Resolution of ortho- and meta-substituted 1-phenylethylamines with isopropylidene glycerol hydrogen phthalate

Article abstract of DOI:10.1016/S0957-4166(01)00156-2

The hydrogen phthalate of isopropylidene glycerol 1, previously reported as an efficient resolving agent of p-substituted 1-phenylethylamines, was also found to resolve selected o- and m-isomers. In particular, the (S)-enantiomers of 1-(2-methylphenyl)eth

Full text of DOI:10.1016/S0957-4166(01)00156-2

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 1071–1075
Resolution of ortho- and meta-substituted 1-phenylethylamines
with isopropylidene glycerol hydrogen phthalate
a,
a
a
b
Marco Pallavicini, * Ermanno Valoti, Luigi Villa and Oreste Piccolo
a
Istituto di Chimica Farmaceutica e Tossicologica, Universit a` di Milano, viale Abruzzi 42, I-20131 Milan, Italy
b
Studio di Consulenza Scientifica, via Born o` 5, I-22060 Sirtori (LC), Italy
Received 23 March 2001; accepted 2 April 2001
Abstract—The hydrogen phthalate of isopropylidene glycerol 1, previously reported as an efficient resolving agent of p-substituted
-phenylethylamines, was also found to resolve selected o- and m-isomers. In particular, the (S)-enantiomers of 1-(2-
methylphenyl)ethylamine 2, 1-(3-methylphenyl)ethylamine 3, 1-(2-chlorophenyl)ethylamine 4 and 1-(3-methoxyphenyl)ethylamine
were obtained in good yields and very high enantiomeric excess (e.e.) by selective crystallization of the respective salts with (S)-
1
5
or (R)-1. The e.e.s of the resolved substrates were determined by chiral HPLC analysis. The (S)-configuration of (−)-3 was
established according to Raban’s procedure. Optical rotations of non-racemic free amines 2 and 3 are reported. The success of
the resolutions presented and of the precedent ones using 1 indicate that the position of the substituent on the 1-phenylethylamine
framework does not affect the resolution, showing the uncommon versatility of 1 in the resolution of monosubstituted
1
-phenylethylamines. © 2001 Elsevier Science Ltd. All rights reserved.
1
. Introduction
bromine, methyl, nitro and methoxyl, or b-substituents,
such as one or two methyl residues, does not preclude
1,2
The hydrogen phthalate of isopropylidene glycerol 1
has proved to efficiently resolve a wide range of amines,
including a number of 1-arylalkylamines, 1-phenyl-2-
propenylamine, 1-phenyl-2-propynylamine, 1-methyl-
the resolution by 1.
In this context, it seemed pertinent to extend the inves-
tigations on the resolution properties of 1 to ortho- and
meta-substituted 1-phenylethylamines, which have not
previously been examined. On the basis of the previous
1–3
allylamine and 1-(1-cyclohexenyl)ethylamine.
The
diastereoisomeric salt between (S)-1 and the (S)-isomer
of the amine is generally the less soluble and was
reported to precipitate from methanol or propan-2-ol,
allowing isolation of the (S)-amine from the racemate
in high yield and e.e. The opposite stereochemical
outcome was observed only in the case of 1-phenyl-2-
propynylamine, where the enantiomer producing the
less soluble diastereomeric salt with (S)-1 had (R)-
1,2
results, the resolution of such amines with 1 was
likely, but not certain, to succeed as recent resolution
studies on 1-phenylethylamines bearing a b-, p-, m- or
o-substituent using mandelic acid and its derivatives as
resolving agents demonstrated that the position of the
substituent on the 1-phenylethylamine (in particular the
methyl and methoxy analogues) can dramatically affect
the resolution efficiency, sometimes preventing the reso-
3
configuration. The effects of structural modification of
the parent amine, 1-phenylethylamine, on the resolution
process have been examined showing, in particular, that
the introduction of para-substituents, such as chlorine,
4–6
lution of the amine.
Therefore, we selected four
substrates, 1-(2-methylphenyl)ethylamine 2, 1-(3-
O
O
O
R
NH2
O
COOH
1
2
3
R = 2-Me
R = 3-Me
4
5
R = 2-Cl
R = 3-MeO
*
0
Corresponding author. E-mail: marco.pallavicini@unimi.it
957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00156-2

1
072
M. Palla6icini et al. / Tetrahedron: Asymmetry 12 (2001) 1071–1075
1
–3
methylphenyl)ethylamine 3, 1-(2-chlorophenyl)ethyl-
amine 4 and 1-(3-methoxyphenyl)ethylamine 5, as
candidates for the resolution with 1. The three differ-
ent substituents were chosen out of those present in
the previously resolved 1-phenylethylamines so as to
allow direct comparisons with the analogous resolu-
tions of the corresponding p- or b-substituted amines.
In particular, preference was given to methyl,
methoxyl and chlorine, considering the ready
availability of (±)-2–5 and the recent interest in the
the precedent resolutions.
In our study, methanol,
which was employed as the solvent in the resolutions
of 1-(p-methylphenyl)-, 1-(p-chlorophenyl)- and 1-(p-
methoxyphenyl)-ethylamine,
2
was replaced with
propan-2-ol due to the high solubility in methanol of
the diastereomeric salt mixtures. However, for the
second crystallization of the (R)-1·(S)-3 salt,
methanol was preferred to propan-2-ol but was used
in a very low ratio to the salt (1.2 mL/g). In this
case, despite the similar total efficiencies, recrystalliza-
tion of the salt of (S)-3 with 31.2% e.e. from propan-
4–10
preparation of pure enantiomers of these amines.
2
-ol raised the e.e. to only 55.4%, while crystallization
from methanol raised the e.e. to 98.9%.
2. Results
Reversed-phase chiral HPLC analysis of the salts
recovered from the first and second crystallizations
and of the liberated (S)-amines allowed the progress
of the resolution to be monitored. Under the chosen
chromatographic conditions, the following behavior
was observed: (S)-2, (S)-3, (S)-4 and (S)-5 eluted
after the respective (R)-isomers and in the case of
directly analyzed diastereomeric salt mixtures, both of
the base-resolved enantiomeric pairs were far pre-
The results of the resolutions of 2–5 are summarized
in Table 1, where the yields and e.e.s are reported. In
the case of amines 2, 4 and 5, selective precipitation
of the (S,S)-salt took place analogously to the stereo-
chemical course we had observed for all the previ-
ously
resolved
amines,
except
1-phenyl-
2-propynylamine.
Interestingly, 3 behaved similarly to 1-phenyl-2-prop-
ynylamine, (S)-1 giving the less soluble diastereomeric
salt with the (R)-base. Under the chiral HPLC condi-
tions optimized to determine the e.e.s, the (R)-enan-
tiomers of 2–5 elute before the (S)-enantiomers and
so their low percentage in highly (S)-enriched samples
was more accurately measurable than that of the (S)-
enantiomers in the reverse case. Therefore, in order to
facilitate chromatographic detection of the minor
enantiomer in the resolved amine, 3 was treated with
ceded by (S)-1 or (R)-1, which exhibited k%$0.
1
The (R)-enriched 2, 3, 4 and 5 were not isolated from
the alcoholic solutions remaining from the original
crystallizations of the salts of the respective (S)-iso-
mers. Nevertheless, the e.e.s were also determined by
HPLC analysis of the residues resulting from the con-
centration of the mother liquors and found consistent
with those expected on the basis of the chemical yield
and the diastereomeric composition of the precipi-
tates.
(
R)-1, which led to preferential precipitation of the
salt of (S)-3.
Equivalent amounts of racemic amine and enantio-
pure 1 were used in all experiments, as reported for
To the best of our knowledge, the optical rotations of
both the enantiomers of 2 and 3 are not reported in
Table 1. Preparation of (S)-2, (S)-3, (S)-4 and (S)-5 from the corresponding racemates by selective crystallization of the
a
respective salts with enantiopure 1 from propan-2-ol
Yieldb (%)
e.e. (%)
c
hDd
Entry
Amine
Resolving
agent
First cryst.e
Second cryst.
Recovered amine
68.2
First cryst.
Second cryst.
1
2
3
4
5
(S)-2
(S)-3
(S)-3
(S)-4
(S)-5
(S)-1
(R)-1
(R)-1
(S)-1
(S)-1
73.8
110.5
110.5
88.2
–
97.7
31.2
31.2
91.0
92.0
–
−60.7f
–
g
76.3
42.0
71.5
54.5
–
55.4
98.9
\99.6
40.6
66.0
53.6
−31.0f
h
−57.4
−22.3
i
67.2
97.5
a
The second crystallization of (R)-1·(S)-3 (entry 3) was performed from methanol.
Relative to the theoretical amount, i.e. half of starting racemate.
Enantiomeric excess of the amines determined by reversed-phase HPLC analysis on a Chiralcel OD-R column from Daicel (NaClO aq./CH CN
b
c
4
3
mixtures).
Of the recovered amine.
d
e
f
From the solution containing equivalent amounts of racemic amine and (S)- or (R)-1.
Observed rotation at 25°C; neat.
The salt, resulting from the second crystallization from propan-2-ol and containing (S)-3 with 55.4% e.e., was not decomposed. After being
recovered by filtration, it was recombined with the filtrate, propan-2-ol was removed and the resultant solid residue recrystallized from methanol
g
(
entry 3).
h
i
7
Observed rotation at 26°C; neat (lit. +55.9, for the (R)-isomer).
Specific rotation at 20°C; c 2, MeOH (lit. −17.8).
9

M. Palla6icini et al. / Tetrahedron: Asymmetry 12 (2001) 1071–1075
1073
the literature. Therefore, (−)-2 was converted into the
known levo-rotatory (S)-N-[1-(2- methylphenyl)ethyl]-
The results obtained for 3–5 confirm the trend observed
for the previous successful resolutions of 1-arylalkyl-
amines using (S)-1: the separations carried out in
propan-2-ol invariably required two crystallizations (in
contrast to those in methanol, which gave rise to the
precipitation of diastereomerically pure salts directly
from the solutions containing equivalent amounts of
resolving agent and racemic amine). This can be
attributed to the fact that compared to methanol
propan-2-ol tends to attenuate the difference in solubil-
ity between the two diastereoisomeric salts.
1
1
benzamide by treatment with benzoyl chloride and,
on the basis of this result, its absolute configuration
was established as (S). For the elucidation of the
absolute stereochemistry of (−)-3, it was transformed
into the diastereomeric mixture of the corresponding
N-(2,4-dinitrobenzenesulfenyl)-4-methylbenzenesulfon-
12
amide according to Raban’s procedure. The product
had a positive optical rotation sign, thus establishing
that (−)-3 was of (S)-absolute configuration. This
assignment was supported by the fact that the (S)-enan-
tiomers of the ortho- and para-isomers of 3 are also
levo-rotatory and by the elution order of the antipodes
on chiral HPLC. In fact, (−)-3 eluted after (+)-3,
as did (S)-1-phenylethylamine, (S)-1-(4-methylphenyl)-
ethyl-amine and (S)-2 with respect to the corresponding
In the case of 3, recrystallization of the salt containing
the amine with 31.2% e.e. from methanol (Table 1,
entry 3) was preferable to that from propan-2-ol (Table
1
, entry 2). This is consistent with the observations
9
reported for the resolution of 5 with mandelic acid. In
fact, according to the authors of this resolution,
methanol is better than propan-2-ol in the recrystalliza-
tion of salts with low d.e.
(
R)-isomers.
3
. Discussion and conclusion
Comparison between the resolutions of 2–5 and the
previous ones of the corresponding p- or b-substituted
1-phenylethylamines reveals small differences in the
total resolution efficiencies between the substrates (see
Table 2). The o-isomer 2 is more efficiently resolved
than 1-(p-methylphenyl)ethylamine, while 3, 4 and 5
As shown in Table 1 (entries 3–5), amines 3–5 were
isolated with very high e.e. after simple recrystallization
of the respective crude salts of (S)-1. In this context, 2,
which was recovered from the first formed precipitate
with 97.7% e.e. (Table 1, entry 1), represents an
exception.
Table 2. Total efficiencies (E) observed for the resolutions of 1-phenylethylamine and its methyl-, methoxy- and chloro-sub-
stituted analogues with 1
E, a
%
amine
Me
E, a
60 c
72
%
amine
E, a
%
amine
MeO
MeO
NH2
81 b
NH2
NH2
NH2
NH2
NH2
NH2
53
5
Me
62 c
2
3
Cl
Me
Me
NH2
42
71
4
NH2
50 c
81 c
Cl
a
Total resolution efficiency=chemical yield of the diastereomeric salt (%)×e.e. of the liberated amine (%)/100.
Calculated from the data reported in Ref. 1.
b
c
Calculated from the data reported in Ref. 2.

1
074
M. Palla6icini et al. / Tetrahedron: Asymmetry 12 (2001) 1071–1075
are resolved less efficiently than the respective p-iso-
mers. For the four monomethyl substituted 1-
phenylethylamines, the resolution efficiencies of 2 and
7.52 (m, 3H), 7.64 (d, 1H), 7.74 (d, 1H). The salt was
decomposed by treatment with 10% HCl in CH Cl .
2
2
The aqueous phase was separated, made alkaline with
1N NaOH, and extracted with ethyl acetate. Removal
of the solvent from the extract, previously dried over
Na SO , gave a colorless oil (8.22 g), which was dis-
3
are placed best and worst, respectively. However, all
four resolutions reported herein can be considered suc-
cessful new procedures to obtain enantiomerically pure
amines 2–5, the respective efficiencies ranging from a
very high 72%, in the case of 2, to a satisfactory 42%,
in the case of 3. These values can be better appreci-
ated considering that neither mandelic acid nor its
recently designed derivatives, namely 2-naphthylgly-
colic, p-methyl- and p-methoxymandelic acids, are
singly able to resolve all of the methyl- and methoxy-
2
4
tilled under vacuum yielding (S)-2 (8.01 g, 68.2% of
25
D
the theoretical amount): h =−60.7 (neat); e.e. 97.7%
(
(
1
3
0
by HPLC under the same conditions as described for
1
S)-1·(S)-2 salt); H NMR (CDCl ) l 1.36 (d, 3H),
3
.68 (s, 2H), 2.37 (s, 3H), 4.38 (q, 1H), 7.1–7.3 (m,
25
H), 7.48 (d, 1H). Benzamide of (S)-2: [h] =−17.2 (c
D
11
25
D
.4, chloroform) [lit. : [h] =+19 (c 0.38, chloroform)
4–6
substituted 1-phenylethylamines reported in Table 2.
for the (R)-isomer].
.2. (−)-(S)-1-(3-Methylphenyl)ethylamine (S)-3
R)-1 (39.8 g, 142.0 mmol) and (RS)-3 (19.2 g, 142.0
Indeed, the additional examples of resolutions
described in this paper indicate the high versatility of
4
1, which resolves the monomethyl substituted 1-
phenylethylamines irrespective of the position of the
substituent, and the amines 4 and 5 with efficiencies
comparable to those exhibited for the respective p-iso-
mers. On the basis of the results presented and previ-
ously reported, extension of the application of 1 to
the resolution of generically mono-substituted 1-
(
mmol) were combined in propan-2-ol (300 mL) at
room temperature. The white precipitate (32.6 g) was
collected by filtration, rinsed with cold propan-2-ol,
and recrystallized from methanol (40 mL) at 0°C
yielding (R)-1·(S)-3 salt (12.4 g, 42.0% of the theoreti-
cal amount): e.e. of (S)-3 98.9% (31.2% before the
phenylethylamines seems reasonable, though with the2
limitations previously shown for the b-substitution.
recrystallization) (by HPLC of the salts; 85/15 0.4 M
Further investigations are required to establish the
limits of the resolving properties of enantiopure 1,
which yet untested substrates may reveal.
1
NaClO /CH CN, 0.6 mL/min); H NMR (DMSO-d )
4
3
6
l 1.30 (s, 3H), 1.36 (s, 3H), 1.51 (d, 3H), 2.34 (s, 3H),
3
4
.84 (dd, 1H), 4.06 (pseudo t, 1H), 4.19 (d, 2H), 4.25–
.40 (m, 2H), 7.18 (d, 1H), 7.25–7.50 (m, 6H), 7.75 (d,
4
. Experimental
1H). The salt was decomposed in the same way as
described for (S)-1·(S)-2 obtaining a colorless oil (3.98
g), whose distillation under vacuum yielded (S)-3 (3.90
1
H NMR spectra were recorded on a Bruker 200 (200
MHz) instrument. Optical rotations were measured in
a 1 dm cell of 1 mL capacity using a Perking Elmer
25
g, 40.6% of the theoretical amount): h =−31.0 (neat);
D
e.e. 98.9% (by HPLC under the same conditions as
1
2
41 polarimeter. HPLC analyses were performed on a
described for (R)-1·(S)-3 salt); H NMR (CDCl ) l
3
Chiralcel OD-R column (250×4.6 mm I.D.) from
Daicel using a Waters 510 pump and a Pye Unicam
PU 4025 UV detector (analytical wavelength 254 nm).
Chromatographic data were collected and processed
using Maxima 820 software from Waters. Racemic
amines 2–5 were readily synthesized by the Leuckart
reaction from the corresponding ketones according to
the experimental procedure described for 1-phenylethyl-
1
7
.39 (d, 3H), 1.63 (s, 2H), 2.35 (s, 3H), 4.09 (q, 1H),
.0–7.30 (m, 4H).
4
.3. (−)-(S)-1-(2-Chlorophenyl)ethylamine (S)-4
(
S)-1 (16.1 g, 57.4 mmol) and (RS)-4 (8.94 g, 57.4
mmol) were combined in propan-2-ol (100 mL) at
room temperature. The white precipitate (11.04 g) was
collected by filtration, rinsed with propan-2-ol and
recrystallized from the same solvent (50 mL) yielding
1
3
amine. (S)-1 was prepared by resolution of the corre-
sponding racemate with (S)-1-phenylethylamine, as
1
previously reported. (R)-1 was obtained by treatment
(
S)-1·(S)-4 salt (8.95 g, 71.5% of the theoretical
of (S)-isopropylideneglycerol with phthalic anhydride
amount): e.e. of (S)-4 >99.6% (91% before the recrys-
tallization) (by HPLC of the salts; 85/15 1 M NaClO /
CH CN, 1 mL/min); H NMR (DMSO-d ) l 1.30 (s,
in pyridine, as described for the alternative prepara-
tion of (S)-1 from (R)-isopropylideneglycerol.
1
4
1
3
6
3
(
1
(
H), 1.35 (s, 3H), 1.47 (d, 3H), 3.79 (dd, 1H), 4.06
pseudo t, 1H), 4.19 (d, 2H), 4.36 (m, 1H), 4.66 (q,
H), 7.41–7.52 (m, 6H), 7.74–7.82 (m, 2H), 8.0–8.25
br s, 3H). The salt was decomposed in the same way
as described for (S)-1·(S)-2 obtaining a colorless oil
4
.1. (−)-(S)-1-(2-Methylphenyl)ethylamine (S)-2
(
S)-1 (48.7 g, 173.8 mmol) and (RS)-2 (23.5 g, 173.8
mmol) were combined in propan-2-ol (290 mL) at
room temperature. The white precipitate of (S)-1·(S)-2
salt (26.6 g, 73.8% of the theoretical amount) was
collected by filtration and rinsed with cold propan-2-
ol: e.e. of (S)-2 97.7% (by HPLC of the salt; 9/1 0.8
(
(
(
3.07 g), whose distillation under vacuum yielded (S)-4
26
2.95 g, 66.0% of the theoretical amount): h =−57.4
D
7
26
D
neat) [lit. h =+55.9 (neat; for the (R)-isomer with
4% e.e.)]; e.e. >99.6% (by HPLC under the same
1
9
M NaClO /CH CN, 1.6 mL/min); H NMR (DMSO-
d6) l 1.31 (s, 3H), 1.36 (s, 3H), 1.49 (d, 3H), 2.38 (s,
4
3
1
conditions as described for (S)-1·(S)-4 salt); H NMR
(CDCl ) l 1.39 (d, 3H), 1.61 (s, 2H), 4.54 (q, 1H),
7.11–7.35 (m, 3H), 7.52 (dd, 1H).
3
4
H), 3.84 (dd, 1H), 4.06 (pseudo t, 1H), 4.20 (d, 2H),
.37 (m, 1H), 4.56 (q, 1H), 7.22–7.30 (m, 3H), 7.39–
3

M. Palla6icini et al. / Tetrahedron: Asymmetry 12 (2001) 1071–1075
1075
4
.4. (−)-(S)-1-(3-Methoxyphenyl)ethylamine (S)-5
benzenesulfonamide derivatives of (−)-3 (1.40 g, 83.0%)
25
as a yellow solid: mp 67.2–72.9°C; [h] =+126.4 (c 0.5,
D
1
(
S)-1 (7.7 g, 27.5 mmol) and (RS)-5 (4.15 g, 27.5 mmol)
dichloromethane); H NMR (CDCl ) l 1.31 (d, 2.19H),
3
were combined in propan-2-ol (20 mL). After cooling
to 5°C, the white precipitate (3.98 g) was collected by
filtration, rinsed with cold propan-2-ol and recrystal-
lized from the same solvent (10 mL) yielding the (S)-
1.54 (d, 0.81H), 2.15 (s, 0.81H), 2.20 (s, 2.19H), 2.38 (s,
0.81H), 2.51 (s, 2.19H), 5.58 (m, 1H), 6.62 (s, 0.27H),
6.80–7.50 (m, 6.92H), 7.80–8.0 (m, 2.27H), 8.31 (d,
0.27H), 8.57 (dd, 0.27H), 8.92 (d, 0.73H), 9.16 (d,
0.27H). From the positive sign of the optical rotation of
the above derivative, (S)-configuration was assigned to
(−)-3 according to Ref. 12.
1
·(S)-5 salt (3.23 g, 54.5% of the theoretical amount):
e.e. of (S)-5 97.5% (92% before the recrystallization)
(
by HPLC of the salts; 85/15 1.5 M NaClO /CH CN,
4
3
1
0
.8 mL/min); H NMR (DMSO-d ) l 1.31 (s, 3H), 1.37
6
(
(
(
s, 3H), 1.53 (d, 3H), 3.80 (s, 3H), 3.85 (dd, 1H), 4.07
pseudo t, 1H), 4.20 (d, 2H), 4.25–4.43 (m, 2H), 6.93
dd, 1H), 7.09 (d, 1H), 7.18 (s, 1H), 7.27–7.55 (m, 4H),
Acknowledgements
7
.77 (dd, 1H). The salt was decomposed in the same
way as described for (S)-1·(S)-2 obtaining a colorless
This work was supported by the Ministero dell’Univer-
sit a` e della Ricerca Scientifica e Tecnologica of Italy.
oil (1.22 g), whose distillation under vacuum yielded
2
0
(
−
S)-5 (1.11 g, 53.6% of the theoretical amount): [h] =
D
9
20
22.3 (c 2, methanol) [lit. [h] =−17.8 (c 2, methanol)];
e.e. 97.5% (by HPLC under the same conditions as
described for (S)-1·(S)-5 salt); H NMR (CDCl ) l 1.38
d, 3H), 1.57 (s, 2H), 3.81 (s, 3H), 4.09 (q, 1H),
D
References
1
3
(
1. Pallavicini, M.; Valoti, E.; Villa, L.; Piccolo, O. Tetra-
hedron: Asymmetry 1996, 7, 1117.
6.74–6.83 (m, 1H), 6.87–6.92 (m, 2H), 7.25 (t, 1H).
2. Pallavicini, M.; Valoti, E.; Villa, L.; Piccolo, O. Tetra-
4
.5. Determination of the absolute configuration of (−)-3
hedron: Asymmetry 1997, 8, 1069.
3. Pallavicini, M.; Valoti, E.; Villa, L.; Piccolo, O. Tetra-
hedron: Asymmetry 2000, 11, 4017.
by derivatization as the N-(2,4-dinitrobenzenesulfenyl)-4-
methylbenzenesulfonamide
4
. Kinbara, K.; Sakai, K.; Hashimoto, Y.; Nohira, H.;
(
−)-3 (1 g, 7.4 mmol), p-toluenesulfonyl chloride (1.55
Saigo, K. Tetrahedron: Asymmetry 1996, 7, 1539.
g, 8.13 mmol) and triethylamine (1.54 mL, 11.1 mmol)
were dissolved in CH Cl . The solution was heated
under reflux for 3 h. After removal of the solvent, the
residue was purified by column chromatography on
silica gel (el. cyclohexane/ethyl acetate 85/15) yielding
5. Kinbara, K.; Sakai, K.; Hashimoto, Y.; Nohira, H.;
Saigo, K. J. Chem. Soc., Perkin Trans. 2 1996, 2615.
6. Kinbara, K.; Harada, Y.; Saigo, K. J. Chem. Soc., Perkin
Trans. 2 2000, 1339.
7. Pickard, S. T.; Smith, H. E. J. Am. Chem. Soc. 1990, 112,
5741.
8. Chen, C. P.; Prasad, K.; Repic, O. Tetrahedron Lett.
1991, 32, 7175.
9. Sakai, K.; Hashimoto, Y.; Kinbara, K.; Saigo, K.;
Murakami, H.; Nohira, H. Bull. Chem. Soc. Jpn. 1993,
66, 3414.
2
2
the p-toluenesulfonamide derivative of (−)-3 (1.85 g,
1
8
6.4%) as an oil: H NMR (CDCl ) l 1.41 (d, 3H), 2.22
3
(
s, 3H), 2.38 (s, 3H), 4.43 (quint., 1H), 4.81 (d, 1H),
.80–7.23 (m, 6H), 7.62 (d, 2H).
6
The p-toluenesulfonamide (1 g, 3.46 mmol) was dis-
solved in diethyl ether, LiOH (124 mg, 5.2 mmol) and
10. Chapman, D. T.; Crout, D. H. G.; Mahmoudian, M.;
Scopes, D. I. C.; Smith, P. W. J. Chem. Soc., Chem.
Commun. 1996, 21, 2415.
11. Bringmann, G.; Geisler, J. P.; Geuder, T.; K u¨ nkel, G.;
Kinzinger, L. Liebigs Ann. Chem. 1990, 795.
12. Raban, M.; Moulin, C. P.; Lauderback, S. K.; Swilley, B.
Tetrahedron Lett. 1984, 25, 3419.
2,4-dinitrobenzenesulfenyl chloride (1.22 g, 5.2 mmol)
were added and the mixture heated at reflux for 2 h.
Subsequent filtration and concentration of the filtrate
gave a solid residue (1.65 g), which was purified by
column chromatography on silica gel (el. cyclohexane/
ethyl acetate 9/1) yielding a 2.7:1 mixture of the two
diastereomeric N-(2,4-dinitrobenzenesulfenyl)-4-methyl-
13. Ingersoll, A. W. Org. Synth. 1943, Coll. Vol. 2, 503.
.