Highly efficient resolutions with isopropylidene glycerol 3-carboxy-2-naphthoate

Article abstract of DOI:10.1016/S0957-4166(02)00608-0

A number of chiral 1-arylethylamines and 1-alkylethylamines were resolved with the 3-carboxy-2-naphthoate of isopropylidene glycerol 2, previously reported to be an even more efficient resolving agent for 1-phenylethylamine than the corresponding hemiphthalate 1. The results obtained for the 1-arylethylamines confirm such a trend, revealing impressive resolution ability, in particular, for 1-(4-bromophenyl)-, 1-(4-nitrophenyl)- and 1-(2-naphthyl)ethylamine, whose enantiomers were almost quantitatively separated with (S)-2 by a single precipitation of the less soluble (S,S) diastereomeric salt. Additionally, the success of the resolutions of 1-alkylethylamines (1-phenyl-2-propylamine, 1-cyclohexylethylamine and 2-butylamine), which could not be resolved with 1, indicates that the novel carboxy ester 2 has a wider range of application than 1.

Full text of DOI:10.1016/S0957-4166(02)00608-0

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 13 (2002) 2277–2282
Highly efficient resolutions with isopropylidene glycerol
3-carboxy-2-naphthoate
Marco Pallavicini,* Cristiano Bolchi, Laura Fumagalli, Ermanno Valoti and Luigi Villa
Istituto di Chimica Farmaceutica e Tossicologica, Universita` di Milano, viale Abruzzi 42, I-20131 Milano, Italy
Received 4 September 2002; accepted 25 September 2002
Abstract—A number of chiral 1-arylethylamines and 1-alkylethylamines were resolved with the 3-carboxy-2-naphthoate of
isopropylidene glycerol 2, previously reported to be an even more efficient resolving agent for 1-phenylethylamine than the
corresponding hemiphthalate 1. The results obtained for the 1-arylethylamines confirm such a trend, revealing impressive
resolution ability, in particular, for 1-(4-bromophenyl)-, 1-(4-nitrophenyl)- and 1-(2-naphthyl)ethylamine, whose enantiomers were
almost quantitatively separated with (S)-2 by a single precipitation of the less soluble (S,S) diastereomeric salt. Additionally, the
success of the resolutions of 1-alkylethylamines (1-phenyl-2-propylamine, 1-cyclohexylethylamine and 2-butylamine), which could
not be resolved with 1, indicates that the novel carboxy ester 2 has a wider range of application than 1. © 2002 Elsevier Science
Ltd. All rights reserved.
1. Introduction
the racemic amines in methanol. In the case of 2-butyl-
amine, the 2-propanol solvent was replaced with
methanol and, for the recrystallization of the precipi-
tate, ethanol was used due to the high solubility of the
diastereomeric salts of this amine with (S)-2 in
methanol.
It was recently shown that isopropylidene glycerol 3-
carboxy-2-naphthoate 2 resolves 1-phenylethylamine
even more efficiently than the corresponding hydrogen
phthalate 1. Additionally it was found that 2 has more
suitable properties for use as a resolving agent than 1.1
On the basis of these results, it seemed worthwhile to
test its ability to resolve other racemic amines. Firstly,
we selected 1-arylethylamines 3–6 (Fig. 1), which had
previously been resolved with high efficiencies using 1,2
with the aim of confirming the further improvement of
resolution ability. Secondly, we considered 1-alkylethyl-
amines 7–9, whose resolution with 1 was unsuccessful,3
in order to verify the greater versatility of the new acid,
in particular with respect to substrates harder to resolve
than 1-arylethylamines.
As shown in Table 1 (entry 1), treatment of 3 with
equimolar (S)-2 resulted in the near-quantitative pre-
cipitation of the (S,S) salt, which contained (S)-3 with
>99.5% e.e. and was so insoluble that its recrystalliza-
tion was unfeasible, even when using large volumes of
boiling methanol. Such an excellent resolution
efficiency allowed the isolation of (R)-3 with 94% e.e.
from the (S,R) salt quantitatively recovered by simple
concentration of the mother liquors (Table 1, entry 2).
The enantiomeric excesses were accurately determined
by reversed-phase chiral HPLC according to the proce-
dure described previously for the resolution of the same
amine with (S)-1.2
2. Results
The results from the resolutions of 3–9 with (S)-2 are
summarized in Table 1, where the yields and the e.e.s
are reported.
The resolution of 1-(4-nitrophenyl)ethylamine with (S)-
2 was also very efficient (Table 1, entries 3 and 4):
precipitation of the (S,S) salt was immediate, quantita-
tive and highly diastereoselective, allowing both the S
and R enantiomers of the amine to be isolated, from
the precipitate and the mother liquors respectively, with
93.1% and 97.2% e.e. In order to maximise their recov-
ery from the respective salts with the resolving agent,
we decided to isolate them as the neutral sulphate salts
As with our preceding resolution of 1-phenylethyl-
amine,1 we decided to utilise the S enantiomer of 2 and
to combine the latter with stoichiometric quantities of
* Corresponding author. E-mail: marco.pallavicini@unimi.it
0957-4166/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(02)00608-0

2278
M. Palla6icini et al. / Tetrahedron: Asymmetry 13 (2002) 2277–2282
Figure 1.
Table 1. Resolutions of amines 3–9 by selective crystallization of the salts with (S)-2 from methanola
Entry
Compd.
Yield (%)b
E.e. (%)c
E (%)d
Crude salt with Recrystallized salt with
Recovered
substrate
Crude salt with
(S)-2
Recrystallized salt with
(S)-2
(S)-2e
(S)-2
1
2
3
4
5
6
7
8
9
(S)-3
(R)-3
(S)-4
(R)-4
(S)-5
(R)-5
(R)-6
(S)-7
(S)-8
(S)-9
90
–
–
–
–
80.5
–
–
–
–
80.5
95.9
85.9
86.9
78.9
88.6
44.2
70.2
61.0
26.2
\99.5
94.0
93.1f
97.2g
91.6
94.4
97.0
78.0
94.1
43.5
–
–
–
–
97.6
–
–
–
–
90
97
94
94
90
92
45
60
58
38
102.7
101.1
96.7
98.1
97.1
46.9
77.1
61.7
88.4
10
27.3
77.9
a The crude salt of (S)-2-butylamine with (S)-2 (entry 10) was precipitated from 2-propanol and recrystallized from ethanol.
b Relative to the theoretical amount, i.e. half of the starting racemate.
c Enantiomeric excess (determined by HPLC) of the amine liberated from the salt.
d Experimental resolution efficiency calculated from the yield of the crude diastereomeric salt and the enantiomeric excess of the amine liberated
from the salt.
e Precipitated from the solution containing equivalent amounts of racemic amine and (S)-2 or, in the case of (R)-3, (R)-4 and (R)-5 (entries 2,
4 and 6), recovered by concentration of the methanolic solution remaining from the isolation of the crude salt of (S)-2 with the respective S
amine.
f 97.3% after conversion into the neutral sulphate salt.
g 98.9% after conversion into the neutral sulphate salt.
by extraction with equivalent amounts of dilute
aqueous sulphuric acid. The concentration of the
aqueous extracts gave crude (S)-4 and (R)-4. Successive
treatment with boiling ethanol further enhanced their
enantiomeric excesses to 97.3 and 98.9%, respectively,
with minimal reduction of yield. Analytical determina-
tions of the enantiomeric compositions were success-
fully accomplished by HPLC on Crownpack chiral
stationary phase achieving a sensibly higher accuracy
than by the previously reported NMR method in the
presence of Eu(hfc)3.2
The resolution of 5 (Table 1, entries 5 and 6) resembled
those of 3 and 1-(4-nitrophenyl)ethylamine leading to

M. Palla6icini et al. / Tetrahedron: Asymmetry 13 (2002) 2277–2282
2279
the quantitative precipitation of the (S,S) salt with
>91% e.e. Consequently, the (S,R) salt was obtained
from the mother liquor in analogous chemical yield and
diastereomeric purity. Again, recrystallization of the
highly insoluble crude (S,S) diastereomeric salt was
unfeasible. Nevertheless, the enantiomeric excess of the
S amine could be further improved by prolonged tritu-
ration with boiling methanol.
dipeptide (R)-phenylglycyl-(R)-phenylglycine6 or 2-
naphthylglycolic acid,7 resolve the same amines with
much lower efficiency than 1-arylethylamines. This indi-
cates that 1-alkylethylamines are a more severe chal-
lenge for such resolving agents than 1-arylethylamines,
which is probably due to the absence or different
position of the aromatic residue. In particular, the
resolution of 2-butylamine 9 further substantiates the
greater versatility of 2, because the enantiomers of 9 are
particularly difficult to separate via diastereomeric salt
formation, as demonstrated by the limited number of
reported procedures.
The opposite stereochemical outcome was observed for
the resolution of amine 6, which was successful, but not
as efficient as those of substrates 3–5 (Table 1, entry 7).
In fact, (R)-6 was recovered with very high enan-
tiomeric excess (97%) from the crude (S)-2·(R)-6 pre-
cipitate, which, however, separated in a moderate 47%
yield. The use of smaller volumes of methanol led to
lower resolution efficiency, indicating that the initially
adopted solvent/salt ratio (4.9 ml/g) could be consid-
ered nearly optimal. On the other hand, alternative
crystallizations effected in ethanol or 2-propanol
showed a modest diastereoselectivity.
In summary, we have shown that 2 resolves (a) some
1-arylethylamines with extraordinary efficiency, effect-
ing quantitative separation of the two enantiomers by a
single precipitation of the less soluble diastereomeric
salt, and (b) some 1-alkylethylamines with lower, but
always workable efficiency. On the basis of these
results, superior resolving power and improved range of
application can be attributed to the novel acid 2, com-
pared with the parent hemiphthalate 1.
Amines 7 and 8 were resolved by (S)-2 with 60 and 58%
efficiency, respectively (Table 1, entries 8 and 9). The
enantiomeric excesses of the S isomers liberated from
the precipitates were determined by chiral HPLC after
conversion into the benzamide derivatives.
4. Experimental
4.1. General
Finally, crystallization of the diastereomeric salts of
2-butylamine with (S)-2 were performed in 2-propanol
and ethanol, due to their high solubility in methanol
(Table 1, entry 10). After the recrystallization, the S
amine was isolated from the salt with (S)-2 as the
hydrochloride salt (S)-9 with 77.9% e.e. (as determined
by chiral HPLC analysis of the benzamide) and in 26%
yield.
1H NMR spectra were recorded on a Varian Gemini
300 (300 MHz) instrument. Melting points were
recorded on a Bu¨chi Melting Point B-450 apparatus
and are uncorrected. Optical rotations were measured
in a 1-dm cell of 1 ml capacity using a Perkin Elmer 241
polarimeter. HPLC analyses were performed on Chiral-
cel and Crownpack columns from Daicel using a
Waters 510 pump, a Hitachi L-7400 UV detector and a
Hitachi D-7000 HPLC System Manager software.
Some of these experiments were performed using recy-
cled resolving agents recovered in nearly quantitative
yield and with unaltered chemical and enantiomeric
purity from the precipitates and the mother liquors of
previous resolutions.
Starting materials: The resolving agent (S)-2 was pre-
pared from (R)-isopropylidene glycerol and 2,3-naph-
thalic anhydride, as previously reported.1 (RS)-6 and
(RS)-2-butylamine were obtained from commercial
suppliers. (RS)-7 was synthesized from racemic 1-
phenyl-2-propanol by tosylation, conversion to the
azide and reduction. (RS)-3, (RS)-5 and (RS)-8 were
synthesized by the Leuckart reaction from the corre-
sponding ketones according to the experimental
procedure described for 1-phenylethylamine.8 (RS)-1-
(4-Nitrophenyl)ethylamine was prepared from ace-
tophenone by the method previously reported.9
3. Discussion and conclusion
As previously described, replacement of the phenyl
group of 1 with a naphthyl moiety led to the new acid,
carboxy naphthoate 2, which resolved 1-phenylethyl-
amines with very high efficiency. Such a trend is con-
firmed by the results from the resolutions of 1-arylethyl-
amines 3–6 with (S)-2 reported herein. In particular,
amines 3–5 were all resolved with efficiencies higher
than 90%, while, in the corresponding resolutions with
1, the efficiencies were ca.70% for 3 and 4, and 50% for
5.4 The same comparison is in favour of 2 for the
resolution of 6 (45% versus 23%),5 which is significantly
less efficient with both resolving agents. Moreover, in
contrast to the hemiphthalate 1, the carboxynaphthoate
2 was able to resolve the structurally heterogeneous
1-alkylethylamines 7–9. The efficiencies of the resolu-
tions of 7 and 8, 60 and 58%, respectively, with 2 can
be appreciated considering that other acids, like the
4.2. (S)-1-(4-Bromophenyl)ethylamine, (S)-3
The acid (S)-2 (14.36 g, 43.5 mmol) and (RS)-3 (8.7 g,
43.5 mmol) were combined in methanol (230 ml). A
white precipitate immediately separated. The suspen-
sion was heated to boiling temperature for 10 min and
then cooled to 25°C and filtered. The isolated solid was
rinsed with methanol twice and dried to give (S)-2·(S)-3
salt (10.38 g, 90%): mp 171.5–172.5°C; e.e. of (S)-3
>99.5% (determined by HPLC of the amine liberated
from a sample of the salt on a Chiralcel OD-R column;
85/15 1 M NaClO4/CH3CN, 0.8 ml/min); 1H NMR

2280
M. Palla6icini et al. / Tetrahedron: Asymmetry 13 (2002) 2277–2282
(DMSO-d6) l 1.28 (s, 3H), 1.33 (s, 3H), 1.37 (d, 3H),
3.82 (dd, 1H), 4.06 (pseudo t, 1H), 4.19 (q, 1H), 4.24 (d,
2H), 4.36 (m, 1H), 7.39 (d, 2H), 7.52 (d, 2H), 7.58–7.62
(m, 2H), 7.98–8.05 (m, 3H), 8.28 (s, 1H). The salt was
decomposed by treatment with 2N H2SO4 and ethyl
acetate. The aqueous phase was separated, made alka-
line with NaOH and extracted with dichloromethane.
Removal of the solvent from the extract, previously
dried over Na2SO4, gave a colourless oil, which was
distilled under vacuum yielding (S)-3 (3.50 g, 81%) as a
pale yellow oil: [h]2D0=−21.0 (c 2.8, methanol); the e.e.
was identical to that previously determined for the
amine liberated from a sample of the salt with (S)-2; 1H
NMR (CDCl3) l 1.35 (d, 3H), 1.47 (bs, 2H), 4.09 (q,
1H), 7.22 (d, 2H), 7.44 (d, 2H).
column; pH 1.5 HClO4 aq., 0.6 ml/min). The salt was
decomposed by treatment with 1N H2SO4 (22.4 ml),
water (30 ml) and ethyl acetate (50 ml). The aqueous
phase was separated, washed with ethyl acetate twice
and concentrated. The resultant solid residue was dried,
refluxed with ethanol (60 ml), cooled to 5°C, and
filtered to give (R)-4 (4.3 g; 87%) as a white solid: mp
198°C; [h]2D0=+6.6 (c 1.07, 0.05 M NaOH); e.e.: 98.9%
(determined by HPLC as described for the amine liber-
1
ated from a sample of the salt with (S)-2); H NMR
identical to (S)-4.
4.6. (S)-1-(2-Naphthyl)ethylamine, (S)-5
The acid (S)-2 (27 g, 81.7 mmol) and (RS)-5 (14 g, 81.7
mmol) were combined in methanol (270 ml). The white
precipitate (20.11 g) was collected by filtration, rinsed
with methanol, dried, and refluxed with methanol (200
ml) for 5 h. After cooling to 35°C, (S)-2·(S)-5 salt (16.5
g, 81%) was isolated by filtration and rinsed with
methanol: mp 150.5–151.5°C; e.e. of (S)-5: 97.6% (92%
before the treatment with boiling methanol) (deter-
mined by HPLC of the amine liberated from the salts
on a Chiralcel OD-R column; 60/40 0.1 M NaClO4/
4.3. (R)-1-(4-Bromophenyl)ethylamine, (R)-3
The methanolic filtrate resulting from the isolation of
(S)-2·(S)-3 was concentrated to afford (S)-2·(R)-3
(11.84 g, 103%). The salt was decomposed in the same
way as reported for (S)-2·(S)-3 and the resultant crude
amine distilled under vacuum yielding (R)-3 (4.17 g,
96%): [h]2D0=+18.2 (c 2.5, methanol); e.e.: 94.0% (deter-
mined by HPLC under the same conditions as
1
CH3CN, 0.5 ml/min); H NMR (DMSO-d6) l 1.25 (s,
1
described for (S)-3); H NMR identical to (S)-3.
3H), 1.32 (s, 3H), 1.55 (d, 3H), 3.80 (dd, 1H), 4.02
(pseudo t, 1H), 4.21 (d, 2H), 4.33 (m, 1H), 4.46 (q, 1H),
7.45–7.70 (m, 5H), 7.85–8.05 (m, 7H), 8.27 (s, 1H). The
salt was decomposed by treatment with 2N H2SO4 and
ethyl acetate. The aqueous phase was separated, made
alkaline with NaOH, and extracted with ethyl acetate.
Removal of the solvent from the extract, previously
dried over Na2SO4 gave (S)-5 (5.52 g; 79%): [h]2D5=−
20.5 (c 2.5, ethanol); the e.e. was identical to that
previously determined for the amine liberated from a
4.4. (S)-1-(4-Nitrophenyl)ethylamine sulphate, (S)-4
The acid (S)-2 (15.27 g, 46.2 mmol) and (RS)-1-(4-
nitrophenyl)ethylamine (7.68 g, 46.2 mmol) were com-
bined in methanol (280 ml). A white precipitate
immediately separated. The suspension was heated to
boiling temperature for 10 min and then cooled to 28°C
and filtered. The isolated solid was rinsed with
methanol twice and dried to give the salt of (S)-1-(4-
nitrophenyl)ethylamine with (S)-2 (11.6 g, 101%): mp
173–174°C; the e.e. of the amine liberated from a
sample of the salt was 93.1% (by HPLC on a Crown-
pack CR(+) column; pH 1.5 HClO4 aq., 0.6 ml/min);
1H NMR (CD3OD+D2O) l 1.34 (s, 3H), 1.40 (s, 3H),
1.62 (d, 3H), 3.94 (dd, 1H), 4.17 (q, 1H), 4.25–4.40 (m,
2H), 4.46 (m, 1H), 4.55 (q, 1H), 7.58 (m, 2H), 7.68 (d,
2H), 7.93 (m, 2H), 8.06 (s, 1H), 8.23 (s, 1H), 8.29 (d,
2H). The salt was decomposed by treatment with 1N
H2SO4 (23.3 ml), water (30 ml) and ethyl acetate (50
ml). The aqueous phase was separated, washed with
ethyl acetate twice and concentrated. The resultant
solid residue was dried, refluxed with ethanol (60 ml),
cooled to 5°C, and filtered to give (S)-4 (4.25 g, 86%) as
a white solid: mp 198°C; [h]2D0=−6.9 (c 1.07, 0.05 M
NaOH); e.e.: 97.3% (determined by HPLC as described
for the amine liberated from a sample of the salt with
1
sample of the salt with (S)-2; H NMR (CDCl3) l 1.48
(d. 3H), 1.58 (bs, 2H), 4.28 (q, 1H), 7.48 (m, 3H), 7.80
(m, 4H).
4.7. (R)-1-(2-Naphthyl)ethylamine, (R)-5
The methanolic filtrate resulting from the isolation of
the first precipitate of (S)-2·(S)-5 (20.11 g) was concen-
trated to afford the salt of (R)-5 with (S)-2 (19.91 g,
97%): e.e. of (R)-5: 94.4% (determined by HPLC of the
amine liberated from a sample of the salt under the
analytical conditions described for (S)-5). The salt was
decomposed in the same way as reported above for
(S)-2·(S)-5 salt yielding (R)-5 (6.2 g, 88.6%): [h]2D5=
+19.9 (c 2.5, ethanol); the e.e. was identical to that
previously determined for the amine liberated from a
sample of the salt with (S)-2; 1H NMR identical to
(S)-5.
1
(S)-2); H NMR (D2O) l 1.50 (d, 3H), 4.50 (q, 1H),
7.50 (d, 2H), 8.10 (d, 2H).
4.8. (R)-1-(1-Naphthyl)ethylamine, (R)-6
4.5. (R)-1-(4-Nitrophenyl)ethylamine sulphate, (R)-4
The acid (S)-2 (14.9 g, 45 mmol) and (RS)-6 (7.7 g, 45
mmol) were combined in methanol (110 ml) at room
temperature. The white precipitate of (S)-2·(R)-6 salt
(5.3 g, 47%) was collected by filtration and rinsed with
cold methanol: mp 135–136°C; e.e. of (R)-6: 97.0%
(determined by HPLC of the amine liberated from a
sample of the salt on a Chiralcel OD-R column; 60/40
The methanolic solution remaining from the isolation
of the salt of (S)-2 with (S)-1-(4-nitrophenyl)ethylamine
was concentrated to afford the salt of (R)-1-(4-nitro-
phenyl)ethylamine with (S)-2 (11.1 g, 97%): e.e. of the
R amine: 97.2% (by HPLC on a Crownpack CR(+)

M. Palla6icini et al. / Tetrahedron: Asymmetry 13 (2002) 2277–2282
2281
0.1
M
NaClO4/CH3CN, 0.5 ml/min); 1H NMR
dichloromethane. Removal of the solvent from the
extract, previously dried over Na2SO4, gave a colour-
less oil, which was distilled under vacuum yielding
(S)-8 (2.27 g, 61.0%): [h]2D0=+3.8 (neat); the e.e. was
identical to that previously determined for the amine
liberated from a sample of the salt with (S)-2 and
(DMSO-d6) l 1.22 (s, 3H), 1.30 (s, 3H), 1.60 (d, 3H),
3.82 (dd, 1H), 4.02 (pseudo t, 1H), 4.20 (d, 2H), 4.33
(m, 1H), 5.20 (q, 1H), 7.44 (m, 5H), 7.82 (d, 1H),
7.88 (d, 1H), 7.97 (m, 4H), 8.13 (d, 1H), 8.28 (s, 1H).
The salt was decomposed by treatment with 2N
H2SO4 and ethyl acetate. The aqueous phase was sep-
arated, made alkaline with NaOH and extracted with
ethyl acetate. Removal of the solvent from the
extract, previously dried over Na2SO4, gave an oil,
which was distilled under vacuum yielding (R)-6 (1.70
g, 44%): [h]2D0=+53.5 (c 2, ethanol); the e.e. was iden-
tical to that previously determined for the amine lib-
1
benzoylated; H NMR (CDCl3) l 0.80–1.25 (m, 8 H),
0.98 (d, 3H), 1.55–1.78 (m, 5H), 2.60 (m, 1H). The
filtrate from the crystallization of (S)-2·(S)-8 was con-
centrated and the residual salt decomposed in the
same way as described for (S)-2·(S)-8 to afford the
R-enriched amine (5.0 g): [h]2D0=−1.4 (neat); e.e.:
42.0% (determined by HPLC of the corresponding
benzamide).
1
erated from a sample of the salt with (S)-2; H NMR
(CDCl3) l 1.58 (d, 3H), 1.72 (bs, 2H), 4.97 (q, 1H),
7.50 (m, 3H), 7.67 (d, 1H), 7.76 (d, 1H), 7.89 (d, 1H),
8.13 (d, 1H).
4.11. (S)-2-Butylamine hydrochloride, (S)-9
The acid (S)-2 (13.8 g, 41.8 mmol) and (RS)-2-butyl-
amine (3.06 g, 41.8 mmol) were combined in 2-
propanol (50 ml) at room temperature. The white
precipitate (7.28 g) was collected by filtration, rinsed
with cold 2-propanol, dried, and recrystallized from
ethanol (7 ml) at 0°C yielding the salt of (S)-2-butyl-
amine with (S)-2 (2.3 g, 27%): mp 106–107°C; e.e. of
(S)-2-butylamine: 77.9% (43.5% before the recrystal-
lization) (determined by HPLC of the corresponding
benzamides on a Chiralcel OB column; hexane/2-
propanol 95/5, 1.2 ml/min); 1H NMR (DMSO-d6) l
0.88 (t, 3H), 1.17 (d, 3H), 1.28 (s, 3H), 1.33 (s, 3H),
1.45 (m, 1H), 1.64 (m, 1H), 3.05 (m, 1H), 3.82
(pseudo t, 1H), 4.04 (pseudo t, 1H), 4.19 (d, 2H),
4.36 (m, 1H), 7.52 (m, 2H), 7.85–8.0 (m, 3H), 8.21 (s,
1H). The salt was decomposed by treatment with 10%
HCl and ethyl acetate. The aqueous phase was sepa-
rated, washed with ethyl acetate, and concentrated to
give (S)-9 (0.60 g, 26%) as a white solid: mp 145–
147°C; [h]3D7=−2.6 (c 1, ethanol); the e.e. was identi-
cal to that previously determined for the benzamide
of the amine liberated from a sample of the recrystal-
4.9. (S)-1-Phenyl-2-propylamine, (S)-7
The acid (S)-2 (5.01 g, 15.2 mmol) and (RS)-7 (2.05
g, 15.2 mmol) were combined in methanol (20 ml) at
room temperature. After cooling to 5°C, the white
precipitate of (S)-2·(S)-7 salt (2.72 g, 77%) was col-
lected by filtration and rinsed with cold methanol: mp
141–142°C; e.e. of (S)-7: 78.0% (determined by HPLC
of the corresponding benzamide on a Chiralcel-OB
1
column, hexane/2-propanol 9/1, 2 ml/min); H NMR
(DMSO-d6) l 1.08 (d, 3H), 1.24 (s, 3H), 1.30 (s, 3H),
2.62 (dd, 1H), 3.08 (dd, 1H), 3.37 (m, 1H), 3.80 (dd,
1H), 4.02 (pseudo t, 1H), 4.20 (d, 2H), 4.33 (m, 1H),
7.15–7.35 (m, 5H), 7.55 (m, 2H), 7.90–8.02 (m, 3H),
8.25 (s, 1H). The salt was decomposed by treatment
with 2N H2SO4 and ethyl acetate. The aqueous phase
was separated, made alkaline with NaOH, and
extracted with dichloromethane. Removal of the sol-
vent from the extract, previously dried over Na2SO4,
gave a colourless oil, which was distilled under vac-
uum yielding (S)-7 (0.72 g, 70%): [h]2D1=+25.4 (c 2.85,
methanol); [h]2D0=+30.5 (c 1.8, benzene); the e.e. was
identical to that previously determined for the amine
liberated from a sample of the salt with (S)-2 and
1
lized salt with (S)-2; H NMR (DMSO-d6) l 0.86 (t,
3H), 1.15 (d, 3H), 1.45 (m, 1H), 1.62 (m, 1H), 3.01
(m, 1H), 8.12 (bs, 3H).
1
benzoylated; H NMR (CDCl3) l 1.11 (d, 3H), 1.25
(bs, 2H), 2.50 (dd, 1H), 2.72 (dd, 1H), 3.16 (m, 1H),
7.15–7.33 (m, 5H).
Acknowledgements
4.10. (S)-1-Cyclohexylethylamine, (S)-8
The acid (S)-2 (19.33 g, 58.5 mmol) and (RS)-8 (7.44
g, 58.5 mmol) were combined in methanol (70 ml) at
room temperature. After cooling to 5°C, the white
precipitate of (S)-2·(S)-8 salt (8.26 g, 62%) was col-
lected by filtration and rinsed with cold methanol: mp
146–147°C; e.e. of (S)-8: 94.1% (determined by HPLC
of the corresponding benzamide on a Chiralcel-OD
column; hexane/2-propanol 9/1, 0.7 ml/min); 1H
NMR (DMSO-d6) l 0.85–1.80 (m, 11H), 1.10 (d, 3H),
1.28 (s, 3H), 1.33 (s, 3H), 2.94 (m, 1H), 3.82 (pseudo
t, 1H), 4.04 (pseudo t, 1H), 4.20 (d, 2H), 4.37 (m,
1H), 7.52 (m, 2H), 7.85–8.0 (m, 3H), 8.22 (s, 1H).
The salt was decomposed by treatment with 10% HCl
and ethyl acetate. The aqueous phase was separated,
made alkaline with NaOH, and extracted with
This work was supported by the Ministero dell’Uni-
versita` e della Ricerca Scientifica e Tecnologica of
Italy.
References
1. Pallavicini, M.; Valoti, E.; Villa, L.; Piccolo, O. Tetra-
hedron: Asymmetry 2001, 12, 2489.
2. Pallavicini, M.; Valoti, E.; Villa, L.; Piccolo, O. Tetra-
hedron: Asymmetry 1996, 7, 1117.
3. Pallavicini, M.; Valoti, E.; Villa, L.; Piccolo, O. Tetra-
hedron: Asymmetry 1997, 8, 1069.

2282
M. Palla6icini et al. / Tetrahedron: Asymmetry 13 (2002) 2277–2282
4. The experimental efficiencies of the resolutions with (S)-1
are calculated from the chemical yields and the enan-
tiomeric excesses reported in Ref. 2 for the same sub-
strates. In the case of 5, the comparison is made with the
resolution with (S)-1 in 2-propanol, because no precipita-
tion occurred in methanol.
resolution of 6 with (S)-1 in 2-propanol is described and
its efficiency is about 67%.
6. Akazome, M.; Matsuno, H.; Ogura, K. Tetrahedron:
Asymmetry 1997, 8, 2331.
7. Kinbara, K.; Harada, Y.; Saigo, K. J. Chem. Soc., Perkin
Trans. 2 2000, 1339.
8. Ingersoll, A. W. Org. Synth. 1943, Coll. Vol. 2, 503.
9. Pallavicini, M.; Valoti, E.; Villa, L.; Piccolo, O. J. Chem.
Res. (S) 1996, 330.
5. The 23% value is calculated from the chemical yield and
the enantiomeric excess reported in Ref. 2 for the resolu-
tion with (S)-1 in methanol. In the same reference, the