The synthesis of an enantiopure planar-chiral Lewis acid complex via kinetic resolution and its application in stereoselective additions to imines

Article abstract of DOI:10.1016/j.tet.2006.06.048

This report describes the synthesis and an application of a new planar-chiral Lewis acid based on a 1,2-azaborolyl framework. The?enantiopure complex is generated through an intriguing kinetic resolution by a chiral nucleophile. Imines bind in excellent yield to the planar-chiral 1,2-azaborolyl, furnishing crystallographically characterizable adducts that adopt the conformation that had been anticipated on the basis of steric considerations. Nucleophiles add with high stereoselectivity to these complexes, with the predicted sense of asymmetric induction.

Full text of DOI:10.1016/j.tet.2006.06.048

Tetrahedron 62 (2006) 11343–11349
The synthesis of an enantiopure planar-chiral Lewis acid complex
via kinetic resolution and its application in stereoselective
additions to imines
y
*
Shih-Yuan Liu, Michael M.-C. Lo and Gregory C. Fu
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA
Received 6 May 2006; revised 13 June 2006; accepted 14 June 2006
Available online 8 August 2006
Abstract—This report describes the synthesis and an application of a new planar-chiral Lewis acid based on a 1,2-azaborolyl framework.
The enantiopure complex is generated through an intriguing kinetic resolution by a chiral nucleophile. Imines bind in excellent yield to
the planar-chiral 1,2-azaborolyl, furnishing crystallographically characterizable adducts that adopt the conformation that had been anticipated
on the basis of steric considerations. Nucleophiles add with high stereoselectivity to these complexes, with the predicted sense of asymmetric
induction.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
As part of an earlier investigation, we described the first
application of such planar-chiral Lewis acids in asymmetric
synthesis, specifically, Mukaiyama aldol reactions mediated
by an (h5-1,2-azaborolyl)iron complex (Eq. 2).3b,4–6 In these
processes, a p interaction between the Lewis acid and the
aldehyde organizes and activates the adduct for a stereo-
selective nucleophilic addition.
Chiral Lewis acids have become widely used tools in asym-
metric synthesis.1 For most of the designs that have been
described, the Lewis-acidic atom is not stereogenic, although
there are a number of noteworthy exceptions to this generali-
zation.2 Designs in which the Lewis-acidic site is stereogenic
may benefit from the close proximity of the asymmetric en-
vironment to the site of reaction; on the other hand, the issue
of stereochemical lability may arise (e.g., Eq. 1).
t-Bu
N
B
OTs
1)
R1
LA
R1
LA
R1
LA
Fe
OC
O
OH
Ar
OSiR3
3
O
CO
R2
R2
TMS
ð1Þ
R
R1X
R3
R2
R1X
R3
ent-1
R3
H
Ar
2) hydrolysis
R
R R
high
achiral 1
1
X = O, S
ð2Þ
stereoselectivity
Several years ago, we initiated a program directed towards
the synthesis of planar-chiral Lewis acids (for an illustrative
example based on a 1,2-azaborolyl framework, see 2).3 This
design circumvents the aforementioned complication of
configurational lability at the Lewis-acidic site (Eq. 1).
t-Bu
N
B
O
H
Ar
OTs
Fe
OC
TMS
CO
1,2-azaborolyl and aldehyde are co-planar;
nucleophile adds to the top face
R
N
B
MLn
2
In this report, we add a new dimension to the use of these pla-
nar-chiral Lewis acids in asymmetric synthesis. In particular,
we demonstrate their effectiveness in mediating stereoselec-
tive additions to imines through a substrate–Lewis acid
adduct that is conformationally distinct from reactions of
aldehydes (Eq. 3 vs Eq. 2).
* Corresponding author. Tel.: +1 617 253 2664; fax: +1 617 324 3611;
e-mail: gcf@mit.edu
Correspondence concerning X-ray crystallography should be directed to
y
M. M.-C. Lo.
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.06.048

11344
S.-Y. Liu et al. / Tetrahedron 62 (2006) 11343–11349
R
N
t-Bu
N
R1
R
Nu
R1
1) Nu
B
HN
R2
high
stereoselectivity
OTf
R2
2) hydrolysis
(3)
Fe
1,2-azaborolyl and imine are not co-planar;
nucleophile adds to the front face
2. Results and discussion
In our study of the Mukaiyama aldol reaction, we gener-
ated enantiopure Lewis acid complex 3 by resolving
a precursor via preparative chiral HPLC. In parallel
with that investigation, we were exploring the chemistry
of 1,2-azaborolyl analogues of ferrocene (e.g., 4 in Eq.
4).4a Rather than relying on chromatographic resolution
of these adducts, we decided to examine the possibility
that they might be separated via kinetic resolution by
an enantiopure Lewis base. Specifically, we hypothesized
that a planar-chiral DMAP derivative (e.g., 57) might be
effective.
Figure 1. ORTEP illustration, with thermal ellipsoids drawn at the 35%
probability level, of complex 6. For the sake of clarity, the triflate counterion
has been omitted.
by steric interactions,10 leading to preferential population
of the ‘perpendicular’ geometry (Eq. 5).
N
t-Bu
R
t-Bu
N
t-Bu
N
R1
N
N
Me
R2
R1
B
OTf
Fe
N
Me
B
N
R
B
R2
Fe
ð5Þ
Me
Me
MLn
MLn
Me
azaborolyl and imine
are co-planar
azaborolyl and imine
are not co-planar
("perpendicular" geometry)
(+)-4
(+)-5
0.55 equiv
CH2Cl2
–78 °C to r.t.
s >40
ð4Þ
Reaction of (ꢀ)-4 with imines furnishes isolable azaborolyl–
imine complexes in excellent yield (Eq. 6). An X-ray
crystallographic study supports our suggestion that
the perpendicular conformation should be preferred
(Fig. 2).
t-Bu
N
Cp*Fe
N
t-Bu
N
B
OTf
N
B
OTf
Fe
Fe
(–)-4
>95% ee
40% yield
6
45% yield
We were pleased to determine that this strategy is indeed vi-
able. Thus, DMAP derivative (+)-5 reacts with high selectiv-
ity for (+)-4 to generate complex 6 (selectivity factor >40),8
thereby allowing unreacted (ꢀ)-4 to be isolated in excellent
ee (Eq. 4).9 The structure of 6 has been confirmed by X-ray
crystallography (Fig. 1), which nicely illustrates how the
planar chirality of the Lewis acid and the Lewis base impact
upon the atropisomer preference of the adduct (FeCp* and
t-Bu groups avoid each other). To the best of our knowledge,
this is the first 1,2-azaborolyl complex that bears a neutral
ligand on boron.
In the case of azaborolyl–aldehyde complexes, we had antic-
ipated that a co-planar geometry would be preferred (Eq. 2).
On the other hand, we predicted that, for azaborolyl–imine
adducts, the co-planar conformation would be destabilized
Figure 2. ORTEP illustration, with thermal ellipsoids drawn at the 35%
probability level, of an azaborolyl-(2-methyl-1-pyrroline) complex. For
the sake of clarity, the triflate counterion has been omitted.

S.-Y. Liu et al. / Tetrahedron 62 (2006) 11343–11349
Table 1. Stereoselective additions to planar-chiral azaborolyl–imine complexes
11345
Nu
R
N
R
t-Bu
N
t-Bu
N
R1
R1
R
R3N
Nu
R1
B–N
B
N
B
Nu
cleavage
OTf
R2
R2
Fe
Fe
R2
8
7
Entry
1
Imine
Nu
7 Yield (%); de (%)a
96; >90
8
Me
Me
N
94% eeb
86% eec
MgBr
F3CO2S
N
Me
Me
Me
HN
OMe
OMe
OMe
MgBr
2
N
N
83; 87
69; 86
OMe
Me
HN
H
OMe
OMe
OMe
OMe
3
LiBHEt3
87% eec
Determined by 1H NMR.
Determined by chiral GC.
Determined by chiral HPLC.
a
b
c
yield to the planar-chiral 1,2-azaborolyl, furnishing crystal-
lographically characterizable adducts that adopt the confor-
mation that had been anticipated on the basis of steric
considerations. Nucleophiles add with high stereoselectivity
to these complexes, with the predicted sense of asymmetric
induction. This study thus adds a new dimension to the use of
planar-chiral azaborolyls in asymmetric synthesis, comple-
menting our earlier investigation of additions to aldehydes,
which proceed through an orthogonal transition-state geo-
metry, also with good stereoselectivity. Efforts to develop
catalyzed transformations are underway.
t-Bu
N
R
N
t-Bu
N
R1
R
N
B
OTf
R1
B
OTf
Fe
R2
Fe
R2
(–)-4
imine
Me
yield
92%
ð6Þ
N
Me
OMe
OMe
N
3. Experimental
98%
All oxygen- and moisture-sensitive manipulations were
carried out under an inert atmosphere using either standard
Schlenk techniques or a glove box.
Just as the co-planar geometry of a planar-chiral azaborolyl–
aldehyde complex provides an effective asymmetric envi-
ronment for stereoselective addition reactions (Eq. 2), we
anticipated that the perpendicular orientation of an azaboro-
lyl–imine adduct would also furnish good selectivity, due to
blockage of one face by the bulky tert-butyl substituent
(Fig. 2).11 Gratifyingly, this expectation has been fulfilled:
additions of nucleophiles to azaborolyl–imine complexes do
indeed proceed with high levels of stereoinduction (Table 1).
For example, reaction with allylmagnesium bromide affords
a quaternary stereocenter in good yield and ꢁ86% ee (entries
1 and 2).12 Furthermore, addition of LiBHEt3 occurs with
high stereoselection to produce 1R-(+)-salsolidine,13 a natu-
rally occurring tetrahydroisoquinoline alkaloid (entry 3).
The sense of induction is consistent with our prediction.
1,2-Azaborolyl complex 44a and DMAP derivative (+)-514
were prepared according to literature procedures. THF,
Et2O, CH2Cl2, and toluene were purified by passage through
a neutral alumina column under argon. All other chemicals
and solvents were purchased from Aldrich or Strem and
used as received.
Silica gel (230–400 mesh; SiliCycle) was heated under vac-
uum in a 200 ꢂC sand bath for 12 h. Flash chromatography
was performed with this pre-treated silica gel under an inert
atmosphere.
11B NMR spectra were recorded on a Varian Unity 300 or on
a Varian Unity 500 spectrometer at ambient temperature. 31P
NMR spectra and 19F NMR spectra were recorded on a
Varian Mercury 300 spectrometer with complete proton
decoupling at ambient temperature. All chemical shifts
In conclusion, we have described a synthesis and an applica-
tion of a new planar-chiral Lewis acid. The complex is gen-
erated in enantiopure form through an intriguing kinetic
resolution by a chiral nucleophile. Imines bind in excellent

11346
S.-Y. Liu et al. / Tetrahedron 62 (2006) 11343–11349
are referenced to an external standard: 19F NMR to trifluoro-
acetic anhydride (d ꢀ78), 31P NMR to 85% H3PO4 (d 0), and
11B NMR to boron trifluoride diethyl etherate (d 0).
the resulting mixture was filtered through an acrodisc
(washed with pentane) and concentrated under vacuum.
The desired product was isolated by flash chromatography
(pentane / 50% pentane/CH2Cl2), which furnished the
desired product as a red oil (25.3 mg, 91%). The H NMR
resonances of the two diastereomers of this 1:1 mixture
are well distinguished from one another.
1
Kinetic resolution of ( )-4 by (D)-5 (Eq. 4). A solution of
(+)-5 (263 mg, 0.699 mmol) in CH2Cl2 (7.0 mL) was added
dropwise to a ꢀ78 ꢂC solution of (ꢃ)-4 (497 mg,
1.27 mmol) in CH2Cl2 (45 mL). The reaction mixture was
kept at ꢀ78 ꢂC for 2 h, and then it was allowed to warm to
rt over 7 h. The CH2Cl2 was then removed under vacuum un-
til a volume ofw6 mL was reached, at which time pentane
(14 mL) was carefully layered on top of the reaction mixture
in order to induce crystallization. After 48 h, dark purple
crystals of complex 6 had formed, leaving an orange mother
liquor. The solid was washed with Et2O (w20 mL) and dried
under vacuum (483 mg, 45%). Crystals suitable for X-ray
analysis were grown from a CH2Cl2/pentane solution.
1H NMR (500 MHz, CD2Cl2): d 8.08 (s, 1H), 7.98 (app. t,
3
3JHH¼8.5 Hz, 2H), 7.91 (d, JHH¼7.5 Hz, 1H), 7.84–7.80
3
(m, 5H), 7.56–7.43 (m, 5H), 5.41 (q, JHH¼6.5 Hz, 1H),
3
5.29 (q, JHH¼6.0 Hz, 1H), 5.13 (s, 1H), 5.08 (s, 1H), 4.19
(s, 5H), 4.07–4.03 (m, 2H), 3.74 (s, 5H), 2.57–2.51 (m,
2H), 1.89 (d, JHH¼6.5 Hz, 3H), 1.59 (d, JHH¼7.0 Hz,
3
3
3H), 1.35 (s, 9H), 1.33 (s, 9H).
13C NMR (125 MHz, CD2Cl2): d 144.7, 144.5, 134.0, 133.9,
133.4, 133.1, 128.34, 128.33, 128.32, 128.23, 128.20, 128.1,
126.5, 126.4, 126.0, 125.9, 125.5, 125.1, 124.8, 124.1, 76.0,
75.5, 69.0, 68.9, 68.6, 68.5, 68.4, 68.3, 54.4, 54.3, 41.4 (br,
2C), 30.0, 29.9, 26.0, 25.9.
Compound 6 (Eq. 4). 1H NMR (500 MHz, CD2Cl2): d 9.08
3
3
(d, JHH¼7.0 Hz, 1H), 6.18 (d, JHH¼7.0 Hz, 1H), 5.47 (s,
3
1H), 4.71 (d, JHH¼4.0 Hz, 1H), 4.46 (s, 5H), 4.16 (d,
3JHH¼2.5 Hz, 1H), 4.02 (t, JHH¼2.5 Hz, 1H), 3.98–3.78
3
3
(m, 3H), 3.74–3.69 (m, 2H), 3.70 (d, JHH¼5.0 Hz, 1H),
11B NMR (160 MHz, CD2Cl2): d 17.1.
2.19–2.31 (m, 4H), 1.70 (s, 15H), 0.88 (s, 9H). Only one
diastereomer was detected.
FTIR (thin film) 3056, 2972, 2927, 1464, 1437, 1424, 1399,
1385, 1366, 1329, 1269, 1238, 1213, 1085 cmꢀ1
.
13C NMR (125 MHz, CD2Cl2): d 163.6, 150.3, 107.6, 95.8,
81.2, 77.4, 73.3, 72.4, 70.4, 69.2, 67.0, 66.1, 59.3 (br), 56.3,
52.0, 51.7, 30.7, 26.7, 25.1, 10.1.
HRMS (EI) calcd for C24H28BNFeO (M+) 413.1608, found
413.1614.
11B NMR (160 MHz, CD2Cl2): d 13.9.
Determination of the ee of (L)-4 (Eq. 4). The ee was deter-
mined through H NMR analysis of its (S)-(ꢀ)-a-methyl-
1
FTIR (thin film) 3113, 2978, 2909, 1576, 1551, 1512, 1457,
1411, 1370, 1321, 1273, 1239, 1150, 1031 cmꢀ1
2-naphthalenemethanol derivative, which was prepared
according to the general procedure: >95% ee.
.
HRMS (ESI) calcd for C34H45BN3Fe2 (M+) 618.2400, found
618.2371.
1H NMR (500 MHz, CD2Cl2): d 7.83–7.80 (m, 4H), 7.55–
3
7.53 (m, 1H), 7.47–7.42 (m, 2H), 5.41 (q, JHH¼6.5 Hz,
3
1H), 5.11 (s, 1H), 4.18 (s, 5H), 4.05 (d, JHH¼4.0 Hz, 1H),
3
3
[a]D ꢀ60 (c 0.050, CH2Cl2; >95% de).
Mp 135–145 ꢂC (decomposition).
2.54 (d, JHH¼4.5 Hz, 1H), 1.87 (d, JHH¼6.5 Hz, 3H),
1.33 (s, 9H).
13C NMR (125 MHz, CD2Cl2): d 144.7, 133.9, 133.1, 128.3,
128.2, 128.1, 126.4, 125.9, 124.8, 124.1, 75.5, 68.9, 68.6,
68.5, 54.4, 41.4 (br), 30.0, 25.9.
Compound (L)-4 (Eq. 4). The orange mother liquor was fil-
tered through an acrodisc and then concentrated to dryness.
This material was purified by column chromatography
(pentane / 50% pentane/CH2Cl2), which furnished (ꢀ)-4 as
an orange oil (199 mg, 40%).
11B NMR (96 MHz, CD2Cl2): d 17.1.
FTIR (thin film) 3056, 2973, 1424, 1400, 1386, 1366, 1329,
1268, 1239, 1085 cmꢀ1
.
The spectral data for (ꢀ)-4 are identical to those reported for
(ꢃ)-4.4a
HRMS (EI) calcd for C24H28BNFeO (M+) 413.1608, found
413.1625.
[a]D ꢀ500 (c 0.50, CH2Cl2; >95% ee).
General procedure for derivatization to determine the ee
of (h5-1,2-azaborolyl)iron complexes (e.g., 4). In a glove
box, a vial was charged with racemic (h5-1-tert-butyl-2-
chloro-1,2-azaborolyl)(h5-cyclopentadienyl)iron4a (18.6 mg,
0.0670 mmol), (S)-(ꢀ)-a-methyl-2-naphthalenemethanol
(13.8 mg, 0.0800 mmol), and sodium hydride (4.8 mg,
0.20 mmol). THF (0.6 mL) was added, and the reaction mix-
ture was stirred at rt for 22 h. As the reaction proceeded, the
color of the mixture gradually changed from orange to red.
At the conclusion of the reaction, pentane was added, and
[a]D ꢀ500 (c 0.10, CH2Cl2; >95% de).
Recovery of (D)-5 (Ref. 9). In a glove box, a vial was
charged with a solution of (ꢀ)-6 (64 mg, 0.083 mmol) in
CH2Cl2 (0.65 mL). TBAF (1.0 M solution in THF; 92 mL,
0.092 mmol) was added, and then the progress of the reac-
tion was monitored by 11B NMR. Complete conversion
was achieved within 3 h, during which time the reaction
mixture changed from blue-purple to burgundy (the color of
(+)-5). The reaction mixture was concentrated under vacuum

S.-Y. Liu et al. / Tetrahedron 62 (2006) 11343–11349
11347
3
and then purified by silica gel chromatography (ethyl
acetate / ethyl acetate/Et3N). The first compound to
elute was (+)-(h5-1-tert-butyl-2-fluoro-1,2-azaborolyl)(h5-
cyclopentadienyl)iron:4a orange solid, 17 mg (78%), [a]D
+160 (c 0.31, CH2Cl2; >95% ee), mp 47–49 ꢂC. The
second, burgundy-colored fraction contained (+)-5, which
was contaminated with a tetrabutylammonium-containing
impurity. Pure (+)-5 was obtained after aqueous extraction
(water/Et2O) and flash chromatography (ethyl acetate /
ethyl acetate/Et3N) (29.5 mg, 94%).
3H), 3.38 (d, JHH¼4.5 Hz, 1H), 3.31–3.26 (m, 2H), 2.58
(s, 3H), 1.30 (s, 9H).
13C NMR (125 MHz, CD2Cl2): d 181.4, 157.9, 149.4, 134.5,
120.4, 112.4, 111.3, 72.9, 72.6, 70.9, 57.8 (br), 57.3, 57.0,
56.5, 55.6, 31.0, 26.7, 24.1.
11B NMR (160 MHz, CD2Cl2): d 14.0.
FTIR (thin film) 3063, 2980, 2943, 2841, 1605, 1550, 1523,
1467, 1430, 1387, 1345, 1265, 1223, 1153, 1068,
Determination of the ee of (D)-(h5-1-tert-butyl-2-fluoro-
1,2-azaborolyl)(h5-cyclopentadienyl)iron (Ref. 9). The
ee was determined through H NMR analysis of its (S)-
1031 cmꢀ1
.
HRMS (ESI) calcd for C24H32BN2O2Fe (M+) 447.1901,
found 447.1916.
1
(ꢀ)-a-methyl-2-naphthalenemethanol derivative, which
was prepared according to the general procedure: >95% ee.
[a]D ꢀ450 (c 0.020, CH2Cl2; >95% ee).
Mp 120–125 ꢂC (decomposition).
Eq. 6 (2-methyl-1-pyrroline complex). In a glove box,
a vial was charged with a solution of (ꢀ)-4 (40.0 mg,
0.102 mmol) in CH2Cl2 (1.2 mL). 2-Methyl-1-pyrroline
(15 mL, 0.153 mmol) was added via a syringe. The reaction
mixture was allowed to sit at rt for 7 h. Pentane was carefully
layered on top of the reaction mixture to induce crystalliza-
tion. The desired complex crystallized as an orange solid
(44.7 mg, 92%). Crystals suitable for X-ray analysis were
grown from a CH2Cl2/hexanes solution.
Table 1, entry 1. In a glove box, a vial was charged with a
solution of the 2-methyl-1-pyrroline complex (39.3 mg,
0.0829 mmol) in CH2Cl2 (2.1 mL). This solution was cooled
to ꢀ35 ꢂC, and then allylmagnesium bromide (1.0 M solu-
tion in Et2O; 91 mL, 0.091 mmol) was added dropwise.
The reaction mixture was allowed to stand at ꢀ35 ꢂC, with
occasional shaking, for 2 h, and then it was warmed to rt.
Pentane was added, the reaction mixture was filtered through
an acrodisc, and the filtrate was concentrated under vacuum.
The resulting residue was then purified by flash chromato-
graphy (pentane/Et2O/Et3N), which furnished the desired
product as an orange oil (29.0 mg, 96%). 1H NMR analysis
indicated that the de of this mixture was >90%.
1H NMR (500 MHz, CD2Cl2): d 5.41 (s, 1H), 5.08–5.02 (m,
1H), 4.92–4.87 (m, 1H), 4.61 (d, 3JHH¼4.0 Hz, 1H), 4.40 (s,
3
5H), 3.51 (d, JHH¼3.5 Hz, 1H), 3.44–3.39 (m, 2H), 2.56–
2.51 (m, 1H), 2.42–2.32 (m, 1H), 2.34 (s, 3H), 1.23 (s, 9H).
13C NMR (125 MHz, CD2Cl2): d 201.8, 73.3, 73.1, 70.6,
67.5, 59.1 (br), 56.4, 42.3, 30.8, 21.1, 20.9.
1H NMR (500 MHz, C6D6): d 6.15–6.08 (m, 1H), 5.22–5.14
11B NMR (160 MHz, CD2Cl2): d 10.8.
(m, 2H), 4.82 (s, 1H), 4.13 (s, 5H), 4.10 (d, JHH¼4.0 Hz,
3
1H), 3.74–3.69 (m, 1H), 3.63–3.58 (m, 1H), 3.19 (d,
3JHH¼4.0 Hz, 1H), 2.81–2.77 (m, 1H), 2.68–2.64 (m, 1H),
2.09–2.04 (m, 1H), 1.85–1.74 (m, 2H), 1.50–1.44 (m, 1H),
1.18 (s, 3H), 1.10 (s, 9H).
FTIR (thin film) 3094, 2975, 1641, 1374, 1265, 1224, 1156,
1031 cmꢀ1
.
HRMS (ESI) calcd for C17H26BN2Fe (M+) 325.1533, found
325.1524.
13C NMR (125 MHz, C6D6): d 138.6, 116.6, 71.0, 70.9, 69.0,
63.8, 58.2 (br), 55.3, 54.9, 45.5, 38.7, 30.8, 29.0, 25.3.
[a]D ꢀ230 (c 0.10, CH2Cl2; >95% ee).
Mp 185–187 ꢂC.
11B NMR (160 MHz, C6D6): d 14.3.
FTIR (thin film) 3071, 2960, 2868, 1636, 1431, 1382, 1361,
1210, 1152, 1128, 1106, 1081 cmꢀ1
.
Eq. 6 (6,7-dimethoxy-1-methyl-3,4-dihydroisoquinoline
complex). In a glove box, a vial was charged with a solution
of (ꢀ)-4 (50.0 mg, 0.128 mmol) in CD2Cl2 (0.9 mL) and
HRMS (EI) calcd for C20H31BN2Fe (M+) 366.1924, found
366.1933.
then
6,7-dimethoxy-1-methyl-3,4-dihydroisoquinoline15
(52.4 mg, 0.255 mmol). The reaction mixture was stirred
at rt for 2 h and then at 70 ꢂC for 19 h, after which time
1H NMR indicated that the reaction was complete. The reac-
tion mixture was cooled to rt, and then Et2O was carefully
layered on top of the reaction mixture in order to induce
crystallization. The resulting solid was recrystallized from
Et2O/CH2Cl2 (79.6 mg, 103% yield, 95% purity according
to 1H NMR).
[a]D ꢀ510 (c 0.10, CH2Cl2; >90% de).
Table 1, entry 2. In a glove box, a vial was charged with
a solution of the 6,7-dimethoxy-1-methyl-3,4-dihydroiso-
quinoline adduct (40.0 mg, 0.0670 mmol) in CH2Cl2
(1.6 mL). This solution was cooled to ꢀ35 ꢂC, and allyl-
magnesium bromide (1.0 M solution in Et2O; 79 mL,
0.079 mmol) was added dropwise. The reaction mixture
was allowed to stand at ꢀ35 ꢂC, with occasional shaking,
for 3 h, and then it was concentrated under vacuum. The
resulting residue was purified by silica gel chromatography
1H NMR (500 MHz, CD2Cl2): d 7.23 (s, 1H), 6.98 (s, 1H),
5.50 (s, 1H), 5.12–5.06 (m, 1H), 4.98–4.90 (m, 1H), 4.65
3
(d, JHH¼3.5 Hz, 1H), 4.43 (s, 5H), 4.03 (s, 3H), 3.91 (s,

11348
S.-Y. Liu et al. / Tetrahedron 62 (2006) 11343–11349
(pentane/Et2O/Et3N). Repurification of this material by
flash chromatography (pentane/Et2O) furnished the desired
product as an orange foam (27.1 mg, 83%). H NMR anal-
a solution of compound 7 in THF (0.6 mL). The course of
the reaction was followed by 11B NMR. After the starting
material had been consumed, the reaction mixture was con-
centrated under vacuum, and the residue was purified by
flash chromatography (CH2Cl2/Et3N).
1
ysis indicated that the de of this mixture was 87%.
1H NMR (500 MHz, CD2Cl2): d 6.68 (s, 1H), 6.59 (s, 1H),
5.99–5.90 (m, 1H), 5.23 (s, 1H), 5.01–4.94 (m, 2H), 4.24
Table 1, entry 1: B–N cleavage and assay of ee. The
general procedure was followed. The 2-allyl-2-methylpyrro-
lidine was derivatized with trifluoromethanesulfonyl anhy-
dride according to a procedure in lit. 16.
3
(d, JHH¼5.0 Hz, 1H), 4.20 (s, 5H), 4.13–4.03 (m, 2H),
3
3.81 (s, 3H), 3.76 (s, 3H), 3.17 (d, JHH¼5.0 Hz, 1H),
3.11–3.07 (m, 1H), 2.94–2.88 (m, 1H), 2.73–2.69 (m, 2H),
1.29 (app. s, 12H).
The enantiomeric excess of the sulfonamide was determined
to be 94% by GC analysis (Chiraldex b-PH; 20 mꢄ0.25 mm,
105 ꢂC oven temperature, helium as carrier gas; (S) isomer
(minor) 41.9 min, (R) isomer (major) 43.2 min). The configu-
rational assignment (R) was made by comparison with the
optical rotation reported in the literature: [a]D +28 (c 0.36,
CH2Cl2); lit.16 (R) enantiomer, [a]D +29 (c 2.0, CH2Cl2;
100% ee).
13C NMR (125 MHz, CD2Cl2): d 147.4, 146.8, 138.6, 136.9,
128.5, 116.0, 112.0, 111.4, 70.8, 70.1, 69.1, 58.5 (br), 58.5,
56.5, 56.1, 55.4, 48.9, 47.0, 30.8, 30.6, 29.8.
11B NMR (160 MHz, CD2Cl2): d 16.8.
FTIR (thin film) 3070, 2973, 2932, 2908, 2831, 1635, 1610,
1514, 1464, 1362, 1251, 1209, 1189, 1157, 1105 cmꢀ1
.
Table 1, entry 2: B–N cleavage and assay of ee. The
general procedure was followed.
HRMS (EI) calcd for C27H37BN2O2Fe (M+) 488.2292,
found 488.2283.
1H NMR (500 MHz, CD2Cl2): d 6.65 (s, 1H), 6.53 (s, 1H),
5.71–5.63 (m, 1H), 5.10–5.03 (m, 2H), 3.79 (s, 3H), 3.78
(s, 3H), 3.08–2.97 (m, 2H), 2.66–2.60 (m, 3H), 2.36–2.31
(m, 1H), 1.34 (s, 3H).
[a]D ꢀ340 (c 0.10, CH2Cl2; 87% de).
Table 1, entry 3. In a glove box, a vial was charged with a so-
lution of the 6,7-dimethoxy-1-methyl-3,4-dihydroisoquino-
line adduct (18.4 mg, 0.0310 mmol) in CH2Cl2 (0.75 mL).
This solution was cooled to ꢀ35 ꢂC, and lithium triethyl-
borohydride (1.0 M solution in THF; 37 mL, 0.037 mmol)
was added dropwise. The reaction mixture was allowed to
stand at ꢀ35 ꢂC, with occasional shaking, for 2 h, and then
it was concentrated under vacuum. The resulting residue was
purified by flash chromatography (pentane/Et2O/Et3N),
which furnished the desired product as an orange oil
13C NMR (125 MHz, CD2Cl2): d 147.5, 147.4, 135.4, 135.2,
127.7, 118.0, 112.0, 109.7, 56.2, 55.8, 55.0, 47.1, 39.1, 30.4,
29.3.
FTIR (thin film) 3312, 3072, 2931, 2832, 1637, 1610, 1514,
1465, 1401, 1366, 1354, 1325, 1259, 1225, 1153 cmꢀ1
.
HRMS (ESI) calcd for C15H22NO2 (M++H) 248.1645, found
248.1652.
1
(9.6 mg, 69%). H NMR analysis indicated that the de of
this mixture is 86%.
[a]D ꢀ53 (c 0.63, THF; 86% ee).
1H NMR (500 MHz, CD2Cl2): d 6.63 (s, 1H), 6.60 (s, 1H),
3
5.19 (s, 1H), 4.33 (q, JHH¼6.0 Hz, 1H), 4.18 (s, 6H), 3.81
The enantiomeric excess was determined to be 86% by
HPLC analysis (Daicel Chiralpak OD; hexane:i-PrOH¼
97:3, 1.0 mL/min; (R) isomer (minor) 13.8 min, (S) isomer
(major) 15.5 min). The configurational assignment (S) was
made by comparison with the optical rotation reported in
the literature: [a]D ꢀ53 (c 0.63, THF); lit.17 (R) enantiomer,
[a]D +48 (c 2.1, THF; 76% ee).
(s, 3H), 3.80 (s, 3H), 3.56–3.50 (m, 1H), 3.52–3.31 (m,
1H), 3.11–3.05 (m, 1H), 2.84 (d, JHH¼4.5 Hz, 1H),
3
3
2.63–2.58 (m, 1H), 1.49 (d, JHH¼6.5 Hz, 3H), 1.28
(s, 9H).
13C NMR (125 MHz, CD2Cl2): d 147.2, 147.1, 134.8, 127.9,
112.3, 110.4, 70.2, 69.5, 68.6, 56.4 (br), 56.2, 56.0, 55.9,
54.6, 44.9, 30.2, 29.4, 22.9.
Table 1, entry 3: B–N cleavage and assay of ee: (R)-salso-
lidine. The general procedure was followed.
11B NMR (160 MHz, CD2Cl2): d 16.7.
1H NMR (500 MHz, C6D6): d 6.49 (s, 1H), 6.38 (s, 1H), 3.97
(q, 3JHH¼7.0 Hz, 1H), 3.44 (s, 3H), 3.42 (s, 3H), 3.07–3.03
(m, 1H), 2.84–2.79 (m, 1H), 2.71–2.65 (m, 1H), 2.50–2.45
(m, 1H), 1.39 (d, 3JHH¼6.5 Hz, 3H).
FTIR (thin film) 3052, 2963, 2928, 2830, 1610, 1515,
1464, 1393, 1360, 1330, 1246, 1230, 1209, 1118,
1105 cmꢀ1
.
HRMS (EI) calcd for C24H33BN2O2Fe (M+) 448.1979,
found 448.1996.
The enantiomeric excess was determined to be 87%
by HPLC analysis (Daicel Chiralpak OD; hexane:i-PrOH¼
80:20, 0.5 mL/min; (S) isomer (minor) 18.2 min, (R) isomer
(major) 21.4 min). The configurational assignment (R) was
made by comparison with the optical rotation reported in
the literature: [a]D +39 (c 0.070, EtOH); lit.18 (R) enantio-
mer, [a]D +59 (c 2, EtOH; 100% ee).
[a]D ꢀ480 (c 0.10, CH2Cl2; 86% de).
General procedure for B–N cleavage (Table 1). In a glove
box, TBAF (1.0 M solution in THF; 4 equiv) was added to

S.-Y. Liu et al. / Tetrahedron 62 (2006) 11343–11349
11349
7. For leading references, see: Fu, G. C. Acc. Chem. Res. 2004, 37,
542–547.
Acknowledgements
8. Selectivity factor (s)¼(rate of fast-reacting enantiomer/rate of
slow-reacting enantiomer). For a classic review of kinetic
resolutions, see: Kagan, H. B.; Fiaud, J. C. Top. Stereochem.
1988, 18, 249–330.
9. By reacting 6 with TBAF, enantiopure DMAP derivative
(+)-5 can be recovered in 94% yield (along with the
B-fluorinated (h5-1,2-azaborolyl)iron complex: 78% yield
and >95% ee).
Support has been provided by Merck Research Laboratories
and Novartis. Funding for the MIT Department of Chemistry
Instrumentation Facility has been furnished in part by the
National Science Foundation (CHE–9808061 and DBI–
9729592).
References and notes
1. For leading references, see: (a) Comprehensive Asymmetric
Catalysis; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.;
Springer: Heidelberg, 1999; (b) Lewis Acids in Organic
Synthesis; Yamamoto, H., Ed.; Wiley: New York, NY, 2000.
2. For some examples and leading references, see: (a) Gladysz,
J. A.; Boone, B. J. Angew. Chem., Int. Ed. 1997, 36,
551–583; (b) Faller, J. W.; Grimmond, B. J.; D’Alliessi,
D. G. J. Am. Chem. Soc. 2001, 123, 2525–2529.
3. (a) For an overview, see: Fu, G. C. J. Org. Chem. 2004, 69,
3245–3249; (b) Liu, S.-Y.; Hills, I. D.; Fu, G. C. J. Am.
Chem. Soc. 2005, 127, 15352–15353; (c) Amendola, M. C.;
Stockman, K. E.; Hoic, D. A.; Davis, W. M.; Fu, G. C.
Angew. Chem., Int. Ed. 1997, 36, 267–269; (d) Tweddell, J.;
Hoic, D. A.; Fu, G. C. J. Org. Chem. 1997, 62, 8286–8287.
4. For additional investigations of 1,2-azaborolyl chemistry from
our laboratory, see: (a) Liu, S.-Y.; Lo, M.-C.; Fu, G. C. Angew.
Chem., Int. Ed. 2002, 41, 174–176; (b) Liu, S.-Y.; Hills, I. D.;
Fu, G. C. Organometallics 2002, 21, 4323–4325.
10. For a review of peri interactions in naphthalene derivatives,
see: Balasubramaniyan, V. Chem. Rev. 1966, 66, 567–641.
11. For leading references to asymmetric additions to imines, see:
(a) Vilaivan, T.; Bhanthumnavin, W.; Sritana-Anant, Y. Curr.
Org. Chem. 2005, 9, 1315–1392; (b) Kobayashi, S.; Ishitani,
H. Chem. Rev. 1999, 99, 1069–1094.
12. For examples of highly enantioselective allylations of ketone-
derived benzoylhydrazones with a stoichiometric enantiopure
allylsilane reagent, see: Berger, R.; Duff, K.; Leighton, J. L.
J. Am. Chem. Soc. 2004, 126, 5686–5687.
13. For a review of synthetic pathways to salsolidine, see:
Kaufman, T. S. Tetrahedron: Asymmetry 2004, 15, 1203–1237.
14. Ruble, J. C.; Fu, G. C. J. Am. Chem. Soc. 1998, 120, 11532–
11533.
15. This was prepared in analogy to a procedure reported for
6,7-methylenedioxy-3,4-dihydroisoquinoline: Slemon, C. E.;
Hellwig, L. C.; Ruder, J.-P.; Hoskins, E. W.; MacLean, D. B.
Can. J. Chem. 1981, 59, 3055–3060.
5. For leading references to pioneering studies of 1,2-azaborolyl
chemistry, see: (a) Schmid, G. Comprehensive Heterocyclic
Chemistry II; Shinkai, I., Ed.; Elsevier: Oxford, 1996; Vol. 3,
Chapter 3.17; (b) Schmid, G. Comments Inorg. Chem. 1985,
4, 17–32.
16. Burgess, L. E.; Meyers, A. I. J. Am. Chem. Soc. 1991, 113,
9858–9859.
17. Meyers, A. I.; Gonzalez, M. A.; Struzka, V.; Akahane, A.;
Guiles, J.; Warmus, J. S. Tetrahedron Lett. 1991, 32, 5501–
5504.
6. For a recent report on 1,2-azaborolyl chemistry, see: Fang, X.;
Yang, H.; Kampf, J. W.; Holl, M. M. B.; Ashe, A. J., III.
Organometallics 2006, 25, 513–518.
18. Menachery, M. D.; Lavanier, G. L.; Wetherly, M. L.;
Guinaudeau, H.; Shamma, M. J. Nat. Prod. 1986, 49,
745–778.