Efficient kinetic resolution in the asymmetric hydrosilylation of imines of 3-substituted indanones and 4-substituted tetralones

Article abstract of DOI:10.1021/jo991328h

Kinetic resolution of the N-methyl imines of 3-substituted indanones and 4-substituted tetralones could be accomplished by hydrosilylation with a chiral titanocene catalyst. N-Methyl imines of 4-substituted tetralones were resolved to yield, after hydrolysis of the unreacted starting materials, ketones with high ee's and the amine products with high diastereomeric and enantiomeric purity. The utility of this process was demonstrated in the synthesis of sertraline.

Full text of DOI:10.1021/jo991328h

J . Org. Chem. 2000, 65, 767-774
767
Efficien t Kin etic Resolu tion in th e Asym m etr ic Hyd r osilyla tion of
Im in es of 3-Su bstitu ted In d a n on es a n d 4-Su bstitu ted Tetr a lon es
J aesook Yun and Stephen. L. Buchwald*
Department of Chemistry, Massachusettes Institute of Technology, Cambridge, Massachusetts 02139
Received August 20, 1999
Kinetic resolution of the N-methyl imines of 3-substituted indanones and 4-substituted tetralones
could be accomplished by hydrosilylation with a chiral titanocene catalyst. N-Methyl imines of
4-substituted tetralones were resolved to yield, after hydrolysis of the unreacted starting materials,
ketones with high ee’s and the amine products with high diastereomeric and enantiomeric purity.
The utility of this process was demonstrated in the synthesis of sertraline.
In tr od u ction
our work on the application of asymmetric imine hy-
drosilylation to the kinetic resolution of racemic acyclic
imines9 of 3-substituted indanones and 4-substituted
tetralones, including the enantio- and diastereoselective
preparation of sertraline, an important antidepressant
sold under the trade name Zoloft.
A large number of bioactive and pharmaceutically
interesting molecules contain an indan ring framework1
or a tetrahydronaphthalene core.2 Indanones or tetral-
ones possessing chiral centers are also useful molecules
as starting material for the synthesis of biologically active
compounds. These products are of great current interest
in the pharmaceutical industry, and therefore, their
derivatives have been extensively evaluated for their
biological activities.3 There are a number of approaches
to introduce chirality to the indanone and tetralone ring
systems; the most general method is the preparation of
chiral aryl carboxylic acids either by resolution or asym-
metric synthesis followed by Friedel-Crafts cyclization
onto an aromatic ring.4 Kinetic resolution of racemic
substrates has also been employed.5,6
Resu lts
Based on the stereochemical model of the transition
state for the reduction of imines with the (EBTHI)-
titanocene catalyst,10a we reasoned that imines could be
successfully kinetically resolved.10b,11 The two important
predictions of this model are that the substituent on the
nitrogen atom has the greatest influence on the stereo-
chemical outcome of the reaction and the syn and the
anti isomers of an imine react to give opposite enanti-
omers of the product. It seemed necessary to have a
substituent on the nitrogen atom big enough to direct
the facial selectivity but not so large as to slow the
reduction. Therefore, the N-propyl imines of 3-substituted
indanones were chosen as the initial substrates. These
were subjected to the previously described hydrosilylation
conditions using phenylsilane as the stoichiometric re-
ductant (Table 1). The reactions proceeded smoothly at
room temperature; however, it was necessary to use
2.5-5 mol % catalyst and long reaction times (1.5-3 d)
in order to reach conversions that gave high ee of the
ketone derived by hydrolysis of the recovered starting
material. Substrates 2a and 2b were transformed with
similar selectivity12 to give the corresponding ketones13
in >90% ee at ca. 60% conversion. The resolutions of 2c
(R ) n-butyl) and 2d (R ) tert-butyl) were less efficient.
We have recently reported an efficient procedure to
reduce ketones7 and imines8 utilizing a chiral (EBTHI)-
titanocene catalyst, (R,R)-(EBTHI)TiF2 (1) (EBTHI )
ethylenebis(tetrahydroindenyl)). Herein we wish to report
(1) Clark, W. M.; Tickner-Eldridge, A. M.; Huang, G. K.; Pridgen,
L. N.; Olsen, M. A.; Mills, R. J .; Lantos, I.; Baine, N. H. J . Am. Chem.
Soc. 1998, 120, 4550.
(2) (a) Williams, M.; Quallich, G. Chem. Ind. 1990, 315. (b) Kamal,
A.; Gayatri, N. L. Tetrahedron Lett. 1996, 37, 3359.
(3) (a) Welch, W. M.; Kraska, A. R.; Sarges, R.; Koe, B. K. J . Med.
Chem. 1984, 27, 1508. (b) Lowe, J . A.; Hageman, D. L.; Drozda, S. E.;
McLean, S.; Bryce, D. K.; Crawford, R. T.; Zorn, S.; Morrone, J .;
Bordner, J . J . Med. Chem. 1994, 37, 3789. (c) Miyano, S.; Tatsuoka,
T.; Suzuki, K.; Imao, K.; Satoh, F.; Ishihara, T.; Hirotsu, I.; Kihara,
T.; Hatta, M.; Horikawa, Y.; Sumoto, K. Chem. Pharm. Bull. 1990,
38, 1570.
(4) (a) Poras, H.; Stephan, E.; Pourcelot, G.; Cresson, P. Chem. Ind.
1993, 206.
(5) For enzymatic kinetic resolutions of 2-substituted indanones or
tetralones, see: (a) Adam, W.; D´ıaz, M. T.; Fell, R. T.; Saha-Mo¨ller, C.
R. Tetrahedron: Asymmetry 1996, 7, 2207. (b) Kajiro, H.; Mitamura,
S.; Mori, A.; Hiyama, T. Tetrahedron: Asymmetry 1998, 9, 907. For
nonenzymatic kinetic resolutions of 2-substituted tetralones, see: (c)
Geneˆt, J . P.; Pfister, X.; Ratovelomanana-Vidal, V.; Pinel, C.; Laffitte,
J . A. Tetrahedron Lett. 1994, 35, 4559. (d) Kabuto, K.; Ziffer, H. J .
Org. Chem. 1975, 40, 3467. (e) Schmalz, H.-G.; J ope, H. Tetrahedron
1998, 54, 3457. (f) Sugi, K. D.; Nagata, T.; Yamada, T.; Mukaiyama,
T. Chem. Lett. 1996, 1081.
(6) For enzymatic kintetic resolution of 4-substituted tetralones,
see: Be´gue´, J .-P.; Cerceau, C.; Dogbeavou, A.; Mathe´, L.; Sicsic, S. J .
Chem. Soc., Perkin Trans. 1 1992, 3141.
(7) (a) Carter, M. B.; Schiøtt, B.; Gutie´rrez, A.; Buchwald, S. L. J .
Am. Chem. Soc. 1994, 116, 11667. (b) Yun, J .; Buchwald, S. L. J . Am.
Chem. Soc. 1999, 121, 5640.
(9) For an example of kinetic resolution of acylic imines, see:
Lensink, C.; de Vries, J . G. Tetrahedron: Asymmetry 1993, 4, 215.
(10) (a) Willoughby, C. A.; Buchwald, S. L. J . Am. Chem. Soc. 1994,
116, 11703. (b) We began this project by investigating the kinetic
resolution of 3-methylindanone using 1 under hydrosilylation condi-
tons; however, the results were not promising. The recovered ketone
was essentially racemic and the diastereomeric alcohol products were
produced in a ∼1:1 ratio.
(11) For examples of highly efficient kinetic resolution processes
using (EBTHI)Zr derivatives, see: Hoveyda, A. H.; Morken, J . P.
Angew. Chem., Int. Ed. Engl. 1996, 35, 1262.
(12) The selectivity factor, s, was calculated by the equation
published in: Martin, V. S.; Woodard, S. S.; Katsuki, T.; Yamada, Y.;
Ikeda, M.; Sharpless, K. B. J . Am. Chem. Soc. 1981, 103, 6237.
(13) The starting acyclic imines were unstable and could not be
isolated by the usual separation procedure. Therefore, they were
isolated as the corresponding ketones after hydrolysis with HOAc/
NaOAc buffer solution or silica gel chromatography.
(8) (a) Verdaguer, X.; Lange, U. E. W.; Reding, M. T.; Buchwald, S.
L. J . Am. Chem. Soc. 1996, 118, 6784. (b) Verdaguer, X.; Lange, U. E.
W.; Buchwald, S. L. Angew. Chem., Int. Ed. Engl. 1998, 37, 1103.
10.1021/jo991328h CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/14/2000

768 J . Org. Chem., Vol. 65, No. 3, 2000
Ta ble 1. Kin etic Resolu tion of N-P r op yl Im in es of 3-Su bstitu ted In d a n on es
Yun and Buchwald
recovered ketone
(% ee)
amine product
cis (% ee)/trans (% ee)
entry
substrate
cat. (mol %)
time (h)
convna (%)
sb
1
2
3
4
5
6
7
2a c
2a c,e
2b
2.5
2.5
5
5
5
72
36
48
45
36
42
48
64.5
49
60
58
54
94 (R)
66 (R)
90
84
74
10.3
10.5
11.7
10.4
9.5
81 (80)d:19
90 (82):10
85 (74):15 (74)
86 (77.5):14 (70)
87.5 (75):12.5 (53)
85 (78):15 (64)
96 (47):4
2be
2c
2c
2d
5
5
61
60
86
71
9.1
5.7
a
Conversion was determined by GC with an internal standard based on consumption of starting imine substrate. See the Experimental
Section for isolated yields of recovered ketone and amine product. Selectivity factor.12 c [Substrate] ) 1.0 M. The other runs used
b
d
[substrate] ) 0.5 M. Absolute stereochemistry was determined by X-ray analysis of the corresponding (+)-camphorsulfonic acid salt
and is as shown above. e Toluene was used as solvent.
Ta ble 2. Kin etic Resolu tion of N-Meth yl Im in es of 3-Su bstitu ted In d a n on es
diastereomeric
recovered ketone
(% ee)
ratio of amine
products (% ee)
entry
substrate
T (°C)
time (h)
convna (%)
s
1
2
3
4
5
6
5a
15b
1.5
2
20
24
2
57
55
51
44
51
52
92 (R)
93 (R)
86 (R)
68 (R)
83
17.5
24.5
28.8
28.1
22.7
16.0
93.5 (77):6.5 (42)
95 (83):5 (41)
97 (88):3 (30)
98 (94):2 (16)
94.5 (89):5.5 (64)
94.5 (84.5):5.5
0b
-20
-30c
0
5b
0d
2
80
a
Conversion was determined by GC with an internal standard based on consumption of starting imine substrate. See the Experimental
b
Section for isolated yields of recovered ketone and amine products. Reactions were carried out in [substrate] ) 1 M. The other runs used
[substrate] ) 0.5 M. c 2.5 mol % (R,R)-1 and 2 equiv of PhSiH3 were used. Toluene was used as solvent.
d
Next, we examined N-methyl imines of 3-substituted
indanones as substrates in the hope that the smaller
nitrogen substituent would give higher turnover numbers
without affecting the selectivity of the reaction. In fact,
the imine hydrosilylation of substrate 5a proceeded
quickly at room temperature.14 The reaction was thus
carried out at a lower temperature employing a smaller
quantity of catalyst (1 mol %) and phenylsilane (1 equiv)
than used in the previous case (Table 2). When imine
substrates 5a and 5b were subjected to the hydrosilyla-
tion conditions, the ketones derived from the unreacted
imines at ∼50% conversion had a high ee. Lowering the
reaction temperature increased the s value for the process
at the expense of the reaction rate. The reaction pro-
ceeded very slowly at temperatures lower than -30 °C.15
Using phenylsilane as the silane source is important for
reasonable turnover rate; the use of diphenylsilane
significantly slowed the reaction.16
The resolution of N-methyl imines of 4-substituted
tetralones was investigated with the same catalyst
system (Table 3). Better results were achieved with
tetralone substrates than with the indanone ones; the s
values ranged from ∼18 to 78 at room temperature. In
addition, the amine products are formed with high
diastereoselectivity and enantioselectivity. Considering
the diastereomeric ratio of amine products in the NaBH4
reduction of these imines (Table 3, for 7a and 7b), the
(15) With 5 mol % catalyst and 2 equiv of PhSiH3 at -45 °C there
was no product formation after 15 h. When the reaction was warmed
to -30 °C, it gave 19% conversion in 12 h.
(16) With diphenylsilane (3 equiv) as the silane source (2 mol %
catalyst) we observed 8% conversion at room temperature in 24 h.
(14) With 2 mol % catalyst and 3 equiv PhSiH3, the reaction went
to completion at room temperature in 3 h. The distereomeric ratio of
the amine products was 77:23 and the ee’s of the products were 27%
and 91% respectively.

Asymmetric Hydrosilylation of Imines
J . Org. Chem., Vol. 65, No. 3, 2000 769
Ta ble 3. Kin etic Resolu tion of N-Meth yl Im in es of 4-Su bstitu ted Tetr a lon es
diasteromeric
ratio of amine
products (% ee)
recovered ketone
(% ee)
entry
substrate
Ta (°C)
time (h)
convnb (%)
s
1
2
3
4
5
6
7a
rt
13
rt
13
rt
24
24
15
15
36
48
54
54
61.5
53
49
96 (R)
99 (R)
98
99
89
38.5
60.8
17.9
80.1
78.3
95 (90):5 (99)
96 (93):4 (99)
89 (88):11
95.5 (96):4.5
98 (97):2 (99)
98.5 (98):1.5 (99)
7b
7c
13
44
75.5
114
a
b
rt ) ∼24 °C; 13 °C was the temperature in the glovebox. Determined by GC with an internal standard based on consumption of
starting material. Isolated yields were calculated on the basis of conversion and reported in the Experimental Section. All the reactions
were carried out in [substrate] ) 0.5 M with the exception of entries 1 and 2; [substrate] ) 1 M (1 mmol/1 mL THF) and 0.8 M (2
mmol/2.5 mL THF) were used.
Ta ble 4. Syn th esis of Ser tr a lin e via Kin etic Resolu tion
diastereoselectivity obtained in this kinetic resolution is
of r a c-10
impressive.17
Since the amine products are produced with high de’s
and high ee’s in the kinetic resolution of N-methyl imines
of tetralones, this method could be also used for asym-
metric synthesis of chiral amine products. Sertraline ((+)-
cis-(1S,4S)-1-methylamino-4-(3,4-dichlorophenyl)tetra-
lin) (11), an antidepressant sold by Pfizer under the trade
name Zoloft, is a competitive inhibitor of synaptosomal
serotonin uptake. It is produced commercially by the
resolution of the racemate with D-mandelic acid.2a Even
though a few synthetic routes to the sertraline penulti-
cat.
PhSiH3 T (°C)/ conversion
ee (isolated
yield)a
ketone
(% ee)
mate tetralone (4S)-1218 and sertraline 1119 have been
reported, a classical resolution method is still the pre-
ferred industrial means of its production. In view of the
structural resemblance of sertraline precursor 10 to 7b,
we decided to synthesize sertraline using our kinetic
resolution method (Table 4).
(mol %) (equiv) time (h)
(%)
5
4
2
2
rt, 24
54.5
dr 94.5:5.5
ee 91%
96
(48%)b
15, 13
51
dr 96:4
ee 95%
(39%)
dr 95.5:4.5
ee 94% (38%)
dr 96.5:3.5
ee 97% (40%)
89
Exposing 1020 to a catalytic quantity of (S,S)-1 using
PhSiH3 as the stoichiometric reductant led to the forma-
tion of (1S,4S)-11 in a highly diastereo- and enantiose-
lective manner. During the course of the reaction, the
active catalytic species was slowly deactivated by the
substrate 10,21 necessitating the use of 2.5-4 mol % of
the catalyst to obtain useful conversions (∼45%) in e24
h. Overall, sertraline could be isolated in high ee in ∼40%
average yield.
4c
3
3
rt, 5
47
44
83.5
71
2.5c
rt, 24
a
Pure cis compound was isolated by silica gel chromatography,
b
and this yield represents that of the cis product. This sample
showed a 96:4 cis and trans ratio by GC and 1H NMR. c (R,R)-1
was used. The absolute stereochemistry of this reaction is opposite
to that shown in the diagram.
1H NMR. X-ray analysis of imine 1022 showed that the
major isomer is E (anti) where the methyl group points
away from the phenyl ring. Since the absolute stereo-
chemistry of several of the reaction products (both
recovered ketone and the major amine product) have been
determined, we can comment on the origins of selectivity
of the kinetic resolution process (Figure 1). Of the four
possible ways for the racemic substrate to approach
(R,R)-(EBTHI)Ti-H, it appears that path A is the most
favorable. In this case, the N-methyl group is positioned
in an empty quadrant and the R group is oriented away
from the bulk of the catalyst. This orientation should
Discu ssion
All the imine substrates used in this work were
obtained exclusively as one isomer (>95%) as judged by
(17) It has been reported that similar diastereomeric mixtures of
racemic amine products are often difficult to resolve.3a
(18) (a) Corey, E. J .; Gant, T. G. Tetrahedron Lett. 1994, 35, 5373.
(b) Quallich, G. J .; Woodall, T. M. Tetrahedron 1992, 48, 10239.
(19) (a) Lautens, M.; Rovis, T. J . Org. Chem. 1997, 62, 5246. (b)
Chen, C.-Y.; Reamer, R. A. Org. Lett. 1999, 1, 293.
(20) Quallich, G. J .; Williams, M. T.; Friedmann, R. C. J . Org. Chem.
1990, 55, 4971.
(21) An experiment with equal amounts of 7b (0.5 mmol) and 10
(0.5 mmol) using 2 mol % catalyst and phenylsilane (2 equiv) was
carried out. In 4 h, both substrates reached ∼50% conversion, and the
conversion had not increased after an additional 11 h (7b is one of the
fast reacting substrate in Table 3).
(22) While the resolution of this crystal structure was not high, it
was sufficient to show that the methyl group is in an anti position.

770 J . Org. Chem., Vol. 65, No. 3, 2000
Yun and Buchwald
F igu r e 1.
minimize steric interactions with the ligand framework
to yield the major amine product with the observed
absolute and relative configuration. The high ee’s and de’s
of the amine products at ∼50% conversion in the kinetic
resolution of N-methyl imines of 4-substituted tetralones
reflect the fact that path A predominates over the other
pathways.
Unlike in cases of simple kinetic resolution where one
enantiomer reacts faster than the other enantiomer with
no creation of a new chiral center, the mathematical
relationships correlating the relative amounts and the
enantiomeric excesses of reaction products and recovered
starting material are complex in instances in which
kinetic resolution is combined with the creation of an
asymmetric center. Horeau and Guette´ described a
relationship that correlates the ee’s of the two diastere-
omeric products and the diastereomeric ratio in cases
where complete conversion occurred.23 This equation
shows that the diastereomeric ratio is equal to the inverse
ratio of the ee’s of the diastereomers. Our data are in
good agreement with that predicted by their equation.
At complete conversion the values are as follows: 77/23
) 3.35 for the diastereomeric ratio and 91/27 ) 3.37 for
the inverse ratio of the ee’s.14 In cases of partial conver-
sion, the mathematical equations dealing with analogous
relationships have been discussed by Kagan.24 These
equations are useful to calculate or to refine data not
easily measured and to check the self-consistency of a
set of data coming from a kinetic resolution. We have
applied these equations to the data from the experiments
described in Table 2 for which we could obtain all the
information25 necessary to check the consistency. The
theoretical value of the ee of recovered starting material
is in good agreement with the experimental value.26 In
the case of the data in Table 3, we have not determined
the absolute configuration of the minor amine product,
which was produced as a single enantiomer. However,
applying Kagan’s equation27 with both possible sign
conventions of the ee of the minor amine product gave
quite different calculated diastereomeric ratios of the
(23) Guette´, J .-P.; Horeau, A. Bull. Soc. Chim. Fr. 1967, 1747.

Asymmetric Hydrosilylation of Imines
J . Org. Chem., Vol. 65, No. 3, 2000 771
addition30 of n-butylcuprate to ethyl trans-cinnamate in the
presence of TMSCl and HMPA, followed by hydrolysis of the
ester (NaOH), transformation to the acid chloride (oxalyl
chloride, catalytic DMF, benzene), and Friedel-Crafts cycliza-
tion of the acid chloride with AlCl3 in refluxing benzene. 3-tert-
Butyl-1-indanone was prepared by the Friedel-Crafts cycliza-
tion of â-tert-butylhydrocinnamic acid.31 4-Phenyl-1-tetralone
was prepared from γ-phenyl-γ-butyrolactone using triflic acid
following a reported procedure.20 4-Cyclopentyl-1-tetralone was
prepared by following the literature procedure3b starting from
commercially available R-phenylcyclopentaneacetic acid in-
stead of R-phenylcyclohexylacetic acid, and the Friedel-Crafts
amine products. The calculated diastereomeric ratio is
very similar to the observed ratio only when the absolute
stereochemistry of the minor amine product is assumed
to be the one (e.g., 1R,4R for 9a ) obtained by path B
(Figure 1).
We can also draw useful information about the dias-
tereoselectivity of the process creating a new asymmetric
center in each enantiomer.28 For example, with the data
for 5a (Table 2, entry 2), the estimated diastereomeric
product ratio ([A]/[A′]) of the matched imine enantiomer
is 59:1 and that of the mismatched one is 2.3:1([B′]/[B])
(Figure 1). For the tetralone substrate (Table 3, entry
1), the estimated diastereomeric product ratio of the
matched imine enantiomer is >104:1 and that of the
mismatched one is 0.95:1. This means that if we had
enantiomerically pure 4-substituted tetralone, we could
transform it to the corresponding cis-N-methylamine
product with very high diastereoselectivity (>99% de).
Combining previously reported asymmetric syntheses of
tetralone (4S)-12,18 the imine hydrosilylation reaction
using the (EBTHI)titanocene catalyst could provide ser-
traline very efficiently.
In summary, we have achieved the efficient kinetic
resolution of imine derivatives of indanones and tetral-
ones having a substituent at a position remote from the
reaction center. The amine product was formed with high
diastereoselectivity using 1 under hydrosilylation condi-
tions. We have also applied this methodology to the
enantiomeric synthesis of sertraline.
32
cyclization was carried out with ClSO3H in CH2Cl2 instead
of the original reaction conditions utilizing (CF3CO)2O in
refluxing benzene. 4-(3,4-Dichlorophenyl)-1-tetralone was pre-
pared by the literature procedure.20
(3-Meth yl-in d a n -1-ylid en e)p r op yla m in e (2a ): Gen er a l
P r ocedu r e for th e P r epar ation of N-P r opylim in es. 3-Meth-
yl-1-indanone (1.80 g, 12.4 mmol) was placed in a dry Schlenk
flask under argon. Anhydrous ether (60 mL) was added, and
the reaction flask was cooled to -35 °C. N-Propylamine (5.0
mL, 62 mmol) was introduced via syringe, followed by the
addition of TiCl4 (0.75 mL, 6.6 mmol). The reaction mixture
was allowed to warm to room temperature slowly and stirred
overnight. The reaction mixture was filtered through a pad of
Celite, thoroughly washed with ether, and concentrated. The
product was purified by Kugelrohr distillation to give the title
product as a yellow oil in 95% yield. It was stored in a N2-
filled glovebox.33 In cases where the imine product was a solid,
the crude product was dried under vacuum and washed under
Ar with a small amount of cold pentane or hexane to remove
colored impurities: 1H NMR (300 MHz, CDCl3) 7.78 (d, J )
7.69 Hz, 1H), 7.42-7.25 (m, 3H), 3.41 (t, J ) 7.19 Hz, 2H),
3.44-3.33 (m, 1H), 2.99 (ddd, J ) 0.64, 8.15, 17.86 Hz, 1H),
2.23 (ddd, J ) 1.12, 4.05, 17.84 Hz, 1H), 1.76 (sext, J ) 7.30
Hz, 2H), 1.34 (d, J ) 7.05 Hz, 3H), 0.99 (t, J ) 7.37 Hz, 3H);
13C NMR (75 MHz, CDCl3) 172.84, 154.33, 139.38, 131.15,
127.20, 124.44, 122.17, 56.01, 37.55, 35.28, 24.34, 22.05, 12.47;
IR (neat) 1656 cm-1; HRMS (EI) m/z calcd for C13H17N1
187.1361, found 187.1365.
Exp er im en ta l Section
Gen er a l Meth od s. THF and diethyl ether were distilled
under argon from sodium benzophenone ketyl. Toluene was
distilled under nitrogen from molten sodium. Phenylsilane and
pyrrolidine were stored under argon in Schlenk tubes and were
manipulated under an argon atmosphere.
Ketones were prepared according to known literature
procedures except for 3-methyl-1-indanone,29 3-phenyl-1-in-
danone, and 4-methyl-1-tetralone, which were available from
Aldrich. 3-n-Butyl-1-indanone was prepared by conjugation
(3-P h en ylin d a n -1-ylid en e)p r op yla m in e (2b). Using the
general procedure, the title compound was prepared as a white
1
solid (the isolated yield was not determined): mp 70 °C; H
NMR (300 MHz, CDCl3) 7.88 (dd, J ) 2.53, 6.68 Hz, 1H), 7.39-
7.20 (m, 5H), 7.14-7.11 (m, 3H), 4.50 (dd, J ) 4.32, 8.89 Hz,
1H), 3.43 (t, J ) 7.17 Hz, 2H), 3.30 (dd, J ) 8.64, 18.10 Hz,
1H), 2.67 (dd, J ) 4.40, 18.10 Hz, 1H), 1.78 (qt, J ) 7.17, 7.23
Hz, 2H), 0.99 (t, J ) 7.23 Hz, 3H); 13C NMR (75 MHz, CDCl3)
172.54, 152.43, 145.41, 139.97, 131.54, 128.95, 127.85, 127.72,
126.85, 126.12, 122.26, 56.04, 47.02, 39.37, 24.27, 12.41; IR
(neat) 1656 cm-1. Anal. Calcd for C18H19N1: C, 86.70; H, 7.68.
Found: C, 86.47; H, 7.65.
(3-n -Bu tylin d a n -1-ylid en e)p r op yla m in e (2c). Using the
general procedure, the title compound was prepared as a
yellow oil in 92% yield: 1H NMR (500 MHz, CDCl3) 7.79 (d, J
) 7.63 Hz, 1H), 7.40-7.26 (m, 4H), 3.43 (t, J ) 7.17 Hz, 2H),
3.29-3.25 (m, 1H), 2.90 (dd, J ) 8.09, 17.85 Hz, 1H), 2.34 (dd,
J ) 3.81, 17.85 Hz, 1H), 1.87-1.84 (m, 1H), 1.81-1.73 (m, 2H),
1.47-1.32 (m, 5H), 1.00 (t, J ) 7.32 Hz, 3H), 0.92 (t, J ) 7.17
Hz, 3H); 13C NMR (75 MHz, CDCl3) 173.05, 153.45, 139.84,
(24) (a) El-Baba, S.; Poulin, J .-C.; Kagan, H. B. Tetrahedron 1984,
40, 4275. (b) Kagan, H. B.; Fiaud, J . C. In Topics in Stereochemistry;
Eliel, E. L., Wilen, S. H., Ed.; J ohn Wiley & Sons: New York, 1988;
Vol. 18; p 249.
X2/X3 ) x ) (C(Y1 - Y3) - Y1)/(C(Y2 - Y1) + Y1)
Y1 ) C(xY2 + Y3)/((C - 1)(1 + x))
(1)
(2)
(3)
(4)
(5)
(6)
Y2 ) ((C - 1)/C) × (1 + 1/x)Y1 - Y3/x
Y3 ) ((C - 1)/C) × (1 + x)Y1 - xY2
a ) x(1 + Y2)/(1 + Y3) ) [RS′]/[RR′]
b ) x(1 - Y2)/(1 - Y3) ) [SR′]/[SS′]
Y1: ee of R,S after kinetic resolution ) (R - S)/(R + S) > 0. Y2 and Y3:
ee of diastereomers. Y2 ) (RS′ - SR′)/(RS′ + SR′), Y3 ) (RR′ - SS′)/
(RR′ + SS′). X1, X2, and X3 represent the fractional amounts after
partial conversion of 1 mol of a compound. C: fractional conversion of
one mole of the initial mixture. The asymmetric centers created are
labeled R′ or S′. The stereochemical assignment of the equations in
the original paper has been modified to that of Figure 1 (R ) Me).
(25) The ee’s of the minor trans-amine products were determined
for 6a and 6b. For the imines of indanones 5a and 5b, the higher the
conversion, the lower the ee of the cis-amine and the higher the ee of
the trans-amine. From this information, the absolute configuration of
the minor amine product of 6a was determined as 1R,3R; see
Figure 1.
(29) One-step synthesis from crotonic acid and AlCl3 in benzene,
see: Koelsch, C. F.; Hochmann, H.; Le Claire, C. D. J . Am. Chem. Soc.
1943, 65, 59.
(30) Yamamoto, K.; Ogura, H.; J ukuta, J .; Inoue, H.; Hamada, K.;
Sugiyama, Y.; Yamada, S. J . Org. Chem. 1998, 63, 4449.
(31) Wotiz, J . H.; Matthews, J . S.; Greenfield, H. J . Am. Chem. Soc.
1953, 75, 6342.
(32) The Friedel-Crafts cyclization proceeded smoothly (1-2 h) at
room temperature under these conditions,18a yielding the cyclized
product in good yield (84%), while under the conditions originally
reported3b the cyclization did not proceed to completion.
(33) The imines of 3-substituted indanones are very sensitive to air
and moisture, especially if they are liquids. Therefore, as soon as they
were purified by Kugelrohr distillation, the products were transferred
to a glovebox. All of the prepared imines were stored in a -40 °C freezer
in a glovebox.
(26) Experimental/calculated ee value of recovered ketone according
to eq 2.24 For data in Table 2: 3a , 92/91.8, 93/93.9, 86/87.9; 3b, 83/
83.9. For data in Table 3: 8a , 96/94.5, 99/100; 8c, 89/89.4, 75.5/74.
(27) Equation 1 in ref 24 was used.
(28) Equations 5 and 6 in ref 24 were used.

772 J . Org. Chem., Vol. 65, No. 3, 2000
Yun and Buchwald
131.07, 127.29, 124.80, 122.30, 55.98, 40.69, 36.48, 35.23,
29.91, 24.32, 23.05, 14.27, 12.42; IR (neat) 1656 cm-1. Anal.
Calcd for C16H23N1: C, 83.79; H, 10.11. Found: C, 83.63; H,
10.18.
25.10, 24.00; IR (neat) 1629 cm-1. Anal. Calcd for C16H21N1:
C, 84.53; H, 9.31. Found: C, 84.42; H, 9.34.
(4-(3,4-Dich lor op h en yl)-3,4-d ih yd r o-2H-n a p h th a len -1-
ylid en e)m eth yla m in e (10). Using the general procedure, the
title compound was prepared as a white solid in 91% yield:
(3-ter t-Bu tylin d a n -1-ylid en e)p r op yla m in e (2d ). Using
the general procedure, the title compound was prepared as a
colorless oil in 89% yield: 1H NMR (300 MHz, CDCl3) 7.81 (d,
J ) 7.61 Hz, 1H), 7.44 (br d, J ) 7.65 Hz, 1H), 7.37-7.26 (m,
2H), 3.49-3.39 (m, 2H), 3.14 (dd, J ) 2.51, 7.86 Hz, 1H), 2.70
(dd, J ) 1.74, 17.46 Hz, 1H), 2.60 (dd, J ) 7.85, 17.46 Hz,
1H), 1.84-1.74 (m, 2H), 1.01 (t, J ) 7.42 Hz, 3H), 0.92 (s, 9H);
13C NMR (125 MHz, CDCl3) 173.38, 150.78, 141.01, 130.24,
127.40, 127.25, 122.15, 55.87, 51.55, 34.93, 32.63, 27.81; IR
(neat) 1657 cm-1; HRMS (EI) m/z calcd for C16H23N1 229.1831,
found 229.1834.
1
mp 145-146 °C; H NMR (500 MHz, CDCl3) 8.25 (br s, 1H),
7.37-7.27 (m, 3H), 7.20 (d, J ) 1.83 Hz, 1H), 6.92-6.89 (m,
2H), 4.17 (dd, J ) 4.62, 6.55 Hz, 1H), 3.34 (s, 3H), 2.61-2.45
(m, 2H), 2.35-2.25 (m, 1H), 2.19-2.08 (m, 1H); 13C NMR (75
MHz, CDCl3) 165.77, 144.54, 140.76, 134.97, 132.69, 130.64,
130.57, 130.52, 130.32, 128.95, 128.07, 127.53, 125.79, 44.40,
38.81, 30.58, 24.72; IR (neat) 1625 cm-1. Anal. Calcd for
C
17H15N1Cl2: C, 67.12; H, 4.97. Found: C, 66.85; H, 5.03.
Gen er a l P r oced u r e for th e Kin etic Resolu tion . To a
solution of (R,R)-1 (3.5 mg, 0.01 mmol) in anhydrous THF (1
mL) were added sequentially phenylsilane (12 µL, 0.1 mmol),
pyrrolidine (8 µL, 0.1 mmol), and MeOH (4 µL, 0.1 mmol)
under an atmosphere of argon in a resealable Schlenk tube.
The reaction mixture was then heated at 60 °C until the color
of the solution turned from yellow to green. The Schlenk tube
was removed from the oil bath and cooled to room temperature.
In a glovebox, the imine substrate (1 mmol) and internal
standard (dodecane or tetradecane) were added to a flask fitted
with a septum. The imine substrate was dissolved in THF (1
mL) and was added to the activated catalyst via pipet in a
glovebox (Table 3). The Schlenk tube was sealed, removed from
the glovebox, and allowed to stir for an indicated amount of
time. For low-temperature (0 °C or below) reactions, the
substrate in THF was transferred via cannula to the Schlenk
flask containing a solution of the activated catalyst with the
aid of a Schlenk line (Table 2). The Schlenk tube was resealed
was allowed to stir for an indicated amount of time.
Wor k u p : P r oced u r e A.34 The reaction mixture was diluted
with either pentane or hexane (10 mL) and was added slowly
to a NaOAc/HOAc buffer solution35 (10 mL) with vigorous
stirring (Ca u tion : bubbling!). The biphasic mixture was
stirred for 5-12 h. At this point, the organic layer containing
ketone from hydrolysis of the unreacted imine was separated,
ether (10 mL) was added, and the combined organic layers
were washed with brine. The organic layer was dried over
MgSO4 and concentrated. Chromatography gave ketone in
>95% purity as judged by 1H NMR and GC. The aqueous layer
containing the amine products was basified with 2 N NaOH
and was extracted with ether (3 × 15 mL). The combined
organic layers were dried over Na2SO4 and purified by chro-
matography. P r oced u r e B (Th e P r efer r ed Meth od for
Im in es of Tetr a lon es). The reaction mixture was diluted
with ether (10 mL) and was added to a 1 N NaOH solution
(10 mL) with vigorous stirring (Ca u tion : bubbling!). The
reaction mixture was stirred for 10-30 min until the bubbling
stopped, and the aqueous layer was extracted with ether (3 ×
20 mL). The combined ether extracts were dried over Na2SO4
and concentrated. Both ketone and amine products were
purified by silica gel chromatography. After elution of ketone
substrate from the unreacted imine with hexane and ether,
the polarity of the eluant was increased to elute the amine
products36 (CH2Cl2/MeOH/triethylamine (or ammonium hy-
droxide solution) ) 10:1:0.1). The isolated products were >95%
pure as judged by GC and 1H NMR. The absolute configuration
of the product was determined by optical rotation in the cases
where it had previously been determined. The ee was deter-
mined by GC analysis using a Chiraldex BPH (20 m length,
0.25 mm ID) or GTA column (20 m length, 0.25 mm ID) or
Meth yl(3-m eth ylin d a n -1-ylid en e)a m in e (5a ): Gen er a l
P r oced u r e for th e P r ep a r a tion for N-Meth ylim in es.
3-Methyl-1-indanone (2 g, 13.7 mmol) was placed in a dry
Schlenk flask under argon. Anhydrous ether (45 mL) was
added, and the reaction flask was cooled to -35 °C. Condensed
methylamine (2.0 mL, 68 mmol) was introduced via cannula,
followed by the addition of TiCl4 (0.83 mL, 7.5 mmol). The
reaction mixture was allowed to warm to room temperature
slowly and stirred overnight. The reaction mixture was filtered
through a pad of Celite and washed with ether. The combined
filtrates were concentrated, and the product was purified by
Kugelrohr distillation to give the title product as a pale pink
oil in 91% yield: 1H NMR (300 MHz, CDCl3) 7.75 (d, J ) 7.66
Hz, 1H), 7.44-7.26 (m, 3H), 3.42-3.32 (m, 1H), 3.33 (s, 3H),
3.00 (dd, J ) 8.15, 18.10 Hz, 1H), 2.24 (dd, J ) 5.38, 18.10
Hz, 1H), 1.36 (d, J ) 7.17 Hz, 3H); 13C NMR (75 MHz, CDCl3)
175.12, 154.53, 139.28, 131.32, 127.36, 124.57, 121.95, 41.02,
37.27, 35.20, 22.02; IR (neat) 1659 cm-1; HRMS (EI) m/z calcd
for C11H13N1 159.1048, found 159.1050.
Meth yl(3-p h en ylin d a n -1-ylid en e)a m in e (5b). Using the
general procedure, the title compound was prepared as a white
solid in 88% yield: mp 96-97 °C; 1H NMR (300 MHz, CDCl3)
7.83 (br d, J ) 6.60 Hz, 1H), 7.39-7.20 (m, 5H), 7.13-7.10
(m, 3H), 4.52 (dd, J ) 4.12, 8.61 Hz, 1H), 3.34 (s, 3H), 3.30
(dd, J ) 8.70, 18.59 Hz, 1H), 2.65 (dd, J ) 4.03, 17.58 Hz,
1H); 13C NMR (75 MHz, CDCl3) 174.57, 152.47, 145.25, 139.75,
131.61, 128.92, 127.82, 127.77, 126.85, 126.16, 121.91, 46.95,
41.15, 39.19; IR (neat) 1657 cm-1. Anal. Calcd for C16H15N1:
C, 86.84; H, 6.83. Found: C, 86.85; H, 6.94.
Met h yl(4-m et h yl-3,4-d ih yd r o-2H -n a p h t h a len -1-ylid e-
n e)a m in e (7a ). Using the general procedure, the title com-
pound was prepared as a colorless oil in 98% yield: 1H NMR
(300 MHz, CDCl3) 8.09 (dd, J ) 2.01, 7.78 Hz, 1H), 7.32 (dt, J
) 1.50, 7.33 Hz, 1H), 7.23 (dt, J ) 0.9, 7.51 Hz, 2H), 3.32 (s,
3H), 2.97-2.91 (m, 1H), 2.71-2.49 (m, 2H), 2.10-2.00 (m, 1H),
1.81-1.70 (m, 1H), 1.32 (d, J ) 6.96 Hz, 3H); 13C NMR (75
MHz, CDCl3) 166.71, 145.12, 134.12, 129.95, 126.86, 126.46,
125.53, 38.68, 33.00, 29.77, 24.97, 20.82; IR (neat) 1630 cm-1
.
Anal. Calcd for C12H15N1: C, 83.19; H, 8.73. Found: C, 83.01;
H, 8.71.
Meth yl(4-ph en yl-3,4-dih ydr o-2H-n aph th alen -1-yliden e)-
a m in e (7b). Using the general procedure, the title compound
was prepared as a white solid in 96% yield: mp 69-70 °C; 1H
NMR (300 MHz, CDCl3) 8.21-8.18 (m, 1H), 7.32-7.20 (m, 5H),
7.09-7.06 (m, 2H), 6.94-6.91 (m, 1H), 4.20 (dd, J ) 4.44, 6.82
Hz, 1H), 3.31 (s, 3H), 2.53 (br t, J ) 6.50 Hz, 2H), 2.30 (dtd,
J ) 4.40, 6.89, 13.05 Hz, 1H), 2.18 (qd, J ) 6.55, 13.10 Hz,
1H); 13C NMR (75 MHz, CDCl3) 166.42, 144.21, 142.13, 135.12,
130.05, 129.15, 128.70, 128.60, 127.05, 126.57, 125.47, 45.18,
38.80, 30.78, 24.97; IR (neat) 1625 cm-1. Anal. Calcd for
(34) This procedure was mostly used for experiments with more
unstable indanone imines.
(35) Meyers, A. I.; Williams, D. R.; Erickson, G. W.; White, S.;
Druelinger, M. J . Am. Chem. Soc. 1981, 103, 3082.
(36) Many of the minor amine products produced from the kinetic
resolution are more polar than the major amine products assumed to
be the cis isomer. Therefore, special care must be taken in order to
isolate the minor amine products during chromatography. Thus, the
diastereomeric ratio of amine products was determined on the crude
reaction mixture. The ee of the products was often determined after
transformation of the crude mixture of amines to proper derivatives
for chiral analysis.
C
17H17N1: C, 86.77; H, 7.28. Found: C, 86.67; H, 7.23.
(4-Cyclop en tyl-3,4-d ih yd r o-2H-n a p h th a len -1-ylid en e)-
m eth yla m in e (7c). Using the general procedure, the title
compound was prepared as a white solid in 97% yield: mp 61
1
°C; H NMR (300 MHz, CDCl3) 8.03 (dd, J ) 1.88, 7.28 Hz,
1H), 7.29-7.21 (m, 2H), 7.12 (br d, J ) 6.41 Hz, 1H), 3.31 (s,
3H), 2.60-2.50 (m, 3H), 2.18-2.09 (m, 1H), 1.98-1.76 (m, 3H),
1.71-1.54 (m, 3H), 1.46-1.38 (m, 2H), 1.33-1.24 (m, 2H); 13
C
NMR (75 MHz, CDCl3) 166.94, 144.62, 134.11, 129.26, 128.65,
126.70, 125.72, 44.70, 43.67, 38.79, 32.43, 31.47, 25.80, 25.28,

Asymmetric Hydrosilylation of Imines
J . Org. Chem., Vol. 65, No. 3, 2000 773
HPLC analysis using a Chiralcel OJ or OD column. The major
enantiomer of each compound indicated below reflects the
major enantiomer of the reaction products with (R,R)-1 except
in the case of sertraline where (S,S)-1 was used. The H NMR
analysis of the amine products reported below is that of the
major amine product.
J ) 3.26, 19.07 Hz, 1H), 1.94-1.89 (m, 1H), 1.60-1.31 (m, 5H),
0.92 (t, J ) 6.85 Hz, 3H); 13C NMR (125 MHz, CDCl3) 206.64,
159.22, 137.01, 134.79, 127.64, 125.81, 123.72, 43.36, 38.46,
36.05, 29.94, 22.96, 14.18; IR (neat) 1711 cm-1; HRMS (EI)
m/z calcd for C13H16O1 188.1201, found 188.1205.
1
(3-n -Bu tylin d a n -1-yl)p r op yla m in e (4c). This was iso-
lated in 64% yield based on 54% conversion (Table 1, entry 5).
The ee was measured by chiral GC on a BPH column with the
corresponding trifluoroacetamide derivative (110 °C isother-
mal, 1.0 mL/min; major isomer, 90.16 min (major), 92.87 min;
minor isomer, 77.30 min (major), 80.30 min): 1H NMR (300
MHz, CDCl3) 7.37-7.32 (m, 1H), 7.23-7.16 (m, 3H), 4.18 (t, J
) 7.78 Hz, 1H), 3.01-2.90 (m, 1H), 2.77-2.63 (m, 3H), 2.05-
1.97 (m, 1H), 1.66-1.47 (m, 3H), 1.46-1.28 (m, 6H), 0.96 (t, J
) 7.33 Hz, 3H), 0.96-0.90 (m, 3H); 13C NMR (75 MHz, CDCl3)
147.10, 145.75, 127.32, 126.55, 123.71, 123.48, 62.13, 49.51,
42.33, 41.56, 34.97, 30.11, 24.00, 23.22, 14.43, 12.21; IR (neat)
3311 cm-1; HRMS (EI) m/z calcd for C16H25N1 231.1987, found
231.1982.
3-Meth yl-1-in d a n on e (3a ). This was isolated in 73% yield37
based on 49% conversion (Table 1, entry 2). The ee was
measured by chiral GC on a BPH column (90 °C isothermal,
1.0 mL/min, 25.6 min, 26.9 min (major, R)): 1H NMR (300
MHz, CDCl3) 7.73 (d, J ) 7.79 Hz, 1H), 7.62 (ddd, J ) 1.18,
7.21, 7.70 Hz, 1H), 7.51 (qd, J ) 0.92, 7.82 Hz, 1H), 7.38 (ddt,
J ) 0.62, 0.93, 7.42 Hz, 1H), 3.45 (dquint, J ) 3.50, 7.21 Hz,
1H), 2.95 (dd, J ) 7.50, 19.07 Hz, 1H), 2.29 (dd, J ) 3.42, 19.07
Hz, 1H), 1.42 (d, J ) 7.17 Hz, 3H); 13C NMR (75 MHz, CDCl3)
206.50, 160.04, 136.51, 134.87, 127.52, 125.41, 123.56, 45.60,
33.07, 21.65; IR (neat) 1710 cm-1; [R]D -6.67° (for a sample of
91.6% ee) (c 1.05, EtOH) (lit.4 [R]D 10.1° (c 3, EtOH), 100% ee
(S)).
3-ter t-Bu tyl-1-in d a n on e (3d ). This was isolated in 72%
yield based on 60% conversion (Table 1, entry 7). The ee was
measured by chiral GC on a BPH column (120 °C isothermal,
1.0 mL/min, 20.62 min, 21.94 min (major)): 1H NMR (300
MHz, CDCl3) 7.74 (d, J ) 7.47 Hz, 1H), 7.60-7.53 (m, 2H),
7.41-7.36 (m, 1H), 3.24 (dd, J ) 2.13, 7.20 Hz, 1H), 2.70 (dd,
J ) 7.27, 19.35 Hz, 1H), 2.61 (dd, J ) 1.64, 19.42 Hz, 1H),
0.94 (s, 9H); 13C NMR (125 MHz, CDCl3) 207.00, 156.49,
137.97, 133.92, 128.11, 127.74, 123.54, 49.48, 41.30, 34.71,
27.87; IR (neat) 1713 cm-1; HRMS (EI) m/z calcd for C13H16O1
188.1201, found 188.1207.
(3-ter t-Bu tylin d a n -1-yl)p r op yla m in e (4d ). This was iso-
lated in 82% yield based on 60% conversion (Table 1, entry 7).
The ee was measured by chiral GC on a BPH column with the
corresponding trifluoroacetamide derivative (130 °C isother-
mal, 1.0 mL/min; major isomer, 25.17 min (major), 27.11 min;
minor isomer, not separated): 1H NMR (300 MHz, CDCl3)
7.46-7.43 (m, 1H), 7.38-7.35 (m, 1H), 7.23-7.15 (m, 2H), 4.08
(t, J ) 7.91 Hz, 1H), 2.92 (dd, J ) 7.66 Hz, 1H), 2.79-2.69 (m,
2H), 2.48 (td, J ) 7.25, 12.23 Hz, 1H), 1.76 (br s, 1H), 1.63-
1.45 (m, 3H), 1.05 (s, 9H), 0.97 (t, J ) 7.42 Hz, 3H); 13C NMR
(125 MHz, CDCl3) 147.09, 144.78, 126.92, 126.39, 125.80,
123.92, 61.18, 53.32, 49.60, 37.47, 33.39, 28.66, 24.01, 12.16;
IR (neat) 3311 cm-1; HRMS (EI) m/z calcd for C16H25N1
231.1987, found 231.1984.
(3-Meth ylin d a n -1-yl)p r op yla m in e (4a ). This was isolated
in 69% yield based on 49% conversion (Table 1, entry 2). The
ee was measured by chiral GC on a BPH column with the
corresponding trifluoroacetamide derivative (120 °C isother-
mal, 1.0 mL/min, cis isomer, 12.6 min (major (1R,3S)), 13.5
min;, trans isomer, not separated): 1H NMR (300 MHz, CDCl3)
7.36-7.31 (m, 1H), 7.26-7.17 (m, 3H), 4.19 (dd, J ) 7.09, 8.89
Hz, 1H), 3.08-2.98 (m, 1H), 2.78-2.64 (m, 3H), 1.66-1.39 (m,
3H), 1.34 (d, J ) 6.68 Hz, 3H), 1.36-1.26 (m, 1H), 0.96 (t, J )
7.42 Hz, 3H); 13C NMR (125 MHz, CDCl3) 147.95, 145.81,
127.42, 126.52, 123.60, 123.27, 62.17, 49.54, 44.20, 36.86,
23.99, 19.89, 12.11; IR (neat) 3304 cm-1. Anal. Calcd for
C
13H19N1: C, 82.48; H, 10.12. Found: C, 82.63; H, 9.74.
3-P h en yl-1-in d a n on e (3b). This was isolated in 81% yield
based on 58% conversion (Table 1, entry 4). The ee was
measured by chiral HPLC on a OJ column (10% i-PrOH/90%
hexanes, 0.5 mL/min, 19.75 min, 25.00 min (major)): 1H NMR
(300 MHz, CDCl3) 7.81 (dd, J ) 0.50, 7.69 Hz, 1H), 7.57 (dt, J
) 1.28, 7.46 Hz, 1H), 7.42 (t, J ) 7.40 Hz, 1H), 7.34-7.22 (m,
4H), 7.14-7.11 (m, 2H), 4.58 (dd, J ) 3.84, 7.97 Hz, 1H), 3.24
(dd, J ) 7.97, 19.23 Hz, 1H), 2.70 (dd, J ) 3.85, 19.32 Hz,
1H); 13C NMR (125 MHz, CDCl3) 205.93, 158.05, 143.88,
136.99, 135.18, 129.07, 128.03, 127.78, 127.14, 127.04, 123.57,
47.01, 44.66; IR (neat) 1710 cm-1; [R]D 61.2° (for a sample of
90% ee) (c 2.50, EtOH). Anal. Calcd for C15H12O1: C, 86.51;
H, 5.81. Found: C, 86.26; H, 5.89.
(3-P h en ylin d a n -1-yl)p r op yla m in e (4b). This was isolated
in 76% yield based on 58% conversion (Table 1, entry 4). The
ee was measured by chiral GC on a BPH column with the
corresponding trifluoroacetamide derivative (140 °C for 40 min,
ramp of 0.5 °C/min to 150 °C and then isothermal for 30 min;
major isomer, 82.11 min (major), 83.37 min; minor isomer,
78.44 min (major), 80.03 min): 1H NMR (300 MHz, CDCl3)
7.42 (d, J ) 7.66 Hz, 1H), 7.35-7.15 (m, 7H), 6.91 (d, J ) 7.50
Hz, 1H), 4.31 (dd, J ) 7.25, 8.72 Hz, 1H), 4.17 (dd, J ) 7.50,
10.11 Hz, 1H), 2.93 (td, J ) 6.68, 12.23 Hz, 1H), 2.73 (t, J )
7.01 Hz, 2H), 1.79 (ddd, J ) 9.08, 10.30, 11.98 Hz, 1H), 1.64-
1.51 (m, 3H), 0.95 (t, J ) 7.48 Hz, 3H); 13C NMR (75 MHz,
CDCl3) 146.20, 145.96, 144.77, 128.67, 128.54, 127.60, 126.92,
126.62, 125.07, 123.63, 62.23, 49.38, 49.13, 45.82, 23.93, 12.09;
IR (neat) 3308 cm-1. Anal. Calcd for C18H21N1: C, 86.01; H,
8.42. Found: C, 86.34; H, 8.29.
3-Meth yl-1-in d a n on e (3a ). This was isolated in 84% yield
based on 55% conversion (Table 2, entry 2).
Meth yl(3-m eth yl-in d a n -1-yl)a m in e (6a ). This was iso-
lated in 87% yield based on 55% conversion (Table 2, entry 2).
The ee was measured by chiral GC on a GTA column with the
corresponding trifluoroacetamide derivative (110 °C isother-
mal, 1.0 mL/min; major isomer, 36.30 min (major), 38.19 min;
minor isomer, 31.66 min (major), 33.92 min): 1H NMR (300
MHz, CDCl3) 7.35-7.30 (m, 1H), 7.27-7.18 (m, 3H), 4.13 (dd,
J ) 7.51, 8.61 Hz, 1H), 3.09-3.01 (m, 1H), 2.71 (td, J ) 7.05,
12.14 Hz, 1H), 2.55 (s, 3H), 1.43 (br s, 1H), 1.35 (d, J ) 6.84
Hz, 3H), 1.37-1.24 (m, 1H); 13C NMR (75 MHz, CDCl3) 147.91,
145.13, 127.47, 126.49, 123.60, 123.25, 63.66, 43.39, 36.85,
34.04, 20.06; IR (neat) 3311 cm-1; HRMS (EI) m/z calcd for
C
11H15N1 161.1204, found 161.1201.
3-P h en yl-1-in d a n on e (3b). This was isolated in 81% yield
based on 52% conversion (Table 2, entry 6).
Meth yl(3-p h en yl-in d a n -1-yl)a m in e (6b). This was iso-
lated in 81% yield based on 52% conversion (Table 2, entry 6).
The ee was measured by chiral GC on a BPH column with the
corresponding trifluoroacetamide derivative (150 °C isother-
mal, 1.0 mL/min; major isomer, 67.15 min, 70.10 min (major);
minor isomer, 58.38 min (major), 59.72 min)): 1H NMR (300
MHz, CDCl3) 7.42 (d, J ) 7.24 Hz, 1H), 7.35-7.21 (m, 7H),
6.92 (d, J ) 7.42 Hz, 1H), 4.25 (dd, J ) 7.05, 8.71 Hz, 1H),
4.19 (dd, J ) 7.60, 10.03 Hz, 1H), 2.96 (td, J ) 7.01, 12.36 Hz,
1H), 2.56 (s, 3H), 1.77 (ddd, J ) 9.43, 10.07, 12.36 Hz, 1H),
1.58 (br s, 1H),; 13C NMR (75 MHz, CDCl3) 146.25, 145.47,
144.71, 128.68, 128.50, 127.69, 126.94, 126.64, 125.10, 123.66,
63.72, 49.07, 45.12, 33.95; IR (neat) 3319 cm-1; HRMS (EI)
m/z calcd for C16H17N1 223.1361, found 223.1356.
3-n -Bu tyl-1-in d a n on e (3c). This was isolated in 92% yield
based on 54% conversion (Table 1, entry 5). The ee was
measured by chiral GC on a BPH column with the correspond-
ing alcohols after reduction with NaBH4 (110 °C, isothermal,
1.0 mL/min; major alcohol, 95.24 min (major), 99.24 min; minor
alchol, 86.91 min (major), 91.20 min): 1H NMR (300 MHz,
CDCl3) 7.73 (d, J ) 7.66 Hz, 1H), 7.60 (dt, J ) 1.14, 7.17 Hz,
1H), 7.51 (br d, J ) 6.68 Hz, 1H), 7.37 (br t, J ) 7.58 Hz, 1H),
3.38-3.32 (m, 1H), 2.86 (dd, J ) 7.50, 19.07 Hz, 1H), 2.37 (dd,
(37) The maximum yield is 100% when the amount of (1 - %
conversion/100) × (initial mol of the starting imine substrate) is
recovered for ketone substrate and (% conversion/100) × (initial mol
of the starting imine substrate) for amine products.

774 J . Org. Chem., Vol. 65, No. 3, 2000
Yun and Buchwald
4-Meth yl-1-tetr a lon e (8a ). This was isolated in 90% yield
based on 54% conversion (Table 3, entry 2). The ee was
measured by chiral GC on a GTA column (105 °C isothermal,
1.0 mL/min, 42.68 min (major (4R)), 48.52 min): 1H NMR (300
MHz, CDCl3) 8.03 (dd, J ) 1.47, 7.82 Hz, 1H), 7.51 (ddd, J )
1.47, 7.41, 7.58 Hz, 1H), 7.35-7.28 (m, 2H), 3.10 (dq, J ) 4.56,
7.01 Hz, 1H), 2.81 (ddd, J ) 4.60, 8.68, 17.40 Hz, 1H), 2.61
(ddd, J ) 4.85, 8.60, 17.40 Hz, 1H), 2.26 (tdd, J ) 4.64, 8.64,
13.36 Hz, 1H), 1.97-1.86 (m, 1H), 1.41 (d, J ) 7.01 Hz, 3H);
13C NMR (75 MHz, CDCl3) 198.56, 149.05, 133.77, 131.94,
127.57, 127.36, 126.65, 36.57, 33.04, 30.77, 20.97; IR (neat)
1684 cm-1; [R]D 21.6° (for a sample of 99% ee) (c 1.99, MeOH)
(lit.6 [R]D -7° (c 2.5, MeOH), 23% ee (S)). Anal. Calcd for
min): 1H NMR (300 MHz, CDCl3) 7.36-7.33 (m, 1H), 7.24-
7.12 (m, 3H), 3.71 (t, J ) 6.11 Hz, 1H), 2.67-2.61 (m, 1H),
2.49 (s, 3H), 2.25-2.16 (m, 1H), 1.94-1.36 (m, 12H), 1.24-
1.14 (m, 1H); 13C NMR (75 MHz, CDCl3) 141.97, 139.04,
129.02, 128.64, 126.39, 125.95, 57.73, 45.10, 42.52, 33.80,
31.90, 29.66, 25.66, 25.10, 23.40; IR (neat) 3328 cm-1. Anal.
Calcd for C16H23N1: C, 83.79; H, 10.11. Found: C, 83.59; H,
10.01.
Kin etic Resolu tion of r a c-10 (Ta ble 4, F ir st En tr y). To
a solution of (S,S)-(EBTHI)TiF2 (8.7 mg, 0.025 mmol) in
anhydrous THF (1 mL) were sequentially added phenylsilane
(12 µL, 0.1 mmol), pyrrolidine (8 µL, 0.1 mmol), and MeOH (4
µL, 0.1 mmol) under an atmosphere of argon in a resealable
Schlenk tube. The reaction mixture was then heated at 60 °C
until the color of the solution turned from yellow to green. The
Schlenk tube was removed from the oil bath and cooled to room
temperature. Imine substrate 10 (152 mg, 0.5 mmol) and
dodecane (0.11 mL) were added to a flask fitted with a septum
in a glovebox. The mixture was then dissolved in THF (1.5
mL)38 and was added to the active catalyst via pipet in the
glovebox. The reaction tube was sealed and stirred at room
temperature for 24 h. The workup procedure B gave the
corresponding ketone (56 mg, 0.19 mmol) and the amine
products (73 mg, 0.24 mmol, 96:4 ) cis/trans).
4-(3,4-Dich lor op h en yl)-1-t et r a lon e (12). The ee was
measured by chiral HPLC on a OD column (10% i-PrOH/90%
hexanes, 0.7 mL/min, 10.77 min, 12.02 min (major, (R)): mp
84 °C; 1H NMR (300 MHz, CDCl3) 8.13 (dd, J ) 1.63, 7.82 Hz,
1H), 7.48 (dt, J ) 1.71, 7.46 Hz, 1H), 7.40 (d, J ) 8.31 Hz,
1H), 7.38 (t, J ) 8.30 Hz, 1H), 7.23 (d, J ) 2.12 Hz, 1H), 6.97-
6.93 (m, 2H), 4.28 (dd, J ) 4.48, 8.23 Hz, 1H), 2.72 (ddd, J )
4.81, 7.66, 17.27 Hz, 1H), 2.63 (ddd, J ) 4.72, 8.80, 17.28 Hz,
1H), 2.52-2.42 (m, 1H), 2.32-2.20 (m, 1H); 13C NMR (75 MHz,
CDCl3) 197.51, 145.00, 144.14, 134.04, 132.93, 132.84, 131.12,
130.77, 130.68, 129.45, 128.12, 127.71, 127.55, 44.77, 36.79,
31.93; IR (neat) 1677 cm-1; [R]D -63.0° (for a sample of 96.3%
ee) (c 1.0, acetone) (lit.18b [R]D 55.7° (c 1.01, acetone), 84% ee
(S)). Anal. Calcd for C16H12O1Cl2: C, 66.00; H, 4.15. Found:
C, 65.88; H, 4.27.
C
11H12O1: C, 82.46; H, 7.55. Found: C, 82.39; H, 7.38.
N -Me t h yl-4-m e t h yl-1,2,3,4-t e t r a h yd r o-1-n a p h t h yl-
a m in e (9a ). This was isolated in 92% yield based on 54%
conversion (Table 3, entry 2). The ee was measured by chiral
GC on a GTA column with the corresponding trifluoroaceta-
mide derivative (110 °C isothermal, 1.0 mL/min; major isomer,
65.68 min (major), 68.20 min; minor isomer, 71.91 min (major),
74.63 min): 1H NMR (300 MHz, CDCl3) 7.34-7.31 (m, 1H),
7.27-7.13 (m, 3H), 3.64 (t, J ) 4.81 Hz, 1H), 2.88-2.81 (m,
1H), 2.50 (s, 3H), 1.97-1.78 (m, 3H), 1.72-1.60 (m, 2H), 1.33
(d, J ) 7.01 Hz, 3H); 13C NMR (75 MHz, CDCl3) 142.44, 138.72,
128.80, 127.83, 127.06, 125.82, 57.73, 34.21, 32.90, 27.66,
25.84, 22.51; IR (neat) 3315 cm-1; HRMS (EI) m/z calcd for
C
12H17N1 175.1361, found 175.1356.
4-P h en yl-1-tetr a lon e (8b). This was isolated in 85% yield
based on 53% conversion (Table 3, entry 4). The ee was
measured by chiral HPLC on an OJ column (10% i-PrOH/90%
hexanes, 1.0 mL/min, 9.74 min, 18.96 min (major)): 1H NMR
(300 MHz, CDCl3) 8.12 (dd, J ) 1.63, 7.66 Hz, 1H), 7.44 (dt, J
) 1.55, 7.46 Hz, 1H), 7.39-7.23 (m, 4H), 7.13-7.09 (m, 2H),
6.99 (br d, J ) 7.66 Hz, 1H), 4.31 (dd, J ) 4.64, 7.90 Hz, 1H),
2.74 (ddd, J ) 4.64, 7.98, 17.11 Hz, 1H), 2.63 (ddd, J ) 4.48,
9.13, 17.11 Hz, 1H), 2.53-2.43 (m, 1H), 2.37-2.25 (m, 1H);
13C NMR (75 MHz, CDCl3) 198.25, 146.41, 143.82, 133.79,
132.93, 129.71, 128.80, 128.76, 127.25, 127.21, 126.95, 45.53,
36.99, 32.09; IR (neat) 1682 cm-1. Anal. Calcd for C16H14O1:
C, 85.45; H, 6.35. Found: C, 86.27; H, 6.46.
N -Me t h yl-4-p h e n yl-1,2,3,4-t e t r a h yd r o-1-n a p h t h yl-
a m in e (9b). This was isolated in 86% yield based on 53%
conversion (Table 3, entry 4). The ee was measured by chiral
GC on a BPH column with the corresponding trifluoroaceta-
mide derivative (150 °C isothermal, 1.0 mL/min; major isomer,
90.30 min, 94.39 min (major); minor isomer, not separated)):
1H NMR (300 MHz, CDCl3) 7.36 (d, J ) 7.82 Hz, 1H), 7.31-
7.01 (m, 7H), 6.83 (d, J ) 7.17 Hz, 1H), 4.01 (dd, J ) 5.58,
8.79 Hz, 1H), 3.73 (t, J ) 3.73 Hz, 1H), 2.54 (s, 3H), 2.15-
1.98 (m, 1H), 1.89-1.79 (m, 1H), 1.60 (br s, 1H); 13C NMR (75
MHz, CDCl3) 147.09, 140.14, 139.01, 130.23, 129.16, 129.00,
128.49, 127.26, 126.36, 126.26, 57.56, 46.26, 34.22, 28.87,
25.85; IR (neat) 3345 cm-1; HRMS (EI) m/z calcd for C17H19N1
237.1518, found 237.1523.
4-Cyclop en tyl-1-tetr a lon e (8c). This was isolated in 86%
yield based on 44% conversion (Table 3, entry 6). The ee was
measured by chiral GC on a BPH column (150 °C isothermal,
1.0 mL/min, 30.39 min (major), 32.59 min): mp 56-58 °C; 1H
NMR (300 MHz, CDCl3) 8.01 (dd, J ) 1.39, 7.74 Hz, 1H), 7.45
(dt, J ) 1.39, 7.46 Hz, 1H), 7.31 (dt, J ) 1.14, 7.58 Hz, 1H),
7.26 (br d, J ) 7.82 Hz, 1H), 2.87-2.55 (m, 4H), 2.24-2.18
(m, 2H), 2.05-1.87 (m, 2H), 1.74-1.42 (m, 4H), 1.38-1.27 (m,
2H); 13C NMR (75 MHz, CDCl3) 198.80, 148.57, 133.05, 131.99,
129.18, 127.55, 126.88, 44.22, 44.20, 34.53, 32.48, 31.63, 26.80,
25.31, 25.02; IR (neat) 1677 cm-1. Anal. Calcd for C15H18O1:
C, 84.07; H, 8.47. Found: C, 83.98; H, 8.50.
N-Meth yl-4-(3,4-d ich lor op h en yl)-1,2,3,4-tetr a h yd r o-1-
n a p h th yla m in e (11). The ee was measured by chiral HPLC
on a OD column with the corresponding 1-naphthoylamide
derivative (10% i-PrOH/ 90% hexanes, 0.6 mL/min; cis isomer,
27.16 min, 30.41 min (major, (1S,4S)); trans isomer, 22.24 min,
1
38.52 min): mp 59-61 °C; H NMR (300 MHz, CDCl3) 7.35
(dd, J ) 1.14, 7.66 Hz, 1H), 7.34 (d, J ) 8.31 Hz, 1H), 7.25 (d,
J ) 2.12 Hz, 1H), 7.20 (ddt, J ) 0.81, 1.46, 7.58 Hz, 1H), 7.11
(dt, J ) 1.47, 7.50 Hz, 1H), 6.97 (dd, J ) 2.12, 8.15 Hz, 1H),
6.79 (d, J ) 7.66 Hz, 1H), 3.98 (dd, J ) 5.46, 9.37 Hz, 1H),
3.73 (t, J ) 4.16 Hz, 1H), 2.54 (s, 3H), 2.09-1.96 (m, 3H), 1.87-
1.75 (m, 1H), 1.64 (br s, 1H); 13C NMR (75 MHz, CDCl3) 147.46,
139.09, 138.88, 132.40, 130.86, 130.46, 130.20, 129.98, 129.43,
128.45, 127.51, 126.79, 57.48, 45.60, 34.42, 28.63, 25.76; IR
(neat) 3346 cm-1. Anal. Calcd for C17H17N1Cl2: C, 66.68; H,
5.60. Found: C, 66.62; H, 5.63.
The major amine product (1R,4R) from the resolution with
(R,R)-1 was converted to the corresponding HCl salt and its
absolute stereochemistry was determined by optical rotation:
[R]D -32.5° (c 0.77, MeOH) (from the amine of 91.5% ee) (lit.3a
[R]D -37.2° (c 2, MeOH), 100% ee (1R,4R)).
Ack n ow led gm en t. This research was generously
supported by the National Institutes of Health (GM-
46059). Additional unrestricted support from Pfizer,
Merck, and Novartis is gratefully acknowledged.
N-Meth yl-4-cyclop en tyl-1,2,3,4-tetr a h yd r o-1-n a p h th y-
la m in e (9c). This was isolated in 86% yield based on 44%
conversion (Table 3, entry 6). The ee was measured by chiral
GC on a BPH column with the corresponding trifluoroaceta-
mide derivative (130 °C for 40 min, ramp of 0.2 °C/min to 150
°C and then isothermal for 5 min; major isomer, 117.66 min,
124.63 min (major); minor isomer, 124.41 min (major), 127.01
J O991328H
(38) The imine substrate (0.5 mmol) showed lower solubility than
those of the other imine substrates in this work and was not completely
soluble in 1.5 mL of THF. However, after its addition to the activated
catalyst via pipet and the mixture became homogeneous (0.5 mmol
substrate/ 2.5 mL solvent in total).