3606.
34. Upthagrove, A. L.; Nelson, W. L. Drug Metab. Dispos. 2001, 29, 1377.
35. Robin, A.; Brown, F.; Bahamontes-Rosa, N.; Wu, B.; Beitz, E.; Kun, J. F. J.; Flitsch,
S. L. J. Med. Chem. 2007, 50, 4243.
36. Das, U.; Crousse, B.; Kesavan, V.; Bonnet-Delpon, D.; Begue, J. P. J. Org. Chem.
2000, 65, 6749.
[
a
]
22 = À81 (c 0.15, CHCl3); Rf (CH2Cl2/MeOH 98:2) 0.24; IR
D
(neat) 3322 (w, br.), 2923 (w, br.), 1454 (w), 1274 (m), 1260 (m)
cmÀ1 1H NMR (400 MHz, CDCl3) d 3.30 (1H, app. spt, J 7.0 Hz),
;
3.20 (1H, td, J 9.8, 4.5 Hz), 2.82–2.60 (1H, m, disappeared upon
D2O shake), 2.36 (1H, ddd, J 11.5, 9.5, 4.2 Hz), 2.12–1.90 (2H, m),
1.80–1.67 (2H, m), 1.63–1.41 (1H, m, disappeared upon D2O
shake), 1.36–1.15 (6H, m), 1.13–0.95 (1H, m) ppm; 13C NMR
(100 MHz, CDCl3) d 126.7 (q, J 281.0 Hz) 74.1, 61.2, 52.7 (q, J
28.6 Hz), 33.4, 31.1, 25.0, 24.3, 14.9 (q, J 2.20 Hz) ppm; 19F NMR
(376 MHz, CDCl3) d À78.0 (3F, d, J 6.9 Hz, simplified as s upon pro-
ton decoupling) ppm; MS [ESI+] 212 [M+H]+; HRMS [ESI+] for C9-
37. Gnanamani, E.; Someshwar, N.; Sanjeevi, J.; Ramanathan, C. R. Adv. Synth. Catal.
2014, 356, 2219.
38. Zhou, Y.; Wang, J.; Gu, Z. N.; Wang, S. N.; Zhu, W.; Acena, J. L.; Soloshonok, V. A.;
Izawa, K.; Liu, H. Chem. Rev. 2016, 116, 422.
H
16F3NO [M+H]+ calcd. 212.1257, found 212.1261.
39. Wang, J.; Sánchez-Roselló, M.; Aceña, J. L.; del Pozo, C.; Sorochinsky, A. E.;
Fustero, S.; Soloshonok, V. A.; Liu, H. Chem. Rev. 2014, 114, 2432.
40. O’Hagan, D. J. Fluorine Chem. 2010, 131, 1071.
Acknowledgements
41. Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V. Chem. Soc. Rev. 2008, 37,
320.
42. Fujiwara, T.; O’Hagan, D. J. Fluorine Chem. 2014, 167, 16.
43. Jeschke, P. Pest Manag. Sci. 2010, 66, 10.
The European Community (INTERREG IVa Channel Programme,
AI-Chem, project 4494/4196), the EPSRC (EP/K503150/1 and core
capability EP/K039466/1), and the University of Southampton are
thanked for financial support.
Please cite this article in press as: Packer, G.; et al. Tetrahedron: Asymmetry (2017), http://dx.doi.org/10.1016/j.tetasy.2017.03.003