Biselector enantioselective stationary phases for HPLC: Dependence of the chiral discrimination properties on stereochemistry and chemical nature of each unit of the chiral auxiliary

Article abstract of DOI:10.1016/S0957-4166(02)00402-0

Optically active 1-(1-naphthyl)ethylamine, N-3,5-dinitrobenzoylphenylglycine and N-3,5-dinitrobenzoylleucine were used as chiral building blocks to prepare three new enantiomerically pure bifunctional chiral auxiliaries for enantioselective HPLC, belongin

Full text of DOI:10.1016/S0957-4166(02)00402-0

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 13 (2002) 1805–1815
Biselector enantioselective stationary phases for HPLC:
dependence of the chiral discrimination properties on
stereochemistry and chemical nature of each unit of the chiral
auxiliary
Anna Iuliano, Emanuele Attolino and Piero Salvadori*
ICCOM-CNR-Sezione di Pisa, Dipartimento di Chimica e Chimica Industriale, via Risorgimento 35, 56126 Pisa, Italy
Received 10 July 2002; accepted 12 July 2002
Abstract—Optically active 1-(1-naphthyl)ethylamine, N-3,5-dinitrobenzoylphenylglycine and N-3,5-dinitrobenzoylleucine were
used as chiral building blocks to prepare three new enantiomerically pure bifunctional chiral auxiliaries for enantioselective
HPLC, belonging to a family of biselector systems, the first example of which (CSP1) has been described previously. These
compounds were covalently linked to silica gel to produce three chiral stationary phases (CSPs 2–4), whose enantiodiscriminating
capability towards the HPLC resolution of selected racemic compounds was assessed. The obtained results allowed us to establish
the influence of the stereochemistry and/or the chemical structure of each chiral moiety of the biselector system on their
enantiorecognition properties. © 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
co-workers,3 who prepared some independent CSPs,
showing good effectiveness in the resolution of a wide
range of racemic compounds.4 Recently, we faced this
problem directing our attention towards the prepara-
tion of a novel chiral auxiliary, obtained by linking two
different selectors [(S)-NEA and (S)-Leu] to one
another, that, once attached to silica gel through the
s-triazine moiety, afforded a so-called biselector CSP.5
Because of the complementarity of the two selectors in
the enantiodiscrimination of different chiral com-
pounds, CSP1 (Fig. 1) had wider applicability than the
two monoselector CSPs. Indeed, CSP1 did work as a
biselector CSP5 since it was able to separate the enan-
tiomers of racemic compounds resolved by Pirkle’s
(S)-Leu CSP6—reproduced in the derivatized amino
acid moiety of CSP1—(Fig. 1) as well as those enan-
tiodiscriminated by Oi’s CSP7—reproduced in the s-tri-
azine-(S)-1-(1-naphthyl)ethylamine moiety—(Fig. 1). In
addition, it was able to resolve racemic compounds
which are not enantiodiscriminated by either of the
CSPs presented by Pirkle and Oi.5
Selective molecular interactions are the essence of vital
biochemical processes, such as molecular transport,
genetic information processing and protein assembly.
An elucidation of the phenomena which govern these
interactions is important not only for the understanding
of these processes but also for their manipulation.
Studies in molecular recognition have given chemists
the tools needed to design, build and evaluate enan-
tioselective molecular systems to be used for synthetic,
separative and analytical purposes. In this field, a great
deal of attention has been devoted to chiral stationary
phases (CSPs), which allow the direct liquid chromato-
graphic separation of enantiomers. Since this has been
recognized as one of the most powerful means for the
determination of the enantiomeric composition of chi-
ral compounds, the preparation of a CSP having high
efficiency and versatility has became more and more
important.1 This problem has been approached in a
rational fashion in the case of independent CSPs,2
whose chiral recognition mechanism is fairly well
understood,2 mainly due to the work of Pirkle and
Prompted by these results, we decided to evaluate
whether the two portions of the biselector system are
independent, or if changing of the stereochemistry and/
or chemical structure of one chiral moiety influences the
enantiodiscriminating capability of the other. This
* Corresponding author. Tel.: +39-50-918203; fax: +39-50-918409;
e-mail: psalva@dcci.unipi.it
0957-4166/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(02)00402-0

1806
A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
Figure 1. Structure of CSPs 1–4, Pirkle CSPs (A) and Oi’s CSP (B).
could afford some useful insights for the design of
‘broad spectrum’ selectors capable of discriminating the
enantiomers of several structurally different com-
pounds, and could also give further information about
the mechanisms governing enantiorecognition. To this
end, we synthesized the three CSPs reported in Fig. 1:
CSP2 was obtained starting from (R)-NEA and (S)-
Leu, i.e. its selector is in a diastereoisomeric relation-
ship to that of CSP1, whereas CSP3 and CSP4 derive
from (S)-NEA and (S)-phenylglycine and (R)-NEA
and (S)-phenylglycine, respectively, i.e. they constitute
pairs of diastereoisomeric CSPs.
trobenzoyl amino acid in the presence of EEDQ:9 the
chiral auxiliaries 9a–c were obtained in 35–38% yield
after chromatographic purification.
The three selectors were covalently linked to silica gel
by means of the terminal double bond.10 They were
treated with a five-fold excess of 3-mercaptopropyl-
trimethoxysilane, in the presence of a catalytic amount
of AIBN in refluxing CHCl3, affording the correspond-
ing chemically pure silanes in quantitative yield. The
derivatized silica materials were obtained by reacting
the silanes 10a–c with silica gel in refluxing of toluene
and were slurry packed in stainless steel columns using
conventional techniques. The loading of the organic
material on silica gel, determined by elemental analysis,
ranges from 0.23 to 0.29 mmol/g for the three different
CSPs and is comparable to that found for CSP1.
The aim of the present work is to compare the enan-
tiodiscriminating capability of the four CSPs in the
HPLC resolution of selected racemic compounds, in
order to determine the influence of both the chemical
structure and the stereochemistry of the two chiral
components of the biselector system on the enan-
tiorecognition properties of the phase.
2.2. Comparison of the enantiodiscriminating capability
of CSPs 1–4 in the separation of enantiomers
The racemic compounds resolved by CSP1 (Chart 1)
were used to compare the enantiodiscrimination prop-
erties of CSPs 1–4. Table 1 reports on the chromato-
graphic resolution of p-acceptor racemates 11–14. Their
enantiodiscrimination on CSP1 has been attributed to
the presence, in the structure of the biselector system, of
the p-donor 1-(1-naphthyl)ethylamino-s-triazine moi-
ety,5 which reproduces the structure of Oi’s CSP.7 The
same moiety is present on CSPs 2–4, therefore the
comparison of the data concerning the resolution of
this class of racemates, allows us to obtain some infor-
mation about the influence of the chemical structure
and the stereochemistry of the amino acid moiety on
the chiral recognition exhibited by the 1-(1-naph-
thyl)ethylamine-s-triazine fragment.
2. Results and discussion
2.1. Synthesis of the CSPs
CSPs 2–4 were synthesized according to the method
used for preparing CSP1:5 Scheme 1 shows the general
synthetic route to the three CSPs. The three chlorine
atoms of s-trichlorotriazine were displaced in succes-
sion by N-tert-butoxycarbonyl ethanolamine, optically
active 1-(1-naphthyl)ethylamine and allylamine, afford-
ing the trisubstituted s-triazine scaffold 7. The first two
nucleophilic displacements were carried out in the same
reaction vessel,8 adding the second nucleophile, 1-(1-
naphthyl)ethylamine, when complete conversion of s-
triclorotriazine had occurred. The disubstituted
derivative 6, isolated and purified by column chro-
matography, was then treated with an excess of allyl-
As previously reported,5 CSP1 is able to resolve all the
p-acceptor racemic compounds 11–14 (Table 1): it was
found to be more versatile towards this class of race-
mates than Oi’s CSP, which showed enantiodiscrimina-
tion for only the 3,5-dinitrophenyl derivatives of
amine, affording chemically pure
quantitative yield. After removal of the BOC protecting
group, the free amine was reacted with the 3,5-dini-
7
in nearly

A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
1807
Scheme 1. Reagents and conditions: (a) N-BOCethanolamine, DIPEA, CH3CN, 60°C; (b) 1-(1-naphthyl)ethylamine, DIPEA,
25°C; (c) allylamine (4 equiv.), CH3CN, 50°C; (d) TFA, CH2Cl2, 25°C; (e) (S)-3,5-dinitrobenzoyl amino acid, EEDQ, THF, 25°C;
(f) 3-mercaptopropyltrimethoxysilane (5 equiv.), AIBN, CHCl3, reflux; (g) silica gel, toluene, reflux.
1-arylpropionic acids and (alkylaryl)methylamines.7
CSP2, which differs from CSP1 in the absolute
configuration at the stereogenic center of 1-(1-naph-
thyl)ethylamine, behaves in a rather different way. In
fact, CSP2 is able to resolve only the 3,5-DNB deriva-
tives of (alkylaryl)methylamines 12 and the 3,5-dini-
troanilide of ibuprofen 13 (entries 3–5). In contrast, no
separation is observed for the enantiomers of the 3,5-
dinitrobenzoyl derivatives of amino acid alkyl esters 14
(entries 6–16) and 4-nitrobenzamides 11 (entries 1 and
2); only a very poor resolution is obtained for the
phenylalanine derivative 14i (entry 14). These results
demonstrate that the stereochemical relationship
between the two chiral moieties of the biselector system
deeply influences the enantiodiscrimination properties
of the 1-(1-naphthyl)ethylamino-s-triazine fragment
towards p-acceptor racemic compounds: the (S)-amine/
(S)-amino acid, which is the chiral selector in CSP1,
behaves as a matched pair11 of this chiral auxiliary,
whereas the diastereoisomeric (R)-amine/(S)-amino
acid system, from which CSP2 has been obtained,
ehave as a mismatched pair.11
The influence of the chemical structure of the deriva-
tized amino acid moiety on the enantiodiscriminating
capability of the 1-(1-naphthyl)ethylamine-s-triazine
fragment can be assessed by comparing the chromato-
graphic data obtained with CSP3, derived from (S)-
NEA and (S)-phenylglycine, with the results from the
separations on CSP1, derived from (S)-NEA and (S)-
Leu. The retention times of the analytes are higher on
CSP3 than on CSP1 in all the examined cases: this is
not surprising if one takes into account that the amino
acid portion of CSP3 possesses a phenyl group that can

1808
A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
Chart 1.
engender attractive interactions with the p-acceptor
racemic compounds.2 As far as the enantioseparations
are concerned, CSP3 resolves the N-3,5-dinitrobenzoyl
derivatives of (alkylaryl)methylamines 12 and the 3,5-
dinitroanilide of ibuprofen 13 (entries 3–5), although to
a reduced extent with respect to CSP1; in addition, few
amino acid derivatives and only one of the two 4-
nitrobenzamides are resolved on CSP3 with lower
enantioselectivity factors than those found for the same
compounds on CSP1. It also noteworthy that the
amino acid derivatives 14f–h and 14m are enan-
tiodiscriminated only if the alkyl ester group is more
complex than methyl (entries 11–13 and 16). Compari-
son of these chromatographic data suggests that the
enantiodiscriminating behavior of the 1-(1-naph-
thyl)ethylamino-s-triazine fragment depends on the
chemical structure of the amino acid moiety, as well as
its stereochemistry: in fact the resolving power of the
biselector system towards p-acid racemates, which is
good when the amino acid moiety is leucine (CSP1),
decreases when the amino acid partner is phenylglycine.
Changing the configuration at the stereogenic center of
the amino acid affects the enantiodiscriminating prop-
erties of the 1-(1-naphthyl)ethylamine-s-triazine also in
the case of the chiral auxiliaries containing the deriva-
tized phenylglycine. CSP4, prepared starting from the
diastereoisomer (R)-NEA/(S)-phenylglycine, shows
slightly better performances in the chromatographic
resolution of amino acid derivatives than CSP3. The

A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
1809
Table 1. Chromatographic resolutiona of p-acidic racemic compounds on CSPs 1–4
CSP 1
hc (e.o.)d
CSP 2
hc(e.o.)d
CSP 3
hc (e.o.)d
CSP 4
hc (e.o.)d
Entry
Compound
kI,b
2.02
kI,b
4.03
kI,b
3.73
kI,b
3.06
1
2
3
4
5
6
7
8
11a
11b
12a
12b
13
14a
14b
14c
14d
14e
14f
14g
14h
14i
1.10 (−)
1.38 (−)
1.28 (−)
1.51 (−)
1.49 (+)
1.10
1.30
1.19
1.30
1.27
1.44
1.39 (−)
1.64
1.27
1.26
1.00
1.00
1.14 (+)
1.50 (+)
1.35 (−)
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.05
1.00
1.01
1.00 (+)
1.09 (+)
1.13 (−)
1.11 (−)
1.34 (+)
1.00
1.00
1.00
1.00
1.00
1.19
1.13 (−)
1.13
1.00
1.00
1.00
1.00
1.09 (+)
1.30 (+)
1.35 (−)
1.00
1.10
1.00
1.00
1.00
1.15
1.12 (+)
1.13
1.18
1.10
3.01
4.42
3.40
2.00
10.07
3.64
1.93
2.57
3.06
1.77
1.44
2.13
4.22
10.78
5.05
4.27
5.60
3.31
2.63
13.45
3.70
1.86
2.56
3.32
1.67
1.59
1.10
4.12
4.34
1.88
3.89
5.39
4.97
2.68
10.13
4.73
2.12
2.96
3.50
1.88
1.94
1.39
5.29
5.25
2.79
4.35
5.85
3.80
2.66
12.50
4.68
2.52
3.64
4.30
2.33
2.36
1.70
6.11
6.23
3.45
9
10
11
12
13
14
15
16
14l
14m
1.50
1.14
1.18
a Chromatographic conditions: UV detection (u=254 nm), T=25°C, flow 1 ml/min, eluent: hexane–dichloromethane–propan-2-ol, 70:30:1.
b Retention factor of the first eluted enantiomer.
c Enantioselectivity factor.
d Sign of the circular dichroism at 254 nm of the first eluted enantiomer.
3,5-dinitrobenzoyl derivative of phenylalanine is
resolved even as methyl ester, 14i, and iso-butyl ester,
14l, (entries 14 and 15) and separation of the enan-
tiomers of the leucine derivative 14b is also observed
(entry 7).
azine moiety is affected by the stereochemistry of the
amino acid fragment to a greater extent when it is
leucine, than in the case of phenylglycine.
In order to establish whether the enantioseparation
properties of the derivatized amino acid moiety are
influenced by the stereochemistry of the 1-(1-naph-
thyl)ethylamine-s-triazine unit, some p-basic com-
pounds, such as the binaphthyl derivatives 15,
2,2,2-trifluoro-1-(9-anthryl)ethanol 16 and the amides
17 (Chart 1), i.e. compounds belonging to classes of
racemates resolved by Pirkle’s CSPs A,6 were analyzed
on CSPs 1–4.
As far as enantiodiscrimination of the amide derivatives
is concerned, it is difficult to establish a general trend.
In fact, the enantiomers of the 4-nitrobenzamide 11b
resolved by CSP3 (entry 2) are not separated by CSP4,
whereas only one of the two 3,5-dinitrobenzamides
(12b, entry 4) shows a higher enantioselectivity factor
on CSP4.
The data reported in Table 1 demonstrate that both the
nature and stereochemistry of the amino acid derivative
influence the enantiorecognition properties of the 1-(1-
naphthyl)ethylamino-s-triazine moiety in the biselector
systems of CSPs 1–4. Furthermore, the enantiodiscrimi-
nating capability of the 1-(1-naphthyl)ethylamino-s-tri-
The data from the chromatographic resolutions of
binaphthyl
derivatives
and
2,2,2-trifluoro-1-(9-
anthryl)ethanol are listed in Table 2. CSP1, as previ-
ously reported,5 shows good enantiodiscriminating
capability towards these racemates5 and the h values
show the same trend observed with the leucine-based
Table 2. Chromatographic resolutiona of binaphthylic derivatives and alcohol 16 on CSPs 1–4
CSP 1
CSP 2
CSP 3
CSP 4
Entry
Compound
kI,b
hc (e.o.)d
kI,b
hc (e.o.)d
kI,b
hc (e.o.)d
kI,b
3.55
10.45
3.45
2.31
6.92
hc
1
2
3
4
5
15a
15b
15c
15d
16e
2.91
7.87
2.87
1.64
4.78
1.22 (−)/R
1.36 (−)/R
1.20 (−)/R
1.16 (−)/R
1.19
5.76
6.43
2.19
4.84
4.52
1.09 (+)/S
1.12 (+)/S
1.03 (+)/S
1.03 (+)/S
1.00
3.23
8.10
3.11
2.08
6.15
1.20 (−)/R
1.28 (−)/R
1.18 (−)/R
1.13 (−)/R
1.35
1.00
1.00
1.00
1.00
1.16
a Chromatographic conditions: UV detection (u=254 nm), T=25°C, flow 1 ml/min, eluent: hexane–dichloromethane–propan-2-ol, 70:30:1.
b Retention factor of the first eluted enantiomer.
c Enantioselectivity factor.
d Sign of the circular dichroism at 240 nm/absolute configuration of the first eluted enantiomer.
e Eluent: hexane–dichloromethane–propan-2-ol, 75:20:5.

1810
A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
Pirkle CSP.6d When the same compounds were ana-
lyzed on CSP2, the enantioseparations were poorer in
all cases examined. In fact, although the enantiomers of
the binaphthyl derivatives are still separated on CSP2,
their h values are lower, whereas the carbinol 16 is not
resolved at all (entry 5).
when the amino acid is phenylglycine: in fact CSP2 is
still able to resolve binaphthyl derivatives whereas
CSP4 does not.
Data concerning the resolution of the last class of
racemates, i.e. p-basic amides 17 and 18, are listed in
Table 3. CSP1 is able to resolve all the compounds,
showing h values that depend on the nature of the
amide. On the contrary, its diastereoisomeric phase,
CSP2, is unable to resolve any of the p-basic amides of
Table 3, although the k% values, higher than those
found upon CSP1, suggest that the racemates-selector
interaction is not negligible on CSP2. Even CSP3 is
able to separate the enantiomers of p-basic amides
(Table 3), although to a lesser extent than CSP1 (the
sole exception is compound 17d that shows a higher h
value); furthermore, two amides are not resolved. The
trend observed going from CSP1 to CSP2 is confirmed
also with the diastereoisomeric couple CSP3/CSP4. In
fact, CSP4 resolves only three compounds, 17a–b and
17d (entries 1, 2 and 4) with lower enantioselectivity
factors than those observed on CSP3. Therefore, even
towards this class of racemates the change in stereo-
chemistry of the 1-(1-naphthyl)ethylamine-s-triazine
moiety influences the enantiodiscriminating capability
of the 3,5-dinitrobenzoyl amino acid fragment in the
same sense either when the amino acid is leucine or
phenylglycine. In fact, the matched pair11 towards the
resolution of p-basic amides for both the biselector
systems is the (S)-amine/(S)-amino acid, whereas the
(R)-amine/(S)-amino acid represents the mismatched
pair.11
The chromatographic separations on CSP3 (Table 2)
show that this phase retains compounds more than
CSP1, with binaphthyl derivatives or the alkylaryl-
carbinol. This behavior can be attributed to the change
in the amino acid structure: in this case the presence of
a phenyl instead of an iso-butyl group can engender
attractive p–p interactions4 with this class of racemates,
which result in higher k% values. The enantioselectivities
are good and comparable to those obtained with CSP1.
Only 16 exhibits a higher h value: this is not surprising
if we take into account that the leucine-based Pirkle
CSP shows lower enantioselectivity towards alkylaryl
carbinols than the analogous CSP containing phenyl-
glycine. The different behavior of CSP1 and CSP3
towards this compound, therefore, is in keeping with
that observed for the two different Pirkle’s CSPs.6e By
comparing the data obtained using CSP3 with those
found upon the diastereoisomeric CSP4 (Table 2) one
can observe that not only compounds 15a–d and 16 are
more retained than upon CSP3, but also more signifi-
cant differences are present as far as the enantioselectiv-
ities are concerned: in fact, CSP4 does not resolve any
binaphthyl derivative, whereas the enantiomers of 16
are separated with a lower h value. The results obtained
with these racemates demonstrate that the stereochem-
istry of the 1-(1-naphthyl)ethylamino-s-triazine moiety
influences the enantioresolution capability of the
derivatized amino acid fragment in the biselector sys-
tem. The matched pair11 towards the resolution of
binaphthyl derivatives and carbinol 16 is (S)-amine/(S)-
amino acid, whereas (R)-amine/(S)-amino acid repre-
sents the mismatched couple,11 independently of the
chemical nature of the amino acid. Furthermore, the
change in stereochemistry of the 1-(1-naph-
thyl)ethylamino-s-triazine moiety has a greater influ-
ence on the enantioselectivity of the biselector system
The analysis of compound 19 (Table 3) on CSPs 1–4 is
rather interesting. This compound shows the features of
a carbinol, resolved by the N-3,5-dinitrobenzoyl amino
acid based CSPs,6e and those of a 3,5-dinitrobenzamide,
a compound related to a class of racemates discrimi-
nated by the Oi’s CSP.7 This substrate is well resolved
on CSPs 1 and 3 whereas poor or no separation is
observed on CSPs 2 and 4. This result confirms that
(S)-amine/(S)-amino acid is the matched pair: in fact,
the synergistic action of the two moieties helps the
Table 3. Chromatographic resolutiona of p-basic amides and 19 on CSPs 1–4
CSP 1
CSP 2
CSP 3
CSP 4
Entry
Compound
kI,b
hc (e.o.)d
kI,b
hc (e.o.)d
kI,b
hc (e.o.)d
kI,b
hc (e.o.)d
1
2
3
4
5
6
7
9
17a
17b
17c
17de
17e
17f
18
6.14
7.54
4.86
2.06
1.86
1.95
2.50
7.96
1.27 (−)
1.26 (−)
1.07 (+)
1.15 (−)
1.40
1.37
1.21 (−)
1.41 (−)
7.89
9.64
6.05
3.65
2.63
2.67
3.26
7.11
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.10 (+)
6.64
9.73
2.03
2.23
1.93
1.91
2.49
7.75
1.17 (+)
1.19 (+)
1.04 (−)
1.31 (−)
1.00
1.00
1.11 (−)
1.44 (−)
7.59
10.46
4.95
2.51
2.19
2.17
2.86
7.04
1.07 (+)
1.07 (+)
1.00
1.09 (−)
1.00
1.00
1.00
1.00
19e
a Chromatographic conditions: UV detection (u=254 nm), T=25°C, flow 1 ml/min, eluent: hexane–dichloromethane–propan-2-ol, 90:10:1.
b Retention factor of the first eluted enantiomer.
c Enantioselectivity factor.
d Sign of the circular dichroism at 254 nm (230 nm for the naphthalene containing compounds) of the first eluted enantiomer.
e Eluent: hexane–dichloromethane–propan-2-ol, 75:20:5.

A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
1811
The elution order of the binaphthyl derivatives is
shown in Table 2: the measurements were performed at
240 nm, a wavelength corresponding to the low-energy
component of the exciton couplet present in the CD
spectra of these derivatives. On the basis of the CD
sign,13 by using the ECCD method,14 it is possible to
determine the absolute configuration of the eluted
enantiomers. Since the elution order of these derivatives
on the Pirkle’s phases is known6d it is possible to
compare it with those obtained upon CSPs 1–4, which
allows us to verify if the enantiorecognition mechanism
of the derivatized amino acid moiety inserted in the
biselector system of the CSPs 1–4 is the same as in the
Pirkle’s CSPs or not. All the CSPs possess a (S)-
configured amino acid moiety. The elution order
observed on CSPs 1 and 3 (Table 2), constituted by the
(S,S)-couple, is the same as the (S)-configured Pirkle
CSPs,6d whereas it is opposite on CSP2 (Table 2),
whose absolute configuration at the stereogenic center
of the 1-(1-naphthyl)ethylamino-s-triazine moiety is
(R). This result suggests that changing the stereochem-
istry of the 1-(1-naphthyl)ethylamino-s-triazine moiety
dramatically affects the enantiorecognition properties
of the derivatized amino acidic fragment: when the
amino acid is phenylglycine the couple (S)-amino acid/
(R)-amine is unable to separate the enantiomers of the
binaphthyl derivatives (Table 2), in the case of leucine a
change in the elution order (Fig. 2) together with lower
enantioselectivity with respect to the (S)-amino acid/
(S)-amine couple (Table 2) is observed. This indicates
that the (S)-3,5-dinitrobenzoylleucine moiety coupled
to (R)-1-(1-naphthyl)ethylamine in the biselector system
resolves the binaphthyl derivatives by means of a differ-
ent enantiorecognition mechanism with respect to that
operating in the case of the Pirkle’s CSP, whose chiral
selector has the same absolute configuration.
resolution of 19 affording a higher h value than that
obtained when the two moieties of the biselector sys-
tem are mismatched.
2.3. Analysis of the elution orders on CSPs 1–4
The comparison of the chromatographic data obtained
on CSPs 1–4 has shown that the behavior of these CSPs
depends on the relative stereochemistry of the two
chiral moieties of the biselector system as well as on the
chemical structure of the amino acidic fragment. This
should be attributable to some differences in the enan-
tiorecognition mechanism exhibited by these CSPs. In
order to gain some insights about this point we have
determined the elution order of the racemic compounds
resolved upon the four biselector CSPs. In fact, if a
class of racemates exhibits the same elution order on
different CSPs whose chiral moiety responsible of its
enantiodiscrimination has the same stereochemistry (for
example compounds listed in Table 1 upon CSPs 1 and
3), we can reasonably conclude that these CSPs exhibit
a similar enantiorecognition mechanism towards the
racemic compounds. In other words, this means that
the two chiral moieties of the biselector system are not
influenced by each other so much that the enan-
tiorecognition mechanism changes.
The elution orders were determined by means of a CD
detector:12 in order to make a comparison, only the
data concerning compounds which are resolved upon
all (or the majority) of the CSPs have been reported.
The measurements were performed at 254 nm in the
case of p-acidic compounds and the results are listed in
Table 1. CSP1 and CSP2, which have opposite abso-
lute configuration at the stereogenic center of the 1-(1-
naphthyl)ethylamino-s-triazine moiety, show opposite
elution orders for the racemic compounds resolved by
both phases (entries 3, 4 and 6). Since the chromato-
graphic resolution of these compounds is attributable
to this moiety, the change in the elution order is only
due to the different stereochemistry of this fragment in
CSP1 and CSP2 and, therefore, the stereochemistry of
the derivatized amino acid group does not influence the
enantiodiscrimination mechanism exhibited by the 1-(1-
naphthyl)ethylamino-s-triazine moiety. The same con-
sideration can be made for CSPs 3 and 4 that are still
epimeric at the stereogenic center of the 1-(1-naph-
thyl)ethylamino-s-triazine fragment: the racemates that
are resolved by both the phases show opposite elution
orders (entries 3–5 and 12). However, the elution order
of the two 4-nitrobenzamides upon CSP1 and CSP3,
(which differ in the nature of the amino acid moiety;
leucine for CSP1 and phenylglycine for CSP3 but have
the same absolute configuration at the two stereogenic
centers) is opposite (entries 1–2): this suggests that the
two phases show a different enantiodiscrimination
mechanism for the two compounds. Therefore, we
should say that the nature of the amino acid group
influences the enantiorecognition mechanism exhibited
by the 1-(1-naphthyl)ethylamino-s-triazine moiety at
least as far as the resolution of the 4-nitrobenzamides is
concerned.
The influence of the stereochemistry of the 1-(1-naph-
thyl)ethylamino moiety on the enantiorecognition
mechanism involved in the resolution of p-basic amides
can be analyzed only in the case of the phenylglycine-
containing systems CSP3 and CSP4, which are both
able to separate the enantiomers of three compounds
(Table 3, entries 1, 2 and 4). The elution order of 17a,
17b and 17d remains unchanged from CSP3 to CSP4,
suggesting that the enantiorecognition mechanism of
the derivatized amino acid moiety is not affected to a
great extent by the change in the stereochemistry of
1-(1-naphthyl)ethylamine residue. However, it is inter-
esting to note that, although the absolute configuration
of the two chiral moieties of the biselector system is the
same, the elution order of 17a and 17b (Table 3) on
CSP1 is opposite to that observed on CSP3. This
suggests that some change in the enantiodiscrimination
mechanism, passing from CSP1 to CSP3, must take
place. Furthermore, the elution order for these com-
pounds on a commercial covalent (S)-DNBPG phase is
the same as that observed on CSP1 and it depends only
on the stereochemistry of the amino acid2 in the case of
this kind of CSPs (it is the same for leucine, phenyl-
glycine and valine based CSPs). Therefore, we can
conclude that the enantiorecognition mechanism exhib-

1812
A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
discussed, the 1-(1-naphthyl)ethylamine-s-triazine frag-
ment dominates in the enantiorecognition process.
3. Conclusions
A comparison of the chromatographic data obtained
using CSPs 1–4 in the HPLC resolution of selected
racemic compounds has allowed us to gain some
insights about the enantiorecognition properties of this
family of biselector systems.
In fact, both the stereochemistry and the chemical
structure of the derivatized amino acid fragment influ-
ence the enantiodiscrimination properties of the 1-(1-
naphthyl)ethylamino-s-triazine moiety: the best results
in terms of efficiency and versatility towards p-acidic
racemic compounds are obtained with CSP1.
The enantioresolution capability of the derivatized
amino acid moiety is dramatically influenced by the
stereochemistry of the 1-(1-naphthyl)ethylamino-s-tria-
zine fragment: either in the case of the leucine-contain-
ing systems (CSP1 and CSP2) or with the
phenylglycine-bearing selectors (CSP3 and CSP4), the
matched couple is the (S,S)-system. The most interest-
ing result is the change in the elution order of binaph-
thyl derivatives going from one CSP to another bearing
diastereoisomeric selectors (Fig. 2). Since the two selec-
tors are diastereoisomeric by virtue of the opposite
absolute configuration at the stereogenic center of the
1-(1-naphthyl)ethylamino moiety, the elution order of
the above mentioned class of racemates, whose resolu-
tion is attributable to the derivatized amino acid frag-
ment, must remain the same. The change in the elution
orders observed with this family of CSPs for the
binaphthyl derivatives suggests that a difference in the
enantiorecognition mechanism should be present pass-
ing from a biselector system to its diastereoisomer.
Taking into account that the enantiodiscrimination
model proposed by Pirkle for the 3,5-dinitrobenzoyl
amino acid based CSPs depends on the preferred con-
formation assumed by the selector,6 the different enan-
tiorecognition mechanism operating in the case of the
diastereoisomeric CSPs, should be attributed to a
change of conformation that the amino acid moiety
undergoes in the biselector system, depending on the
absolute configuration at the stereogenic center of the
1-(1-naphthyl)ethylamine-s-triazine fragment. This last
point, although requiring further examination, is quite
interesting because it suggests that the conformation
and, consequently, the enantiorecognition properties of
chiral moieties that are components of structurally
complex molecules such as these biselector systems, can
be markedly affected by the ‘remote chirality’ present
on the same molecule. This can give helpful insights for
the development of ‘broad spectrum’ chiral auxiliaries.
In fact, systems having wider applicability can be built
up by linking two different and complementary selec-
tors to one another, provided the two chiral units are
‘well coupled’ as far as both chemical nature and
stereochemistry are concerned.15
Figure 2. UV (upper traces) and CD (lower traces) chromato-
graphic detection concerning the separation of 6,6%-dibromo-
2,2%-dihydroxy-1,1%-binaphthyl 15b on CSP2 (a) and CSP1
(b).
ited by the derivatized amino acid moiety of the biselec-
tor system towards these compounds is influenced by
the presence of the other chiral fragment when the
amino acid is phenylglycine so much that their elution
order is opposite to that found on the Pirkle’s CSP.
Also when the amino acid residue is leucine, we observe
different behavior depending on the stereochemistry of
the other moiety. In fact CSP2 is unable to resolve
these compounds at all.
The elution order of 19 changes depending on the
absolute configuration at the stereogenic center of the
1-(naphthyl)ethylamine-s-triazine moiety (Table 3): this
suggests that although both moieties are involved in the
enantiodiscrimination of this compound, as previously

A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
1813
4. Experimental
saturated NaHCO3 solution and water, then dried over
Na2SO4. The solvent was evaporated under reduced
pressure and the crude product was purified by column
chromatography (SiO2, hexane:ethylacetate=2:8).
4.1. General
1H and 13C spectra were recorded in CDCl3 on a 200
MHz NMR spectrometer, using TMS as an external
standard. The following abbreviations are used: s=sin-
glet, d=doublet, dd=double doublet, t=triplet, q=
quartet, m=multiplet, br s=broad signal. TLC
analyses were performed on silica gel sheets; chromato-
graphic separations were carried out on adequate
dimension columns using silica gel 60 (70–230 mesh).
HPLC analyses were performed using an intelligent
HPLC pump equipped with an UV detector and a
spectropolarimeter. Optical rotations were measured
with a digital polarimeter. Melting points are uncor-
rected. The IR spectra were recorded on a Perkin–
Elmer 1710 spectrophotometer. Elemental analyses
were carried out at ICMAT-CNR, AREA della
RICERCA di ROMA. Toluene and THF were refluxed
over sodium benzophenone and distilled before use.
Allylamine, diisopropylethylamine (DIPEA) and
CH3CN were distilled over CaH2; (S)- and (R)-1-(1-
naphthyl)ethylamine were distilled under reduced pres-
sure. Unless otherwise specified the reagents were used
without any purification.
4.2.1.
2[(S)-N-(3,5-Dinitrobenzoyl)amino-iso-butyl-
acetyl] - 2 - {[4 - allylamino - 6 - (R) - (1 - (1 - naphthyl)ethyl-
amino)-1,3,5-triazin-2-yl]oxy}ethylamine, 9a. Yield:
38%; mp=120–124°C; [h]2D4=−26 (c 1, CH2Cl2); 1H
NMR (200 MHz, DMSO-d6 100°C, l/ppm): 0.92 (t,
6H, CH(CH3)2); 1.58 (d, 3H, C*CH3); 1.69 (m, 3H,
CHCH2); 3.39 (dd, 2H, -CH6 2NH); 3.80 (br s, 2H,
-CH2OAr); 4.20 (t, 2H, allylic -CH2); 4.60 (m, 1H,
*CH-iBu); 5.00 (m, 2H, ꢀCH2); 5.78 (m, 1H, CHꢀ); 5.92
(m, 1H, *CH-Naft); 7.00 (br s, 1H, NH allylic); 7.40–
7.82 (m, 5H, aromatic protons superimposed to 1H,
-NH
6
*CH-naph); 7.88–8.00 (d, 1H, aromatic superim-
CO); 8.22 (d, 1H, aromatic);
posed to t, 1H, -CH2NH
6
8.96 (t, 1H, aromatic); 9.02 (d, 1H, *CHNH6 CO); 9.09
(d, 2H, aromatics); 13C NMR (50 MHz, DMSO-d6,
100°C, l/ppm): 21.2–21.3 (CH(CH3)2); 22.2 (-CH3);
24.1 (CHCH2); 38.1 (*CH-naph); 40.2 (CH(CH3)2); 42.1
(CH2CHꢀ); 45.3 (CH2NH); 52.3 (CO*C6 HNH); 69.8
6
(CH2OAr); 114.9, 116.0, 120.1, 122.2, 122.7, 124.8,
124.9, 125.3, 126.5, 127.2, 128.1, 135.2, 171.2; IR
(KBr): 3408, 3292, 3090, 2976, 2931, 2875, 1729, 1660,
1628, 1583, 1543, 1427, 1343, 1258, 1237, 1187, 1152,
1105, 1076, 993, 918, 813, 780, 730. Anal. calcd for
C33H37N9O7: C, 59.01; H, 5.55; N, 18.77. Found: C,
58.92; H, 5.53; N, 18.84%.
Standard procedures were used for preparing racemic
amides, the 3,5-dinitrobenzoyl derivatives of amino
acids and the 3,5-dinitroanilide of ibuprofen.5
The preparation of CSP1 has been previously
4.2.2. 2[(S)-N-(3,5-Dinitrobenzoyl)amino-phenylacetyl]-
2-{[4-allylamino-6-(S)-(1-(1-naphthyl)ethylamino)-1,3,5-
described.
2-{[4-Allylamino-6-(S)-(1-(1-naphthyl)-
ethylamino)-1,3,5-triazin-2-yl]oxy}ethylamine 8b was
prepared as described elsewhere5 and matched the
reported characteristics. Its enantiomer 8a was obtained
according to the same procedure.5
triazin-2-yl]oxy}ethylamine,
9b.
Yield:
38%;
1
mp=92–97°C; [h]D22=+60 (c 0.92, CH2Cl2); H NMR
(200 MHz, DMSO-d6 100°C, l/ppm): 1.59 (d, 3H,
-CH3); 3.40 (dd, 2H,-CH6 2NH); 3.80 (t, 2H, -CH2OAr);
4.22 (t, 2H,-CH2 allylic); 5.00 (m, 2H, ꢀCH2); 5.78 (m,
1H, CHꢀ superimposed to 1H, *CH-Ph); 5.95 (m, 1H,
*CH-naph); 6.80 (t, 1H, NH allylic); 7.20–7.90 (m,
4.1.1. tert-Butyl-2-{[4-chloro-6-(R)-(1-(1-naphthyl)ethyl-
amino)-1,3,5-triazin-2-yl]oxy}ethylcarbamate, 6a. Mp=
67–70°C; [h]2D2=−34.2 (c 0.9, CH2Cl2).
11H, aromatics superimposed to 1H, NH
8.22 (d, 1H, aromatic superimposed to t, 1H, -
CH2NH); 8.95 (t, 1H, aromatic); 9.10 (d, 2H, aromat-
ics); 9.42 (d, 1H, *CHNH
CO); 13C NMR (50 MHz,
6 *CH-naph);
4.1.2. tert-Butyl-2-{[4-allylamino-6-(R)-(1-(1-naphthyl)-
6
ethylamino)-1,3,5-triazin-2-yl]oxy}ethylcarbamate,
7a.
6
Mp=71–74°C; [h]D22=−51.9 (c 0.865, CH2Cl2).
DMSO-d6, 100°C, l/ppm): 20.3 (-CH3); 37.2 (*CH-
naph); 41.0 (CH2CHꢀ); 44.3 (CH2NH); 56.3
4.1.3. 2-{[4-Allylamino-6-(R)-(1-(1-naphthyl)ethylamino)-
1,3,5-triazin-2-yl]oxy}ethylamine, 8a. Mp=60–63°C;
[h]2D5=−64 (c 1, CH2Cl2).
(CO*C6 HNH); 62.6 (CH2OAr); 113.9,119.1, 121.2,
121.6, 123.7, 123.9, 124.3, 125.4, 126.1, 126.2, 126.4,
126.7, 127.0, 132.0, 134.2, 135.7, 136.4, 139.5, 146.7,
161.0, 164.7, 165.5, 168.0, 168.7. IR (KBr): 3408, 3292,
3090, 2976, 2931, 2875, 1729, 1660, 1628, 1583, 1543,
1427, 1343, 1258, 1237, 1187, 1152, 1105, 1076, 993,
918, 813, 780, 730. Anal. calcd for C35H33N9O7: C,
60.77; H, 4.81; N, 18.22. Found: C, 60.92; H, 4.83; N,
18.35%.
4.2. General procedure for the preparation of com-
pounds 9
To a solution of the 3,5-dinitrobenzoyl amino acid
(2.74 mmol) in dry THF (60 ml), 2-ethoxy-1-ethoxycar-
bonyl-1,2-dihydroquinoline (EEDQ) (0.678 g, 2.74
mmol) was added and the reaction mixture was stirred
at room temperature. After 3 h, 8a or 8b (1 g, 2.74
mmol) was added and the reaction mixture was stirred
for 15 h at room temperature. The solvent was evapo-
rated under reduced pressure and the crude product
dissolved in dichloromethane. The organic solution was
washed sequentially with 10% hydrochloric acid, water,
4.2.3. 2[(S)-N-(3,5-Dinitrobenzoyl)amino-phenylacetyl]-
2-{[4-allylamino-6-(R)-(1-(1-naphthyl)ethylamino)-1,3,5-
triazin-2-yl]oxy}ethylamine, 9c. Yield: 36%; mp=
1
123–127°C; [h]D22=−5.4 (c 0.8, CH2Cl2) H NMR (200
MHz, DMSO-d6 100°C, l/ppm): 1.59 (d, 3H, -CH3);
3.40 (dd, 2H, -CH2
2H,-CH2 allylic); 5.00 (m, 2H, ꢀCH2); 5.78 (m, 1H,
6 NH); 3.80 (t, 2H, -CH2OAr); 4.22 (t,

1814
A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
CHꢀ superimposed to 1H, *CH-Ph); 5.95 (m, 1H,
*CH-naph); 6.80 (t, 1H, NH allylic); 7.20–7.90 (m,
2H, -CH6 2NH); 3.35–3.80 (m, 11H, -CH2OAr,
CH3OSi); 4.20 (t, 2H, -CH2NHAr); 5.75 (d, 1H, *CH-
Ph); 5.93 (m, 1H, *CH-naph); 6.80 (t, 1H,
11H, aromatics superimposed to 1H, NH
8.22 (d, 1H, aromatic superimposed to t, 1H, -
CH2NH); 8.95 (t, 1H, aromatic); 9.10 (d, 2H, aromat-
ics); 9.42 (d, 1H, *CHNH
CO); 13C NMR (50 MHz,
6 *CH-naph);
CH2NH
posed to 1H, -NH
8.96 (t, 1H, aromatic); 9.08 (d, 2H, aromatics); 9.40
(d, 1H, *CHNHCO). Anal. calcd for C41H49N9O10SSi:
6
Ar); 7.20–8.30 (m, 12H, aromatics superim-
6
6
*CH-naph and to 1H, CH2NHCO);
6
6
DMSO-d6, 100°C, l/ppm): 20.3 (-CH3); 37.2 (*CH-
naph); 41.0 (CH2CHꢀ); 44.3 (CH2NH); 56.3
6
C, 55.45; H, 5.56; N, 14.20; S, 3.6. Found: C, 55.47;
H, 5.55; N, 14.21; S, 3.60%.
(CO*C6 NH); 62.6 (CH2OAr); 113.9, 119.1, 121.2,
121.6, 123.7, 123.9, 124.3, 125.4, 126.1, 126.2, 126.4,
126.7, 127.0, 132.0, 134.2, 135.7, 136.4, 139.5, 146.7,
161.0, 164.7, 165.5, 168.0, 168.7; IR (KBr): 3408, 3292,
3090, 2976, 2931, 2875, 1729, 1660, 1628, 1583, 1543,
1427, 1343, 1258, 1237, 1187, 1152, 1105, 1076, 993,
918, 813, 780, 730. Anal. calcd for C35H33N9O7: C,
60.77; H, 4.81; N, 18.22. Found: C, 60,68; H, 4.82; N,
18.28%.
4.4. Preparation of CSPs 2–4
A solution of the silane 10, 0.88 mmol) in dry toluene
(15 ml) was added dropwise to LiChrospher Si 100
2
,
silica gel (100 A, 5 mm particle size, 300 m /g, 2.5 g),
previously dried at 180°C at 0.05 mm Hg for 15 h and
slurried in dry toluene (15 ml) and the mixture was
gently stirred under reflux for 24 h. The mixture,
cooled at room temperature, was filtered and washed
with toluene (3×30 ml), dichloromethane (3×30 ml),
methanol (3×30 ml), THF (3×30 ml) and pentane (3×
30 ml), then dried at 50°C at 0.05 mmHg.
4.3. General procedure for preparation of compounds
10a–c
To a solution of 9 (0.9 mmol) in dry CHCl3 (10 ml)
freshly distilled 3-mercaptopropyl trimethoxysilane
(0.85 ml, 4.5 mmol) and AIBN (0.033 g, 0.2 mmol)
were added and the mixture was stirred under reflux
for 48 h. The solvent was eliminated by evaporation
under reduced pressure and the residual oil was dis-
persed in pentane affording a solid, which was filtered
and washed with pentane (5×20 ml). The pure product
10 was obtained in quantitative yield.
The amount of chiral selector linked to silica gel was
determined by elemental analysis.
CSP 2: C, 9.80; H, 1.24; N, 2.62; S, 0.98% corre-
sponding to 0.230 mmol/g (0.66 mmol/m2).
CSP 3: C, 14.48; H, 1.86; N, 2.87; S, 0.91 correspond-
ing to 0.292 mol/g (0.89 mmol/m2).
CSP 4: C, 11.59; H, 1.33; N, 2.53; S, 1.09 correspond-
ing to 0.260 mol/g (0.74 mmol/m2).
1
4.3.1. Compound 10a. [h]2D2=−27 (c 1.08, CH2Cl2); H
NMR (200 MHz, DMSO-d6 100°C, l/ppm): 0.65 (m,
2H, SiCH2); 0.92 (t, 6H, CH(CH3)2); 1.55–1.75 (m,
10H, C*CH3, CHCH2, SiCH2CH2, CH2CH2NHAr);
The derivatized silica gels were slurry-packed into 15
cm stainless steel columns, using conventional
techniques.
2.50 (m, 4H, CH2SCH2); 3.20–3.50 (m, 11H, -CH6 2NH,
CH3OSi); 3.80 (m, 2H, -CH2OAr); 4.20 (m, 2H,-
CH2NHAr); 4.60 (m, 1H, *CH-iBu); 5.95 (m, 1H,
Acknowledgements
*CH-Naft); 6.80 (s sl, 1H, CH2NH
7H, aromatics superimposed to 1H, -NH
and to 1H, CH2NHCO); 8.96 (t, 1H, aromatic); 9.03
(d, 1H, *CHNHCO); 9.11 (d, 2H, aromatics). Anal.
6
Ar); 7.40–8.20 (m,
6
*CH-naph
6
This work was supported by the MIUR (Progetto
Nazionale Stereoselezione in Sintesi Organica:
Metodologie e Applicazioni) and the AmbiSEN (Cen-
tro di Eccellenza dell’Universita` di Pisa). We thank
Dr. Luca Petrilli (ICMAT-CNR Area della Ricerca di
Roma) for the elemental analyses.
6
calcd for C39H53N9O10SSi: C, 53.96; H, 6.15; N, 14.52;
S 3.69. Found: C, 54.01; H, 6.14; N, 14.49; S, 3.68%.
1
4.3.2. Compound 10b. [h]2D2=+62 (c 0.9, CH2Cl2); H
NMR (200 MHz, DMSO-d6 100°C, l/ppm): 0.68 (m,
2H, SiCH2); 1.55–1.75 (m, 7H, C*CH3, SiCH2CH2,
CH2CH2NHAr); 2.50 (m, 4H, CH2SCH2); 3.20 (m,
2H, -CH2
CH3OSi); 4.20 (t, 2H, -CH2NHAr); 5.72 (d, 1H, *CH-
Ph); 5.92 (m, 1H, *CH-Naft); 6.70 (t, 1H, CH2NHAr);
7.22–8.26 (m, 12H, aromatics superimposed to 1H-
NH*CH-naph and to 1H, CH2NHCO); 8.93 (t, 1H,
aromatic); 9.05 (d, 2H, aromatics); 9.38 (d, 1H,
*CHNHCO). Anal. calcd for C41H49N9O10SSi: C,
6
NH); 3.35–3.80 (m, 11H, -CH2OAr,
References
6
1. (a) Satinder, A. Chiral separation: Applications and Tech-
nology; ACS: Washington, 1997; (b) Maier, N. M.;
Franco, P.; Lindner, W. J. Chromatogr. A 2001, 906,
3–33; (c) Fe´lix, G. J. Chromatogr. A 2001, 906, 171–184.
2. (a) Pirkle, W. H.; Pochapsky, T. C. Chem. Rev. 1989, 89,
347–362; (b) Taylor, D. R.; Maher, K. J. Chromatogr.
Sci. 1992, 30, 67–85; (c) Gasparrini, F.; Misiti, D.; Vil-
lani, C. J. Chromatogr. A 2001, 906, 35–50.
3. Welch, C. J. J. Chromatogr. A 1994, 666, 3–26.
4. (a) Pirkle, W. H; Welch, C. J. Tetrahedron: Asymmetry
1994, 5, 777–780; (b) Pirkle, W. H.; Yuelong, L. J.
Chromatogr. A 1996, 736, 31–38.
6
6
6
55.45; H, 5.56; N, 14.20; S, 3.61. Found: C, 55.42; H,
5.57; N, 14.19; S, 3.62%.
4.3.3. Compound 10c. [h]2D4=−16 (c 0.9, CH2Cl2); 1H
NMR (200 MHz, DMSO-d6 100°C, l/ppm): 0.68 (m,
2H, SiCH2); 1.55–1.75 (m, 7H, C*CH3, SiCH2CH2,
CH2CH2NHAr); 2.50 (m, 4H, CH2SCH2); 3.20 (m,

A. Iuliano et al. / Tetrahedron: Asymmetry 13 (2002) 1805–1815
1815
5. Iuliano, A.; Attolino, E.; Salvadori, P. Eur. J. Org. Chem.
2001, 3523–3529.
10. Rosini, C.; Bertucci, C.; Pini, D.; Altemura, P.; Sal-
vadori, P. Tetrahedron Lett. 1985, 26, 3361–3364.
11. Masamune, S.; Choy, W.; Petersen, W.; Sita, L. R.
Angew. Chem., Int. Ed. Engl. 1985, 24, 1–76.
12. Salvadori, P.; Di Bari, L.; Pescitelli, G. In Circular
Dichroism: Principles and Applications; Berova, N.;
Nakanishi, K.; Woody, R. W., Eds.; John Wiley: New
York, 2000; pp. 797–817.
6. (a) Pirkle, W. H.; House, D. W.; Finn, J. M. J. Chro-
matogr. A 1980, 192, 143–158; (b) Pirkle, W. H.; Welch,
C. J.; Hyun, M. H. J. Org. Chem. 1983, 48, 5022–5026;
(c) Pirkle, W. H.; Welch, C. J. J. Org. Chem. 1984, 49,
138–140; (d) Pirkle, W. H.; Schreiner, J. L. J. Org. Chem.
1981, 46, 4988–4991; (e) Pirkle, W. H.; Finn, J. M. J.
Org. Chem. 1981, 46, 2935–2938.
13. Salvadori, P.; Rosini, C.; Bertucci, C. J. Org. Chem.
1984, 49, 5050–5054.
7. Oi, N.; Nagase, M.; Sawada, Y. J. Chromatogr. A 1984,
292, 427–431.
8. Gustafson, G. R.; Baldino, C. M.; O’Donnell, M. E.;
Sheldon, A.; Tarsa, R. J.; Verni, C. J.; Coffen, D. L.
Tetrahedron 1998, 54, 4051–4065.
14. Harada, N.; Nakanishi, K. Circular Dichroic Spec-
troscopy, Exciton Coupling in Organic Stereochemistry;
University Science Books: Oxford Press, 1983.
15. It has been shown for other kinds of CSPs that the
further introduction of suitable moieties on a CSP can
affect its enantiodiscrimination properties (see Ref. 1c).
9. Belleau, B.; Malek, G. J. Am. Chem. Soc. 1968, 90,
1651–1652.