Stereo- and regioselectivity in asymmetric synthesis of α-amino substituted benzocyclic compounds

Article abstract of DOI:10.1016/S0957-4166(98)00467-4

The enantiomerically pure chiral benzocyclic amines 6-8 were obtained by asymmetric transamination of the corresponding prochiral ketones 9a-c. The method involves: (a) formation of chiral imines 10a-c from the prochiral ketones 9a-c and the inexpensive c

Full text of DOI:10.1016/S0957-4166(98)00467-4

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 4369–4379
Stereo- and regioselectivity in asymmetric synthesis of α-amino
substituted benzocyclic compounds
Arie L. Gutman,∗ Marina Etinger, Gennady Nisnevich and Felix Polyak
Technion-Israel Institute of Technology, Technion City, Haifa, 32000, Israel
Received 13 October 1998; accepted 16 November 1998
Abstract
The enantiomerically pure chiral benzocyclic amines 6–8 were obtained by asymmetric transamination of
the corresponding prochiral ketones 9a–c. The method involves: (a) formation of chiral imines 10a–c from the
prochiral ketones 9a–c and the inexpensive chiral auxiliary (R)- or (S)-phenylethylamine(PEA); (b) asymmetrically
induced reduction of these imines to the diastereomeric amines 11a–c and 12a–c; (c) catalytic hydrogenation to
remove the benzylic fragment of the chiral PEA auxiliary. The stereoselectivity of the imine reduction, as well as the
regioselectivity of the catalytic hydrogenation, are strongly dependent on the size of the saturated ring condensed
with the benzene ring. This approach was used to develop a convenient, high yielding, and stereoselective route
to several practically important optically active α-amino substituted benzocyclic compounds. © 1998 Elsevier
Science Ltd. All rights reserved.
1. Introduction
Chiral amines are intermediates for many commercial and developmental drugs, and are also used as
chiral auxiliaries and resolving agents.1 In many cases, the corresponding prochiral ketone is the key
intermediate in the conventional manufacture of the racemic amine. The desired stereoisomer can then
be obtained by the resolution of the racemic amine via: (a) fractional crystallisation of the diastereomeric
salt of that amine with a chiral acid;2 (b) by chromatography on a chiral stationary phase;3 and (c) more
recently, by enzyme catalysed aminolysis in organic solvents.4 The disadvantage of all these resolution
procedures is the fact that the unwanted enantiomer has to be recycled or discarded, which means that
the desired chiral amines may be isolated in yields not higher than 50%. This problem may be avoided if
it is possible to use asymmetric synthesis.
Recognising the strategic importance of the corresponding prochiral ketones, much effort has been
directed toward developing methodologies for their asymmetric transamination into chiral amines,
∗
Corresponding author. E-mail: chgutman@tx.technion.ac.il
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00467-4

4370
A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
using readily available chiral benzylic amines as auxiliaries. This concept involves formation of the
intermediate chiral imine 3 from the key prochiral ketone 1 and the chiral benzylic amine auxiliary 2,
asymmetrically induced catalytic or stoichiometric reduction of this imine into the diastereomeric amine
4, followed by removal of the no longer needed benzylic auxiliary fragment by catalytic hydrogenation
(Scheme 1).
Scheme 1.
The absolute configuration of the newly generated stereogenic centre in the amine 5 is determined by
the configuration of the chiral auxiliary, and since both enantiomers of the phenylethylamine (PEA) are
available, either configuration of this centre could be generated at will.
This approach was successfully used for the synthesis of psychotomimetic phenylisopropylamines,5
chiral alkyl amines,6 derivatives of mescaline,7 as well as taxol’s side chain8 and natural amino acids.9
The case when n=0 and R or R1 is an aromatic moiety is a special problem because the intermediate
diastereomeric amine 4 is ‘dibenzylic’ and its catalytic hydrogenation may cause the cleavage of either of
the two benzylic C–N bonds (Scheme 2). While hydrogenation promoted cleavage according to route (a)
would result in the desired product 5, cleavage according to route (b) would be completely unproductive
and wasteful. Only one attempt was reported to apply this transamination approach to the synthesis
of chiral benzylic amines (n=0, R is an oxygenated phenyl, R1=Me). This is the work of Bringmann
and Geisler,10 who succeeded in obtaining enantiomerically pure oxygenated 1-phenylethylamines using
oxygenated acetophenones as substrates.
Scheme 2.
Chiral benzocyclic amines (Fig. 1) are playing an increasingly important role in pharmaceutical
chemistry and provide key intermediates for a number of valuable pharmaceutical preparations with
mostly neurotropic and psychotropic activity.2d,11,12
Of particular interest to us is the chiral (R)-1-aminoindane 6, which is the key intermediate in
the synthesis of (R)-N-propargyl-1-aminoindan (PAI), a potent irreversible inhibitor of the B form of
monoamine oxidase which could be used in the treatment of Parkinson’s disease.11,12 In view of the
growing importance of this class of compounds and, in particular, because of our continuing interest

A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
4371
Figure 1.
in developing a practical synthesis of (R)-1-aminoindane 6, we undertook a study of the scope and
limitations of the asymmetric synthesis approach to the synthesis of chiral α-substituted benzocyclic
amines.
2. Results and discussion
We now report the applicability of the asymmetric transamination approach, with enantiomerically
pure phenylethylamines (PEA) as chiral auxiliaries, to the synthesis of chiral α-substituted benzocyclic
amines. The practically important (R)-1-aminoindane 6, (R)- and (S)-1,2,3,4-tetrahydro-1-naphthylamine
7 and (R)- and (S)-5-amino-6,7,8,9-tetrahydro-5H-benzocycloheptene 8 were chosen as target molecules
for this study (Fig. 1).
Our approach includes three steps which are summarised in Scheme 3: (a) synthesis of chiral imines
10a–c by condensation of ketone 9a–c with (R)-(+)-PEA; (b) stereoselective reduction of the chiral
imine C_N bond with an achiral reductive agent to predominantly one diastereomer of the mixture
of diastereomeric amines 11a–c and 12a–c; and (c) regioselective removal of the chiral auxiliary by
catalytic hydrogenolysis to obtain 6–8.
2.1. Synthesis and analysis of imines 10a–c
The imines 10a–c were prepared by condensation of ketones 9a–c with (R)-(+)-(PEA) in the presence
of catalytic amounts of trifluoroacetic acid (TFA) by refluxing in toluene or xylene with azeotropic
removal of water. The imines were not isolated from the reaction mixture in pure form because of
decomposition during distillation or purification on silica gel, so the crude reaction mixtures (which
contained according to GC analysis about 5–7% of unreacted PEA, 5–7% of unreacted ketones and
76–85% of the main product) were subjected to IR, GC–MS, HRMS and NMR analysis. These analyses
indicated that imines 10a–c were the only main products (Table 1).
1
Analysis of H NMR spectra of the reaction mixtures revealed that in the case of 10a or 10b only
one quartet for Hα at 4.84 or 4.88 ppm and only one doublet for the Me group at 1.59 or 1.54 ppm
were observed. This indicates that only one geometrical isomer of 10a and 10b is present in the reaction
mixtures. In the case of 10c, two quartets were observed at 4.82 and 4.60 ppm, which could belong
to two geometrical isomers of the imine in the ratio 1.2:1. We suppose that for imines 10a–b, when a
five- or a six-membered ring is fused with a benzene ring, the syn isomer could not exist due to steric
hindrances, whereas in the case of imine 10c, the larger and more flexible seven-membered ring makes
possible the existence of anti as well as syn isomers. At this stage we do not have enough information for
the conclusive identification of various isomers, but this problem is of considerable theoretical interest
and its investigation is currently in progress in our laboratory.

4372
A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
Scheme 3.
2.2. Asymmetric hydrogenation of imines 10a–c
In the next step of the synthetic sequence, the imine can be hydrogenated in several ways to induce the
formation of the second chiral centre. To find the optimal conditions providing the best stereoselectivity
for this transformation, we studied the catalytic reduction of the chiral imines 10a–c over palladium on
charcoal (Pd/C), as well as the stoichiometric reduction with sodium borohydride. The results are given
in Table 2.
Reduction of imines 10a and 10b with sodium borohydride proceeded with a high diastereoselectivity
(de>92%) and gave as the main products (R,R) amines 11a and 11b. The configuration of newly created
stereogenic centres was tentatively assigned as (R) by comparison of specific rotations of 1-aminoindane
and 1,2,3,4-tetrahydro-1-naphthylamine, obtained after hydrogenolysis of 11a and 11b, with specific
rotations of pure enantiomers.2,13
Asymmetric reduction of Shiff’s bases using B2H6 has been reported by Charles et al.14 The steric
course of the reaction was discussed and it was concluded that the conformation of the substrate in the
transition state includes the four-centred complex with the M–H bridge from the less hindered side.
Applying this model permits an explanation of the predominant creation of the (R,R) diastereomer in the
reduction of imine 10a with sodium borohydride (Fig. 2).
If the configuration of 10a is anti, and in the most favoured transition-state conformation the hydrogen

A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
4373
Table 1
Spectral characteristics of crude imines 10a–c
Table 2
The ratio of diastereomers 11a–c/12a–c obtained by catalytic and stoichiometric reduction of imines
10a–c*
Figure 2. Transition state will yield the (R,R)-isomer

4374
A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
from the phenylethyl moiety is lying in the plane of the molecule, the reduction takes place from the less
hindered side, i.e. from the side of the methyl group, which results in an (R) configuration of the new
stereogenic centre. We believe that this model can also be applied to the reduction of 10b, in spite of the
fact that the six-membered ring is not planar. However, if the C_N double bond, together with the two
α-carbons and the hydrogen from the phenylethyl moiety are lying in the same plane, the reduction of
the anti isomer of 10b from the side of the methyl group, will result in (R) induction.
In the case of imine 10c, the diastereoselectivity of the reduction with sodium borohydride was very
poor (de 16%), which is probably due to either the existence of 10c in two isomeric forms anti and
syn, with reduction of each form from the less hindered side resulting in a new stereogenic centre with
an opposite configuration, or because applying the model discussed above is not warranted in this case
because of the complex and unpredictable conformation of the seven-membered ring.
As it is possible to see from Table 2, the diastereoselectivity of the reduction of 10c increased
substantially when catalytic reduction over Pd/C was employed. Moreover, the diastereoselectivity was
influenced by the solvent used. The best results were obtained when hexane was used as a solvent (62%
de), and the diastereoselectivity decreased in more polar solvents: toluene (56% de) ethyl acetate (54%
de) and methanol (42% de). In each case the major product was the (R,R) diastereomer. The absolute
configuration of the new stereogenic centre predominantly formed in 11c was established as (R) by X-
ray analysis of its perchlorate salt (Fig. 3).
To explain the results of the catalytic hydrogenation of 10c, we used the model proposed by Harada
et al.,15 who studied the diastereoselectivity of catalytic reduction of (ꢀ)-N-(ethylbenzylidene)-α-
ethylbenzylamine over Pd/C and found that the ratio of (ꢀ) to meso forms of the product is not
dependent on the anti to syn ratio in the precursor imines. On this basis they proposed a mechanism of
hydrogenation, whereby the substrate is isomerised rapidly by the catalyst bringing about an equilibrium
of syn and anti isomers. Each isomer is then hydrogenated at a different rate from the least hindered side.
Thus, if the catalyst causes the syn/anti isomerisation of 10c (Keq), and every isomer is hydrogenated
from the less hindered side to give with different rates (ksyn, kanti) the (R,R)- and (R,S)-diastereomers, the
ratio between them is defined, according to Curtin–Hammet equation, by Keq, ksyn, kanti
.
If an isomer (syn or anti), which gives the (R,R)-diastereomer is hydrogenated faster than the
equilibrium between isomers (syn/anti) is shifted, then the main product will be the (R,R) diastereomer. In
non-polar hexane, where the interaction between the substrate and the surface of the catalyst is stronger
than in the polar methanol, this process is more active and the diastereoselectivity of the reduction is
better.
2.3. Catalytic debenzylation of amines 11a–c
N-Debenzylations via catalytic hydrogenation are widely used in organic synthesis.16 Competitive N-
debenzylation (as exemplified by routes ‘a’ and ‘b’ in Scheme 2, i.e. the influence of α-substituents and
substituents in the benzene ring on the regioselectivity of N-debenzylation was studied.17 It was found
that the presence of a methyl group as an α-substituent to nitrogen reduces the reactivity of the adjacent
C–N benzylic bond.
We found that the regioselectivity of N-debenzylation depends critically on the size of the saturated
ring condensed with the benzene ring. The regioselectivity of N-debenzylation toward the benzocyclic
amines (R)-6–8 is enhanced in the order 11a<11b<11c (Table 3). In the case of the five-membered
ring 11a, the regioselectivity of the hydrogenolysis was not satisfactory, which means that a planar
five-membered ring unsuccessfully competes with the methyl group from the phenylethyl moiety for
the stabilisation of the adjacent C–N benzylic bond. Enlargement of the saturated ring, which amounts to

A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
4375
Figure 3. X-Ray structure of 11c — perchlorate (only a cation is shown)
increasing the size of alkyl substituent, improves the regioselectivity. Thus, in the case of a six-membered
ring the ratio (R)-7:(R)-PEA was 3.5:1, and in the case of a seven-membered ring the ratio (R)-8:(R)-PEA
increased to 45.5:1.
Catalytic removal of the phenylethyl moiety from the secondary amines (R)-11a–c was performed on
Pd/C, in methanol–water solution with an excess of acetic acid (about 4 equivalents), under a hydrogen
pressure of about 4 atm and at 18°C. The reaction was monitored by HPLC and it was found that the
hydrogenolysis is not accompanied by racemisation. Amines (R)-6–8 were separated from (R)-PEA, via
crystallisation with (R,R)-tartaric or hydrochloric acid.
3. Conclusions
The method of asymmetric transamination of benzocyclic ketones 9a–c using (R)-(+)-PEA as a chiral
auxiliary was applied to the synthesis of enantiomerically pure benzocyclic amines 6–8, which are
potentially useful pharmaceutical syntones.
(i) Imines 10a–c were obtained with high yields from the prochiral ketones 9a–c and the chiral
auxiliary (R)-(+)-PEA. The anti to syn ratio of 10a–c depends on the size of the saturated ring fused

4376
A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
Table 3
Catalytic hydrogenation of amines (R,R)-11a–c
with the benzene ring and changes from predominantly anti (>95%) for five- and six-membered
rings 10a–b to a virtually 50:50 ratio for the seven-membered ring 10c.
(ii) The diastereoselectivity of the reductions of 10a–c with sodium borohydride and catalytic re-
duction over Pd/C was studied. It was found that for 10a–b the diastereoselectivity of reduction
with sodium borohydride is high (de>92%) and gives the (R,R)-diastereomer as a major product
which corresponds to the reduction of the anti-isomer from the less hindered side. For 10c, the
diastereoselectivity of the reduction with sodium borohydride was low (de 16%), but increased
substantially under catalytic hydrogenation conditions with Pd/C (de>42%). A dependence of the
de on the solvent used was observed with the best diastereoselectivity obtained in non-polar solvents
(hexane, de 62%).
(iii) The absolute configuration of the newly created stereogenic centre in 11c was established as (R)
with the help of X-ray analysis, which means that the major product of 10c reduction is also the
(R,R)-diastereomer.
(iv) The regioselectivity of hydrogenolytic cleavage of (R,R)-11a–c to (R)-6–8 over Pd/C was studied.
It was found that the regioselectivity of the reaction increases from 11a [(R)-6:(R)-PEA 2.2:1] to
11c [(R)-8: (R)-PEA 45.5:1].
4. Experimental section
4.1. General details
1
Melting points and boiling points are uncorrected. H NMR spectra were determined with Bruker-
200 AM (200 MHz) and Bruker 400-AM WB (400 MHz) instruments in CDCl3 solutions with TMS
as an internal standard. 13C NMR spectra were determined with a Brucker 400 AM WB (100.614
MHz) instrument in CDCl3 solutions. IR spectra were run on a Nicolet Impact 400. HRMS spectra
were measured on a Varian Mat-711 at 70 eV. GC–MS spectra were measured on a Magnum (Finnigan)
with a DB-5 column. Gas chromatographic analyses were obtained on a Hewlett–Packard 5890 Series
II chromatograph using a DB-210 capillary column (30 m/0.25 mm), stationary phase OV-210, carrier
gas — helium (2.0 ml/min), detector FID (250°C). Program A — 100°C (5 min), 10°/min (14 min),

A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
4377
240°C (10 min). Program B — 100°C (5 min), 1°/min (60 min), 20°/min (4 min), 240°C (10 min).
HPLC was performed on a Merck–Hitachi instrument (pump 6200A, UV detector L-4000, integrator D-
2500) on a Crownpack chiral column, eluent HClO4–water (pH 2). Optical rotations were measured on
a JASCO DIP-370 polarimeter. Commercially available (R)-(+)-phenylethylamine (Fluka) 99%+ (GC),
bp 187–189°C, [α]20 +38.6 (d 0.952) was used as a chiral auxiliary. 1-Indanone (Aldrich) 99%+, bp
D
243–245°C. α-Tetralone (Aldrich) 98%, bp 113–116°C (6 mmHg). 1-Benzosuberon (Aldrich) 99%, bp
270°C. Pd/C catalyst, Type 487, Johnson Matthey, %Pd 9.95; %H2O 0.5.
4.2. Synthesis of amines 11a–b via asymmetric reduction of imines 10a–b with sodium borohydride
4.2.1. (R,R)-N-(1-Indanyl)-1-phenylethylamine 11a
A mixture of 1-indanone (20.0 g, 0.152 mol), (R)-(+)-PEA (18.4 g, 0.152 mol) and 100 mg of
trifluoroacetic acid (TFA) was refluxed in toluene (200 ml) with azeotropic removal of water during
7 h. The solution was cooled to room temperature and added dropwise at −20 to −30°C during 30 min to
the stirred solution of sodium borohydride (6 g, 0.159 mol) in absolute ethanol (150 ml). The resulting
mixture was stirred at −20 to −30°C during an additional 2–3 h period and then kept at −16°C for two
days. Water was added to the mixture and the solvents were removed in vacuo. The residue was extracted
with ether (150 ml), dried over sodium sulphate, concentrated in vacuo and distilled at 115–120°C (0.1
mmHg) to obtain 20 g of the diastereomeric mixture 11a and 12a in the ratio 96:4. A solution of amines
11a and 12a (20 g, 0.084 mol) in methanol (100 ml) was added to the solution of (R,R)-tartaric acid
(12.6 g, 0.084 mol) in methanol, the mixture was stirred for ca. 15 min and the solvent was evaporated.
The solid crude product was crystallised twice from ethanol to obtain 15 g of the tartrate salt of 11a
with de>99%, mp 161–163°C, [α]20 +50 (c 0.85, MeOH). The free amine 11a (9 g, 25% yield based on
D
1-indanone, was isolated from the salt by the standard procedure, [α]20 +91 (c 0.945, MeOH).
D
1H NMR δ ppm 1.46 (d, 3H); 1.60 (s, 1H); 1.74–1.82 (m, 1H); 2.26–2.34 (m, 1H); 2.75–2.82 (m, 1H);
2.98–3.05 (m, 1H); 4.15–4.22 (m, 2H); 7.24–7.52 (m, 9H). 13C NMR δ ppm (aliphatic region) 24.42
(CH3); 30.02 (CH2); 34.87 (CH2); 56.27 (CH); 60.74 (CH). HRMS (m/z) 237.1505. C17H19N calcd
237.1517.
4.2.2. (R,R)-N-1-(1,2,3,4-Tetrahydronaphthyl)-1-phenylethylamine 11b
A mixture of 1-tetralone (20.0 g, 0.14 mol), (R)-PEA (19.8 g, 0.16 mol) and 100 mg of TFA was
refluxed in xylene (150 ml) with azeotropic removal of water during 8 h. The cooled solution was added
dropwise during 40 min, at −20 to −30°C, to the stirred solution of sodium borohydride (5.2 g, 0,18
mol) in absolute ethanol (150 ml) and the mixture was stirred for an additional 2 h at −20 to −30°C and
after this was kept overnight at −16°C. Water was added to the mixture, the solvents were removed in
vacuo, and the product extracted with ether, dried over sodium sulphate and the ether was removed in
vacuo. The crude mixture of 11b and 12b in the ratio 97:3 was dissolved in ethanol (50 ml) and added
to the solution of (R,R)-tartaric acid (24 g, 0.16 mol) in ethanol (250 ml). The resulting solution was
concentrated to dryness to give 38 g of the salt. Additional crystallisation from ethanol gave 16 g of
tartaric salt of (R,R)-isomer with de>99%, mp 72–75°C, [α]20 +59 (c 0.92, MeOH). 11b (9.5 g, the yield
D
27% based on 1-tetralone) was isolated from the salt by standard procedure, [α]20 +94 (c 1.09, MeOH).
D
1H NMR δ ppm 1.36 (d, 3H); 1.65–1.77 (m, 3H); 1.84–1.93 (m, 1H); 2.69–2.80 (m, 2H); 3.70 (dd,
1H); 4.04 (q, 1H); 7.07–7.49 (m, 9H). 13C NMR δ ppm (aliphatic region) 24.77 (CH3); 18.73 (CH2);
29.13 (CH2); 29.56 (CH2); 53.25 (CH); 56.18 (CH). HRMS (m/z) 251.1660. C18H12N calcd 251.1674.

4378
A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
4.3. Hydrogenation of imines 10 a–c on Pd/C in various solvents — general procedure
The hydrogenation was carried out in hexane, toluene, ethyl acetate and methanol at 25–27°C and at a
hydrogen pressure of about 4 atm.
A solution of imine 10a–c (1 g/30 ml) and Pd/C (10% from the weight of imine) was shaken until
absorption of hydrogen ceased. The analysis of the reaction mixtures was performed by GC.
4.3.1. (R,R)-N-[5-(6,7,8,9-Tetrahydro-5H-benzocycloheptenyl)]-1-phenylethylamine 11c
A mixture of benzosuberone (9.42 g, 0.06 mol), (R)-(+)-PEA (7.11 g, 0.06 mol) and TFA (100 mg) in
toluene (150 ml) was refluxed with azeotropic removal of water during 8 h.
The solution was cooled to room temperature, mixed with Pd/C (1.5 g), and shaken under a hydrogen
pressure of about 4 atm during 30 h. The catalyst was filtered off through Celite, the solvent evaporated
and the residue (16.5 g) was dissolved in methanol and mixed with a methanolic solution of (R,R)-
tartaric acid (9.25 g). The solution was stirred during 15 min, methanol was removed and the residue was
triturated with ether and crystallised three times from ethanol to obtain 4.5 g of the salt with de>98%,
mp 192–193°C, [α]20 +69 (c 0.945, MeOH). 11c (2.85 g, 18% yield based on 1-benzosuberone) was
D
isolated from the salt, [α]20 +120 (c 0.94, MeOH).
D
1H NMR δ ppm 1.47 (d, 3H); 1.67–2.11 (m, 7H); 2.80–2.86 (m, 1H); 3.18 (dd, 1H); 3.75–3.82 (m,
2H); 7.23–7.49 (m, 9H). 13C NMR δ ppm (aliphatic region) 25.01 (CH3); 27.00 (CH2); 27.92 (CH2);
34.73 (CH2); 35.50 (CH2); 55.05 (CH); 59.21 (CH). HRMS (m/z) 265.1787. C19H23N calcd 265.1831.
4.4. Hydrogenolysis of amines 11a–c on Pd/C — general procedure
Hydrogenolysis of 11a–c was performed in a methanol–water solution in the presence of 4 equivalents
of acetic acid at a hydrogen pressure of about 4 atm, at 17–20°C, on Pd/C. At the end of the reaction,
the mixture was passed through Celite to remove the catalyst and the methanol was evaporated. Free
amines were isolated by standard procedure and separation of amines 6–8 from PEA were performed via
crystallisation of their tartrates or hydrochloride salts.
4.4.1. (R)-(−)-1-Indanamine 6
A mixture of 11a (de>99%) (6 g, 0.025 mol), Pd/C (0.6 g), acetic acid (5.7 ml), water (6 ml) and
methanol (60 ml) was shaken at 25°C, during 62 h under a hydrogen pressure of about 4 atm. The
catalyst was filtered off, the methanol was removed in vacuo and the residue dissolved in chloroform. The
chloroform solution was extracted with 25% aqueous solution of NH4OH, the organic layer separated,
dried over sodium sulphate and the solvent removed. The crude oil (4.2 g) contained (R)-6 and (R)-PEA
in the ratio about 64:36. The mixture was dissolved in methanol and HCl (gas) was passed through the
solution. Methanol was removed, the crude crystalline residue was refluxed with 40 ml of ethyl acetate
during 15 min, crystals (3 g) with the ratio of (R)-6:(R)-PEA about 83:17 were filtered off, dissolved in
6 ml of hot methanol and ethyl acetate (6 ml) was added to the hot solution. 1.4 g (20% yield) of the
salt with the ratio of (R)-10:(R)-PEA about 96.5:3.5 was separated, ee of (R)-6>99%, [α]20 −17 (c 1.3,
D
MeOH) {lit.,2 [α]20 −16.5 (c 1.5, MeOH)}.
D
1H NMR δ ppm 1.60–1.71 (m, 1H); 1.66 (s, 2H); 2.41–2.54 (m, 1H); 2.69–3.00 (m, 2H); 4.32 (t, 1H);
7.19–7.31 (m, 4H).

A. L. Gutman et al. / Tetrahedron: Asymmetry 9 (1998) 4369–4379
4379
4.4.2. (R)-(−)-1,2,3,4-Tetrahydro-1-naphthylamine 7
A mixture of 11b, de>99%, (8 g, 0.032 mol), Pd/C (0.8 g), water (4 ml), acetic acid (7.3 ml) and
methanol (160 ml) was shaken during 24 h under a hydrogen pressure of about 4 atm. The mixture
was worked up as described for (R)-6 and distilled at 120–140°C/5 mmHg to give 2 g (43%) of (R)-7,
ee>99%, which contained about 10% of PEA. The crude product was dissolved in methanol, mixed with
the methanolic solution of (R,R)-tartaric acid (2 g), stirred for about 15 min, methanol was removed in
vacuo, the residue washed with ether and the solid crystallised twice from ethanol to give 2 g of the (R)-
7-tartrate salt), [α]20 +17 (c 1.10, MeOH), which gave 0.9 g of (R)-7 (yield 18%), [α]20 −26 (c 1.32,
D
D
MeOH), ee>99% {lit.13 [α]20 −42.1 (d 1.0240)}.
D
1H NMR δ ppm 1.63 (s, 2H); 1.67–1.81 (m, 2H); 1.88–2.01 (m, 2H); 3.97 (t, 1H); 7.04–7.41 (m, 4H).
4.4.3. (R)-5-Amino-6,7,8,9-tetrahydro-5H-benzocycloheptene 8
A mixture of 11c (96% de) (1.5 g, 0.0057 mol), Pd/C (0.15 g), water (0.9 ml), acetic acid (1.3 ml) and
methanol (30 ml) were shaken during 25 h at 25°C. The mixture was worked up as described for (R)-6
and 7 to obtain 1 g of crude (R)-8, which was purified via crystallisation of its tartrate. 0.95 g (53%) of
the salt, which did not melt till 210°C was separated, [α]20 +26 (c 1.125, MeOH).
D
1H NMR δ ppm (free base) 1.51 (s); 1.48–2.10 (m, 6H); 2.79–2.84 (m, 2H); 4.21 (d, 1H); 7.08–7.19
(m, 3H); 7.40 (d, 1H).
Acknowledgements
This research was supported by the Fund for the Promotion of the Research at the Technion.
References
1. Jacques, J.; Collet, A.; Wilen, S. H. Enantiomers, Racemates and Resolutions; John Wiley: Chichester, 1981.
2. (a) Fontana, L. P.; Chandramouly, T.; Smith, H. E.; Polavarupu, P. L. J. Org. Chem. 1988, 53, 3379. (b) Smith, H. E.;
Willis, T. C. Tetrahedron 1970, 26, 107. (c) Lawson, W. B.; Rao, G. J. S. Biochemistry 1980, 19, 2133. (d) Smith, R.;
White, R. L.; Krantz, A. J. Med. Chem. 1988, 31, 1558. (e) Kipping, F. S. J. Chem. Soc. 1901, 81, 579–582.
3. (a) Dobashi, A.; Dobashi, Y.; Kinoshita, K.; Hara, S. Anal. Chem. 1988, 60, 1985. (b) Stalcup, A. M.; Gahm, K. H. Anal.
Chem. 1996, 68, 1369. (c) Shitangkoon, A.; Vigh, G. J. High Resolut. Chromatogr. 1993, 16(8), 504.
4. (a) Gutman, A. L.; Meyer, M.; Kalerin, E.; Polyak, F.; Sterling, J. Biotech. Bioeng. 1992, 40, 760. (b) Kitaguchi, H.;
Fitzpatrick, P. A.; Huber, J. E.; Klibanov, A. M. J. Am. Chem. Soc. 1989, 111, 3094.
5. Nichols, D. E.; Barfknecht, C. F.; Rusterholz, D. B.; Benington, F.; Morin, R. D. J. Med. Chem. 1973, 16, 480.
6. (a) Charles, J. P.; Christol, H.; Solladie, G. Bull. Soc. Chim. Fr. 1972, 1124. (b) Demailly, G.; Solladié, G. ibid. 1975,
2128.
7. Standridge, R. T.; Howell, H. G.; Gylys, J. A.; Partyka, R. A. J. Med. Chem. 1976, 19(12), 1400.
8. (a) Bourzat, J. D.; Commerçon, A. Tetrahedron Lett. 1993, 34, 6049; (b) Georg, G. I.; Mashava, P. M.; Akgün, E.; Milstead,
M. W. Tetrahedron Lett. 1991, 32, 3151.
9. (a) Hiskey, R. G.; Northrop, R. C. J. Am. Chem. Soc. 1965, 87, 1753. (b) Harada, K.; Iwasaki, T.; Okawara, T. Bull. Chem.
Soc. Jpn. 1973, 46, 1901.
10. Bringmann, G.; Geisler, J.-P. Tetrahedron Lett. 1989, 30, 317.
11. Youdim, M. B. H.; Finberg, J. P. M.; Levy, R.; Sterling, J.; Lerner, D.; Berger-Paskin, T. Eur. Pat. Appl. EP 436,492.
12. Sterling, J.; Levy, R.; Veinberg, A.; Goldenberg, W.; Finberg, J.; Youdim, M.; Gutman, A. Eur. Pat. Appl. EP 538,134.
13. (a) Ghislandi, V.; Vercesi, D. Farmaco Ed. Sci. 1971, 26, 474. (b) Weidmann, R.; Guette, J. P. C. R. Ser. C 1969, 268, 2225.
(c) Ghislandi, V.; La Manna, A.; Vercesi, D. Farmaco Ed. Sci. 1976, 31, 561.
14. Charles, J. P.; Christol, H.; Solladie, G. Bull. Soc. Chim. Fr. 1970, 4439.
15. Harada, K.; Yoshida, T. Bull. Chem. Soc. Jpn. 1972, 45, 3706.
16. Rylander, P. N. Catalytic Hydrogenation in Organic Syntheses; Academic Press: London, 1979.
17. (a) Baltzly, R.; Buck, J. S. J. Am. Chem. Soc. 1943, 65, 1984. (b) Baltzly, R.; Russel, P. B. ibid. 1953, 75, 5598. (c) Dahn,
H.; Solms, U.; Zoller, P. Helv. Chim. Acta 1952, 35, 2117.