Enantioselective synthesis of optically active homoallylamines by nucleophilic addition of chirally modified allylboranes to N-silylimines

Article abstract of DOI:10.1039/a902635e

Enantioselectivity in the allylboration of W-silylimines with a variety of chirally modified allylboron reagents has been examined. Optically active N-sulfonylarhino alcohols (16,17 and 19) derived from D-camphor and norephedrine were found to be efficient chiral ligands for the allylboration reagent. These reagents smoothly reacted with Ar-silylimines to give the corresponding homoallylic amines in a high level of enantioselectivity up to 96% ee.

Full text of DOI:10.1039/a902635e

View Article Online / Journal Homepage / Table of Contents for this issue
Enantioselective synthesis of optically active homoallylamines
by nucleophilic addition of chirally modified allylboranes to
N-silylimines
Shinichi Itsuno,* Katsuhiro Watanabe, Takeshi Matsumoto, Shizue Kuroda, Ayako Yokoi and
Ashraf El-Shehawy†
Department of Materials Science, Toyohashi University of Technology, Toyohashi, 441-8580,
Japan
Received (in Cambridge) 1st April 1999, Accepted 1st June 1999
Enantioselectivity in the allylboration of N-silylimines with a variety of chirally modified allylboron reagents
has been examined. Optically active N-sulfonylamino alcohols (16, 17 and 19) derived from -camphor and
norephedrine were found to be efficient chiral ligands for the allylboration reagent. These reagents smoothly
reacted with N-silylimines to give the corresponding homoallylic amines in a high level of enantioselectivity
up to 96% ee.
Table 1 Allylboration of various N-substituted imines with triallyl-
borane in THF at room temperature
Introduction
In recent years, considerable progress has been made in the
asymmetric addition of carbon nucleophiles to prochiral
carbonyl compounds. Especially, enantioselective allylation of
aldehydes has been extensively studied and various efficient
chiral allylating agents have been developed.1 Chirally modified
allylboron reagents have been particularly important for allyl-
ations.2 By comparison, the asymmetric addition of carbon
nucleophiles to prochiral imines leading to optically active
amines is a field which is still in its infancy. Optically active
amines are important compounds utilized extensively in
organic synthesis as starting materials in the preparation of
biologically active substances, resolving agents, and chiral aux-
iliaries for asymmetric synthesis.3 Although the enantioselective
synthesis of optically active amines by nucleophilic addition of
organometallic reagents toward imines is a topic of current
interest,4 the enantioselective addition of allylmetallic reagents
Yield of
Reaction
time (t/h)
homoallylamine
Entry
Imine
N-Substituent
(%)a
1
2
3
4
5
6
1
2
3
4
5
6
4-MeOC6H4
Tr
MeO
Ac
PhS
Ts
16
24
24
24
30
20
55b
NRc
36d
NRc
55
Complex
mixture
NRc
90
7
7
8a
Ph2P(O)
Me3Si
20
3
8e
a Isolated yields. b The corresponding 4-methoxyphenylamine was
obtained. c No reaction. d The corresponding methoxyamine was
obtained. e At 0 ЊC.
to the C᎐N double bond is quite underdeveloped.5 There are
᎐
only a few reports on the asymmetric allylation of aldimines.6–8
We have recently found that some chirally modified allylboron
reagents could asymmetrically add to N-metalloimines such
as N-silylimines9 and N-aluminioimines10 to afford the corre-
sponding enantioenriched homoallylamines. Since this method-
ology appeared to be potentially useful for preparing certain
enantioenriched homoallylamines, we have examined the scope
and limitation of this procedure. We report herein the details
of our efforts to find suitable conditions for preparing highly
enantioenriched homoallylamines by nucleophilic addition of
chirally modified allylboron reagents to N-silylimines.
B
H
X
H
HN
N
3
R
R
OMe
OAc
H
Tr
H
OMe
H
N
N
N
N
N
Ph
Ph
Ph
Ph
Ph
H
1
2
3
4
SiR2R3R4
Ph
O
Ph
N
SPh
H
N
Ts
H
Results and discussion
P
H
N
Before our study of asymmetric allylboration of imines, we sur-
veyed the reactivity of various imines toward triallylboranes
(Scheme 1, Table 1). Allylboration of N-arylimine 1 gave the
secondary homoallylic amine in moderate yield after 16 h at
room temperature. Bulky N-substituents such as a trityl group
completely prevented the allylboration reaction even after 24 h
at room temperature (entry 2). Oxime ether 3 reacted with tri-
allylborane slowly to give the corresponding methoxyamine in
low yield. Sulfenimine 5 was converted to the sulfenamide that
H
Ph
Ph
R1
5
6
7
8
8a: R1 = H, R2 = R3 = R4 = Me
8b: R1 = o-Cl, R2 = R3 = R4 = Me
8c: R1 = p-Cl, R2 = R3 = R4 = Me
8d: R 1= o-OMe, R2 = R3 = R4 = Me
8e: R 1= p-OMe, R2 = R3 = R4 = Me
8f: R1 = H, R2 = R3 = Me, R4 = tBu
8g: R1 = H, R2 = R3 = R4 = iPr
† Current address: Chemistry Department, Faculty of Education,
Tanta University, Kafr El-Sheikh, Egypt.
Scheme 1
J. Chem. Soc., Perkin Trans. 1, 1999, 2011–2016
2011

View Article Online
Table 2 Enantioselective allylboration of N-(trimethylsilyl)imines with chirally modified allylborane 10
Homoallylamine 9
Entry
Silylimine
Solvent
T/ЊC
Yield (%)a
ee (%)b
Config.c
1
2
3
4
5
6
7
8
8a
8a
8a
8a
8b
8c
8d
8e
THF
Ether
Ether
Ether
THF
THF
THF
THF
Ϫ78
65
70
76
84
62
72
65
70
64
73
43
29
46
8
S
S
S
S
S
S
S
S
Ϫ78
Ϫ30
0
Ϫ78
Ϫ78
Ϫ78
Ϫ78
52
14
a Isolated yields. b Determined by HPLC analysis with a chiral stationary-phase column. See Experimental section. c The absolute configuration of
the product was determined by comparison of the specific rotation with the reported value (ref. 19).
Table 3 Enantioselective allylboration of N-(trimethylsilyl)benzald-
imine 8a with chirally modified allylboron reagents for 3 h at Ϫ78 ЊC
B
SiMe3
H
1-Phenylbut-3-enamine 9a
N
Chiral
ligand
Entry
Solvent
Yield (%)a
ee (%)b
Config.c
10
1) H3O+
1
2
3
4
5
6
7
8
9
10
11
12
13
11a
11b
11c
12
THF
THF
THF
THF
THF
THF
THF
Ether
Ether
THF
Ether
Ether
THF
89
70
83
92
26
14
90
65
93
81
88
12
72
39
25
32
32
28
14
40
65
72
53
78
10
27
R
R
R
R
R
S
R
S
R
S
R
8
2) NH4OH
13
NH2
14
15
16a
17
18
19a
21
22
R
9
S
R
R
Scheme 2
a Isolated yields. b Determined by HPLC analysis with a chiral
stationary-phase column. See Experimental section. c The absolute
configuration of the product was determined by comparison of the
specific rotation with the reported value (ref. 19).
higher enantioselectivity. The enantioselectivity decreased with
increasing temperature as expected (Table 2, entries 2–4). In all
cases the si-face of the imine was attacked by 10 to yield the
S-homoallylamine. However, the enantioselectivities were very
much affected by the substituents on the aromatic ring of the
imine. Very recently, Brown and co-workers reported a modified
method for the asymmetric allylboration of N-silylimines,
which showed higher enantioselectivities.15 They claimed that
N-trimethylsilylimines were not reactive toward 10, and sug-
gested that aldimines generated in situ from N-trimethyl-
silylimines by addition of 1 mol equiv. of water could react with
10. From these observations, the reaction in our case might
occur during aqueous work-up.
To examine the enantioselectivity in allylboration of N-silyl-
imines, we have prepared various kinds of chirally modified
allylboron reagents other than 10. These reagents were readily
prepared from the reaction of triallylborane with chiral
bidentate ligands such as chiral diols 11 and 12, hydroxy acid
13, N-sulfonylamino acid 14, and N-sulfonylamino alcohols 15–
22. As shown in Table 3, asymmetric allylboration of 8a took
place smoothly with these chirally modified allylboron reagents
at Ϫ78 ЊC to give the corresponding primary homoallylamine
9a. Although tartrate ligands 11 developed by Roush showed
high efficiency in the allylboration of aldehydes,1g,16 not very
good selectivity was obtained in the allylboration of the
imine (entries 1–3). Enantiopure N-sulfonylamino alcohols
derived from -camphor and norephedrine showed relatively
higher enantioselectivity in the allylboration (entries 8–11).
However, contrary to the reaction with 10, addition of water
severely prevented the allylboration with these reagents giving
benzaldehyde instead. N-Trimethylsilylimine is reactive enough
toward these reagents. GC analysis of the reaction mixture
showed that the N-trimethylsilylimine was consumed before
aqueous work-up, which also supported the above mentioned
reactivity of the reagents.
could be elaborated to the primary homoallylamine by acid
treatment. Although N-p-tolylsulfonyl-11,12 and N-diphenyl-
phosphinoylimines13 are known to react with some organo-
metallic reagents to give the corresponding amines, the reaction
of N-tosylimine 6 with triallylborane yielded only a complex
mixture. N-Diphenylphosphinoylimine 7 showed no reaction
with triallylborane at room temperature. Finally we found that
N-trimethylsilylimine 8a was relatively highly reactive toward
triallylborane, and gave the homoallylamine in high yield as
shown in Table 1 (entry 8). The trimethylsilyl group was easily
removed during the usual aqueous work-up procedure to give
the primary amine. The high reactivity of the N-trimethylsilyl-
imine toward triallylborane prompted us to apply this reaction
to its asymmetric version. Since various excellent chiral allyl-
boron reagents have been developed for the enantioselective
allylboration of aldehydes to prepare homoallylic alcohols,1 we
first chose one of these chiral reagents for the enantioselective
allylboration of N-silylimines.
In the first place, enantiopure allyldiisopinocampheyl-
borane14 10 developed by Brown was employed as enantio-
selective allylboration agent for 8a (Scheme 2). This reagent is
known as one of the most effective allylborating agents for
aldehydes. Another advantage of this reagent is its easy accessi-
bility from commercially available (Ϫ)-chlorodiisopino-
campheylborane [(Ϫ)-DIP-Chloride]. Table 2 showed that
enantioselective allylboration of several N-(trimethylsilyl)-
imines with 10. Allylboration of 8a with 10 in THF at Ϫ78 ЊC
afforded the enantioenriched homoallylamine 9a in satisfactory
yield with 64% ee (entry 1). The use of diethyl ether (herein-
after abbreviated simply as ether) as solvent gave somewhat
2012
J. Chem. Soc., Perkin Trans. 1, 1999, 2011–2016

View Article Online
Table 4 Enantioselective allylboration of N-(trialkylsilyl)benzaldimines 8 with chirally modified allylboron reagents derived from 16 and 17 at
Ϫ78 ЊC for 3 h
1-Phenylbut-3-enamine
Trialkylsilyl
group (imine)
Chiral
ligand
Entry
Solvent
Yield (%)a
ee (%)b
Config.c
1
2
3
4
5
6
7
8
TMS (8a)
TMS (8a)
TBDMS (8f)
TIPS (8g)
TMS (8a)
TBDMS (8f)
TIPS (8g)
TMS (8a)
TMS (8a)
TMS (8a)
TMS (8a)
TMS (8a)
16a
16b
16b
16b
16c
16c
16c
17
Ether
Ether
Ether
Ether
Ether
Ether
Ether
Ether
Ether
THF
65
90
83
54
37
73
20
93
92
92
73
67
65
89
67
55
86
83
58
72
76
64
53
35
S
S
S
S
S
S
S
R
R
R
R
R
9d
10
11
12
17
17
17
Toluene
Hexane
17
a Isolated yields. b Determined by HPLC analysis with a chiral stationary-phase column. See Experimental section. c The absolute configuration of
the product was determined by comparison of the specific rotation with the reported value (ref. 19). d At Ϫ100 ЊC.
Table 5 Effect of N-substituent in the chiral ligand 19 on the allyl-
HO
HO
CO2R
CO2R
Ph
Ph
Ph
boration of 8a
Ph
OH
CO2H
14
HO
CO2H
HO
1-Phenylbut-3-enamine
TsHN
Chiral
ligand
12
13
11a R = Me
11b R = Et
11c R = iPr
Entry
Solvent
Yield (%)a
ee (%)b
Config.c
1
2
19a
19b
19b
19b
19b
19b
19c
19d
19e
20
Ether
Ether
Ether
THF
Toluene
Hexane
Ether
Ether
Ether
Ether
88
89
80
91
76
64
91
92
93
90
78
92
96
87
77
68
87
88
91
92
S
S
S
S
S
S
S
S
S
R
3d
4
Ph
Ph
OH
NHSO2R
OH
5
6
7
8
9
10
NHTs
OH
TsHN
15
16a R = Me
17
16b R = Tol
16c R = 2-Naph
a Isolated yields. b Determined by HPLC analysis with a chiral
stationary-phase column. See Experimental section. c The absolute
configuration of the product was determined by comparison of the
specific rotation with the reported value (ref. 19). d At Ϫ100 ЊC.
Me
Ph
Me
Ph
Ph
Ph
NHTs
RO2SNH
OH
OH
TsNH
OH
TsHN
OH
18
20
21
19a R = Me
19b R = Tol
derivative 17 preferred re-face attack to give R-homoallylamine
in all solvents tested.
19c R = o-NO2Ph
19d R = 1-Naph
19e R = 2-Naph
Among the various chiral ligands used in this study, those
represented by structure 19 derived from norephedrine
appeared to be the most suitable for the allylboration of the
imine (Table 3). The effect of the substituent of the enantiopure
N-sulfonylamino alcohols 19 on the reactivity and selectivity
has then been investigated. The results are shown in Table 5.
Of various substituents investigated tolylsulfonyl 19b gave
the best result. It is noteworthy that this reaction occurred even
at Ϫ100 ЊC in high yield (entry 3). Enantioselectivity obtained
at this temperature was always somewhat higher than that
obtained at Ϫ78 ЊC (Table 4 entry 9, Table 5 entry 3, Table 6
entry 5). Thus, the highest ee (96% ee) for the allylboration of
8a was attained by using 19b in ether at Ϫ100 ЊC (Table 5, entry
3). Both enantiomers of norephedrine are commercially avail-
able. From (1R,2S)-norephedrine, S-homoallylamine was ob-
tained, while R-homoallylamine with the same ee was obtained
from (1S,2R)-norephedrine (entries 2, 10).
Not only allylboration, but also methallyl- and prenylbor-
ation were possible in this system. Instead of an allylboron
reagent, methallyl- or prenylboron reagents were prepared
using 17 and 19b as chiral ligands. They also reacted smoothly
with imine 8a to give the corresponding homoallylamines with
high ee as shown in Table 6. In the case of methallylboration of
8a with the chiral reagent prepared from 19b the homo-
allylamine was obtained in 94% yield with 96% ee at Ϫ100 ЊC
(entry 5).
Me
HO
CO2Me
NHTs
22
The effect of the silyl substituent of N-silylimines 8 was then
investigated by using camphor-derived chiral allylboron
reagents (Table 4). The synthesis of N-(trialkylsilyl)imines
possessing bulky silyl groups such as tert-butyldimethylsilyl
(TBDMS) and triisopropylsilyl (TIPS) has recently been
reported by Cainelli et al.17 Enantioselective allylboration of
these N-silylimines with the reagents derived from sulfonamides
16 was tested. In spite of a bulky substituent on the imine nitro-
gen, the reaction occurred at Ϫ78 ЊC. In the case of the TIPS
imine, both the yield and ee decreased. By using these bulky
trialkylsilyl groups, aromatic as well as aliphatic enolizable
imines could be prepared.17 Although we have prepared such
aliphatic N-(tert-butyldimethylsilyl)imines, unfortunately, no
reaction occurred when they were treated with 16a. The solvent
effect was examined by using chiral ligand 17 as shown in Table
4 (entries 8–12). From these results ether is the choice of sol-
vent. A similar tendency was observed when the chiral ligand
19b was employed (Table 5). In the case of camphor-based
enantiopure N-sulfonylamino alcohols, exo-derivatives 16 pre-
ferred si-face attack to give S-homoallylamine, while endo-
J. Chem. Soc., Perkin Trans. 1, 1999, 2011–2016
2013

View Article Online
azide according to the literature procedure.18 N-Trialkyl-
silylimines 8f and 8g were prepared according to Cainelli’s
procedure.17
Table 6 Enantioselective allylboration of N-(trimethylsilyl)benzald-
imine 8a with chirally modified allylboron reagents in ether at Ϫ78 ЊC
Ts
N
NH2 R1
R2
R2
B
R*
General procedure for the preparation of the N-sulfonylamino
alcohols 15–22
8a
+
O
Ph
R1
R2 R2
The preparation of 19b is typical. To a solution of (1R,2S)-
norephedrine (1.51 g, 10 mmol) in THF (50 cm3) were added
triethylamine (1.4 cm3, 10 mmol) and toluene-p-sulfonyl chlor-
ide (1.9 g, 10 mmol) at 0 ЊC. After being stirred for 3 h at room
temperature, the reaction mixture was poured into 1 M aq. HCl
(50 cm3), and the THF was evaporated under vacuum. The
resulting aqueous solution was extracted with ether (3 × 25
cm3). The combined extracts were dried over MgSO4 and evap-
orated at reduced pressure to give the crude product. Recrystal-
lization from ethanol–water gave 19b in 96% yield; [α]D25 Ϫ15.93
(c 2.48 in CHCl3); mp 82 ЊC; νmax(KBr) 3500, 3255, 1591, 1456,
1324, 1160; δH(CDCl3; 270 MHz) 0.82 (3H, d, J 6.8), 2.41 (3H,
s), 2.86 (1H, br), 3.52–3.58 (1H, m), 4.79 (1H, d, J 3.4), 5.16
(1H, d, J 8.8), 7.21–7.33 (7H, m), 7.77 (2H, d, J 8.3) (Found: C,
62.82; H, 6.28; N, 4.25. C16H19NO3S requires C, 62.92, H, 6.28;
N, 4.59%).
Homoallylamine
Chiral
ligand
Entry
R1
R2
Yield (%)a
ee (%)b
Config.c
1d
2
3
17
Me
Me
H
Me
Me
H
H
H
Me
H
H
89
89
91
92
94
89
72
80
68
94
96
87
R
R
R
S
S
S
17
17
4
19b
19b
19b
5e
6
Me
a Isolated yields. b Determined by HPLC analysis with a chiral
stationary-phase column. See Experimental section. c The absolute
configuration of the product was determined by comparison of the
specific rotation with the reported value (ref. 19). d THF was used as
solvent. e At Ϫ100 ЊC.
14. Yield 83%; [α]D25 ϩ17.1 (c 2.23 in CHCl3); mp 150–153 ЊC;
νmax(KBr) 3332, 3178, 2962, 1727, 1691, 1468, 1346, 1141, 1089;
δH(CDCl3; 270 MHz) 0.85 (3H, d, J 6.8), 0.94 (3H, d, J 6.8),
2.09 (1H, br), 2.40 (3H, s), 3.77 (1H, dd, J 9.8 and 4.4), 5.43
(1H, d, J 9.8), 7.26 (2H, d, J 7.8), 7.71 (2H, d, J 8.3) (Found: C,
53.29; H, 6.28; N, 4.86. C12H17NO4S requires C, 53.14; H, 6.27;
N, 5.17%).
In conclusion, we have investigated the enantioselectivity
in the reaction of chirally modified allylboron reagents with
N-silylimines. Of various chiral ligands developed in this study,
N-sulfonylamino alcohols such as 16, 17 and 19 showed high
levels of enantioselectivity with good yields of the primary
homoallylic amines. These chiral ligands were readily prepared
and recovered in high yield after the reaction. The foregoing
results clearly demonstrate that our methodology provides a
practical way of obtaining optically active primary homoallylic
amines with high ee. Related reactions by using the polymer-
supported N-sulfonylamino alcohols as solid phase chiral
ligands are in progress.
15; Yield 84%; [α]D25 ϩ20.06 (c 2.81 in THF); mp 195–198 ЊC;
νmax(KBr) 3530, 3300, 2960, 1455, 1325, 1158, 1098; δH(CDCl3;
270 MHz) 0.67 (3H, t, J 7.3), 0.84–0.95 (1H, m), 0.98 (3H, d,
J 6.8), 1.16–1.22 (1H, m), 1.58–1.64 (1H, m), 2.37 (3H, s), 2.66
(1H, s), 4.37 (1H, d, J 9.8), 4.66 (1H, d, J 9.3), 7.10–7.43 (14H,
m) (Found: C, 70.89; H, 6.94; N, 3.31. C25H29NO3S requires C,
70.92; H, 6.98; N, 3.31%).
16a. Yield 83%; [α]D25 ϩ31.46 (c 2.30 in CHCl3); mp 128–
130 ЊC; νmax(KBr) 3498, 3239, 2959, 1481, 1306, 1151;
δH(CDCl3; 270 MHz) 0.82 (3H, s), 0.95 (3H, s), 1.08 (3H, s),
1.50–1.84 (5H, m), 2.40 (1H, br s), 2.98 (3H, s), 3.48 (1H, t,
J 7.3), 3.70–3.80 (1H, m), 5.08 (1H, br s) (Found: C, 53.44;
H, 8.50; N, 5.67. C11H21NO3S requires C, 53.41; H, 8.56; N,
5.66%).
Experimental
General
All reactions were carried out under an atmosphere of dry
nitrogen. Tetrahydrofuran (THF) and ether were freshly dis-
tilled from sodium benzophenone ketyl under nitrogen
immediately before use. (Ϫ)-Chlorodiisopinocampheylborane
[(Ϫ)-DIP-Chloride] was purchased from Aldrich, Inc. Reac-
tions were monitored by TLC using Merck precoated silica gel
plates (Merck 5554, 60F254). Flash column chromatography was
performed over Wako silica gel (Wakogel C-200, 100–200
mesh). Mps were measured on a Yanaco micro melting point
apparatus and are uncorrected. Microanalyses were obtained
using YANACO MT-3 CHN CORDER. 1H NMR spectra
were measured on a JEOL JNM-GX270 spectrometer using
Me4Si as an internal standard, and J values are reported in Hz.
IR spectra were recorded with a JEOL JIR-7000 FT-IR spec-
trometer and are reported in reciprocal centimeters (cmϪ1).
Optical purity was determined by HPLC analysis with a
TOSOH HLC-8020 or JASCO HPLC system composed of 3-
Line Degasser DG-980-50, HPLC Pump PV-980, and Column
oven CO-965, equipped with a chiral column (Chiralcel OD-H,
Daicel) using hexane–propan-2-ol–diethylamine (90:10:0.1). A
UV detector (TOSOH UV-8011 for TOSOH HPLC system,
JASCO UV-975 for JASCO HPLC system) was used for
chromatographic peak detection. Optical rotations were taken
on a JASCO DIP-140 digital polarimeter using a 10 cm
thermostatted microcell. [α]D-Values are given in units of 10Ϫ1
16b. Yield 90%; [α]D25 Ϫ25.03 (c 1.73 in EtOH); mp 144–
147 ЊC; νmax(KBr) 3516, 3330, 2960, 1597, 1483, 1340, 1095;
δH(CDCl3; 270 MHz) 0.76 (3H, s), 0.89 (3H, s), 1.05 (3H,
s), 1.39–1.68 (5H, m), 2.03 (1H, d, J 4.4), 2.44 (3H, s),
3.24 (1H, t, J 7.3), 3.56 (1H, dd, J 7.3 and 3.9), 5.24 (1H, d,
J 6.3), 7.31 (2H, d, J 8.3), 7.76 (2H, d, J 8.3) (Found: C, 63.15;
H, 7.78; N, 4.33. C17H25NO3S requires C, 63.16; H, 7.74; N,
4.33%).
16c. Yield 75%; [α]D25 ϩ16.1 (c 1.59 in CHCl3); mp 165 ЊC;
νmax(KBr) 3529, 3336, 2956, 1589, 1482, 1344, 1093; δH(CDCl3;
270 MHz) 0.72 (3H, s), 0.86 (3H, s), 1.05 (3H, s), 1.23–1.70
(5H, m), 2.14 (1H, d, J 4.4), 3.26–3.33 (1H, m), 3.53–3.58
(1H, m), 5.42 (1H, d, J 6.3), 7.57–7.97 (7H, m) (Found: C,
66.90; H, 7.00; N, 3.90. C20H25NO3S requires C, 66.85; H,
6.96; N, 3.90%).
17. Yield 79%; [α]D25 ϩ34.4 (c 2.77 in CH2Cl2); mp 134–136 ЊC;
νmax(KBr) 3527, 3342, 2929, 1599, 1495, 1321, 1094; δH(CDCl3;
270 MHz) 0.80 (3H, s), 0.83 (3H, s), 0.85 (3H, s), 1.06–1.74 (5H,
m), 2.03 (1H, d, J 3.9), 2.43 (3H, s), 3.67–3.78 (2H, m), 5.24
(1H, d, J 6.4), 7.30 (2H, d, J 8.3), 7.77 (2H, d, J 8.3) (Found: C,
63.05; H, 7.72; N, 4.17. C17H25NO3S requires C, 63.16; H, 7.74;
N, 4.33%).
19a. Yield 97%; [α]D25 Ϫ27.12 (c 2.81 in CHCl3); mp 98–100 ЊC;
νmax(KBr) 3505, 3280, 1593, 1160; δH(CDCl3; 270 MHz) 1.07
(3H, d, J 6.8), 2.89 (3H, s), 3.18 (1H, d, J 3.9), 3.68–3.80 (1H,
m), 4.82–4.88 (1H, m), 4.97 (1H, d, J 8.8), 7.25–7.34 (5H, m)
deg cm2 gϪ1
.
All N-trimethylsilylimines 8a–8e were obtained by reaction
of the corresponding aldehydes with lithium hexamethyldisil-
2014
J. Chem. Soc., Perkin Trans. 1, 1999, 2011–2016

View Article Online
(Found: C, 52.39; H, 6.58; N, 6.11. C10H15NO3S requires C,
52.38; H, 6.61; N, 6.11%).
solution of (1R,2S)-N-(p-tolylsulfonyl)norephedrine 19b (1.22
g, 4 mmol) and triethylamine (0.05 cm3) was added a THF
solution (15 cm3) of triallylborane (5 mmol) prepared from
BF3ؒOEt2 (0.61 cm3, 5 mmol) and allylmagnesium chloride (15
mmol, 1.2 M) at 0 ЊC. The reaction mixture was stirred at room
temperature for 30 min and then heated under reflux for 1 h to
complete the formation of the oxazaborolidine ring. After cool-
ing to room temperature, the solvent was removed in vacuo and
dry ether (20 cm3) was introduced. At Ϫ78 ЊC an ethereal (2 cm3)
solution of N-silylimine 8a (0.53 g, 3 mmol) was added drop-
wise and the mixture was stirred for 3 h. The reaction was
quenched with 1 M HCl and the organic phase was separated,
from which chiral ligand 19b was recovered (93%). The aqueous
phase was then neutralized with NH4OH and extracted with
ether. The combined ether layer was dried (MgSO4) and evap-
orated. Chromatography (Et2O–hexane, 4:1) gave 1-phenylbut-
3-enamine as a colorless oil. (89%). The enantioselectivity (92%
ee) was determined by HPLC analysis using a chiral stationary-
phase column as shown before.
19c. Yield 96%; [α]D25 ϩ41.97 (c 2.33 in CHCl3); mp 111–
113 ЊC; νmax(KBr) 3527, 3342, 1535, 1417, 1356, 1317;
δH(CDCl3; 270 MHz) 0.97 (3H, d, J 6.8), 2.51 (1H, d, J 4.4),
3.70–3.90 (1H, m), 4.75–4.82 (1H, m), 5.62 (1H, d, J 8.3),
7.21–7.33 (5H, m), 7.69–7.88 (3H, m) (Found: C, 53.54; H,
3.93; N, 8.38. C15H16N2O5S requires C, 53.56; H, 3.90; N,
8.32%).
19d. Yield 94%; [α]D25 Ϫ28.64 (c 2.56 in CHCl3); mp 92 ЊC;
νmax(KBr) 3532, 3317, 1432, 1155, 1093; δH(CDCl3; 270 MHz)
0.74 (3H, d, J 6.8), 2.60 (1H, d, J 3.9), 3.47–3.59 (1H, m),
4.62–4.68 (1H, m), 5.2 (1H, d, J 8.8), 7.12–7.27 (5H, m),
7.52–7.69 (3H, m), 7.93 (1H, d, J 8.3), 8.07 (1H, d, J 8.3),
8.32 (1H, d, J 7.3), 8.62 (1H, d, J 8.8) (Found: C, 66.91; H,
5.58; N, 4.13. C19H19NO3S requires C, 66.84; H, 5.62; N,
4.10%).
19e. Yield 97%; [α]D25 ϩ14.95 (c 2.48 in CH2Cl2); mp 82 ЊC;
νmax(KBr) 3523, 3315, 1433, 1155, 1092; δH(CDCl3; 270 MHz)
0.85 (3H, d, J 6.8), 2.71 (1H, br s), 3.61–3.72 (1H, m), 4.76–
4.84 (1H, m), 5.11 (1H, d, J 8.8), 7.18–7.30 (5H, m), 7.57–
7.67 (2H, m), 7.82–7.96 (4H, m), 8.47 (1H, s) (Found: C,
66.80; H, 5.71; N, 4.11. C19H19NO3S: C, 66.84; H, 5.62; N,
4.10%).
Acknowledgements
This study was supported by a Grant-in-Aid for Scientific
Research (no. 09650955, 17650907) from the Ministry of
Education, Science, Sports and Culture, Japan (Monbusho).
20. Yield 96%; [α]D25 ϩ15.93 (c 2.48 in CHCl3); mp 93–95 ЊC;
νmax(KBr) 3504, 3257, 1591, 1457, 1324, 1160; δH(CDCl3; 270
MHz) 0.82 (3H, d, J 6.8), 2.41 (3H, s), 2.86 (1H, br s), 3.52–3.58
(1H, m), 4.79 (1H, d, J 3.4), 5.16 (1H, d, J 8.8), 7.21–7.33 (7H,
m), 7.77 (2H, d, J 8.3) (Found: C, 62.88; H, 6.21; N, 4.58.
C16H19NO3S requires C, 62.92; H, 6.28; N, 4.59%).
References
1 Reviews on allyl organometallics: (a) Y. Yamamoto and N. Asao,
Chem. Rev., 1993, 93, 2207; (b) W. R. Roush, in Comprehensive
Organic Synthesis, ed. B. M. Trost, I. Fleming and C. F.
Heathcock, Pergamon, Oxford, 1991, vol. 2, ch. 1.1. For leading
references on asymmetric allylboration of aldehydes: (c) R. W.
Hoffmann and T. Herold, Chem. Ber., 1981, 114, 375; (d) T.
Herold, U. Schrott and R. W. Hoffmann, Chem. Ber., 1981, 114,
359; (e) H. C. Brown and P. K. Jadhav, J. Org. Chem., 1984,
49, 4089; ( f ) U. S. Racherla and H. C. Brown, J. Org. Chem.,
1991, 56, 401; (g) W. R. Roush, A. E. Walts and L. K. Hoong,
J. Am. Chem. Soc., 1985, 107, 8186; (h) W. R. Roush, L. K. Hoong,
M. A. J. Palmer and J. C. Park, J. Org. Chem., 1990, 55, 4109; (i)
M. T. Reetz and T. Zierke, Chem. Ind., 1988, 663; (j) R. P. Short
and S. Masamune, J. Am. Chem. Soc., 1989, 111, 1892; (k) E. J.
Corey, C.-M. Yu and S. S. Kim, J. Am. Chem. Soc., 1989, 111,
5495.
21. Yield 87%; [α]D25 ϩ75.3 (c 2.03 in THF); mp 224–228 ЊC;
νmax(KBr) 3470, 3320, 1310, 1150; δH(CDCl3; 270 MHz) 2.29
(1H, d, J 4.9), 2.34 (3H, s), 4.55 (1H, dd, J 7.8 and 4.9), 4.99
(1H, t, J 4.6), 5.20 (1H, d, J 7.8), 6.80–7.22 (12H, m), 7.48 (2H,
d, J 8.3) (Found: C, 68.70; H, 5.75; N, 3.80. C21H21NO3S
requires C, 68.66; H, 5.72; N, 3.81%).
22. Yield 79%; [α]D25 Ϫ4.66 (c 2.13 in CHCl3); mp 102 ЊC;
νmax(KBr) 3400, 3240, 1710, 1370, 1170; δH(CDCl3; 270 MHz)
1.26 (3H, d, J 6.4), 2.17 (1H, br s), 2.42 (3H, s), 3.52 (3H, s),
3.82 (1H, dd, J 9.8 and 2.9), 4.13 (1H, br s), 5.48 (1H, d, J 9.3),
7.29 (2H, d, J 8.8), 7.73 (2H, d, J 8.3) (Found: C, 50.15; H, 5.95;
N, 4.85. C12H17NO5S: C, 50.17; H, 5.92; N, 4.88%).
2 H. C. Brown and P. V. Ramachandran, Pure Appl. Chem., 1991, 63,
General procedure for enantioselective allylboration of 10 to
N-silylimines
307.
3 H. Moser, G. Rihs and H. Santer, Z. Naturforsch., Teil B, 1982, 37,
451; J. K. Whitesell, Chem. Rev., 1989, 89, 1581; J. Seyden-Penne,
Chiral Auxiliaries and Ligands in Asymmetric Synthesis, Wiley,
New York, 1995.
4 E. F. Kleinman and R. A. Volkmann, in Comprehensive Organic
Synthesis, ed. B. M. Trost, I. Fleming and C. H. Heathcock,
Pergamon, Oxford, 1991, vol. 2, ch. 4.3.
The transformation of 8a to 9a is typical. To a solution of 10
(12 mmol), prepared from (Ϫ)-DIP-Chloride (3.8 g, 12 mmol)
and allylmagnesium chloride (12 mmol), was added dropwise a
THF (5 cm3) solution of 8a (1.42 g, 8 mmol) at Ϫ78 ЊC. The
reaction mixture was then stirred for 3 h at Ϫ78 ЊC and
quenched with 1 M HCl. The aqueous phase was separated,
and washed with ether. The aqueous phase was then neutralized
with NH4OH and extracted with ether. The combined ether
layer was dried over MgSO4 and evaporated under reduced
pressure. The residual product was purified by flash chrom-
atography (Et2O–hexane, 4:1) to yield 9a as a colorless liquid
(0.76 g, 65%), δH(CDCl3; 270 MHz) 7.34–7.22 (5H, m), 5.82–
5.68 (1H, m), 5.15–5.06 (2H, m), 3.98 (1H, dd, J 7.8 and 5.4),
2.46–2.32 (2H, m), 1.53 (2H, br s). The enantioselectivity of
64% ee was determined by HPLC analysis using a chiral
stationary-phase column (Daicel, Chiralcel OD-H; hexane–
propan-2-ol–diethylamine 90:10:0.1, flow rate 0.5 cm3
minϪ1); tR = 16.3 min (R), tR = 20.7 min (S). The absolute
configuration of the product was correlated to that described
in the literature.19
5 For a review on the asymmetric addition of nucleophiles to the
C᎐N double bond, see: E. Denmark and O. J.-C. Nicaise, Chem.
᎐
Commun., 1996, 999; D. Enders and U. Reinhold, Tetrahedron:
Asymmetry, 1997, 8, 1895.
6 H. Nakamura, K. Nakamura and Y. Yamamoto, J. Am. Chem. Soc.,
1998, 120, 4242.
7 M. Nakamura, A. Hirai and E. Nakamura, J. Am. Chem. Soc., 1996,
118, 8489.
8 I. Chataigner, F. Zammattio, J. Lebreton and J. Villiéras, Synlett,
1998, 275.
9 K. Watanabe, K. Ito and S. Itsuno, Tetrahedron: Asymmetry,
1995, 6, 1531; S. Itsuno, K. Watanabe, K. Ito, A. A. El-Shehawy and
A. A. Sarhan, Angew. Chem., Int. Ed. Engl., 1997, 36, 109.
10 K. Watanabe, S. Kuroda, A. Yokoi, K. Ito and S. Itsuno,
J. Organomet. Chem., 1999, 581, 103.
11 J. Sisko and S. Weinreb, J. Org. Chem., 1990, 55, 393.
12 G. Andersson, D. Guijarro and D. Tanner, J. Org. Chem., 1997, 62,
7364.
13 K. Soai, T. Hatanaka and T. Miyazawa, J. Chem. Soc., Chem.
Commun., 1992, 1097.
14 H. C. Brown and P. K. Jadhav, J. Am. Chem. Soc., 1983, 105, 2092;
P. K. Jadhav, K. S. Bhat, T. Perumal and H. C. Brown, J. Org.
Chem., 1986, 51, 432.
General procedure for enantioselective allylboration of
N-silylimines with chirally modified allylboron reagents
The transformation of 8a to 9a is typical. To a THF (5 cm3)
J. Chem. Soc., Perkin Trans. 1, 1999, 2011–2016
2015

View Article Online
15 G.-M. Chen, P. V. Ramachandran and H. C. Brown, Angew. Chem.,
Int. Ed., 1999, 38, 825.
16 W. R. Roush, A. D. Palkowitz and K. Ando, J. Am. Chem. Soc.,
1990, 112, 6348; W. R. Roush, K. Ando, D. B. Powers, A. D.
Palkowitz and R. H. Halterman, J. Am. Chem. Soc., 1990, 112, 6339.
17 G. Cainelli, D. Giacomini, P. Galletti and A. Gaiba, Synlett, 1996,
657.
18 D. J. Hart, K. Kanai, D. G. Thomas and T. K. Yang, J. Org. Chem.,
1983, 48, 289.
19 T. Basile, A. Bocoum, D. Savoia and A. Umani-Ronchi, J. Org.
Chem., 1994, 59, 7766.
Paper 9/02635E
2016
J. Chem. Soc., Perkin Trans. 1, 1999, 2011–2016