Preparation and evaluation of novel chiral stationary phases covalently bound with chiral pseudo-18-crown-6 ethers

Article abstract of DOI:10.1016/S0040-4039(03)00020-0

Novel chiral stationary phases consisting of silica gel covalently bound with chiral pseudo-18-crown-6 type hosts, which possess either an OH or OMe group as a binding functionality, were prepared for enantiomer-separation of lipophilic amines.

Full text of DOI:10.1016/S0040-4039(03)00020-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1549–1551
Preparation and evaluation of novel chiral stationary phases
covalently bound with chiral pseudo-18-crown-6 ethers
Keiji Hirose,a,* Takashi Nakamura,a Ryota Nishioka,b Tetsuro Ueshigeb and Yoshito Tobea
aDepartment of Chemistry, Faculty of Engineering Science, Osaka University, Machikaneyama, Toyonaka, Osaka 560-8531,
Japan
bSumika Chemical Analysis Service, 3-1-135 Kasugadenaka, Konohana, Osaka 554-0022, Japan
Received 9 November 2002; revised 28 December 2002; accepted 7 January 2003
This paper is dedicated to Emeritus Professor Soichi Misumi on the occasion of his 77th birthday
Abstract—Novel chiral stationary phases consisting of silica gel covalently bound with chiral pseudo-18-crown-6 type hosts, which
possess either an OH or OMe group as a binding functionality, were prepared for enantiomer-separation of lipophilic amines.
© 2003 Elsevier Science Ltd. All rights reserved.
Although Cram and co-workers reported enantiomer-
separation of amino acids and amino esters by HPLC
using chiral stationary phases (CSPs) consisting of chi-
ral crown ethers attached on polystyrene1 or silica gel in
the 1970s,2 such were not commercially available. In
1987, Shinbo and co-workers reported the separation of
amino acids by using a CSP in which a hydrophobic
chiral crown ether was dynamically coated on an ODS
column.3 This type of CSP was commercialised as
CROWNPAK CR and it has been proven useful for
the resolution of chiral primary amines. However,
because of the dynamically coated nature of this CSP,
solvents that can be used as a mobile phase are limited.
The use of a solvent containing more than 15%
methanol would result in leaching of the chiral crown
selector and deterioration of the CSP.4 Since 1998,
chemical-bonded type CSPs containing chiral crown
ether have been developed actively.5 However, there is
no CSP in which a normal mobile phase is used as an
eluent. Due to a recent analytical demand, there is an
increasing requirement for separation of chiral
lipophilic amino compounds. Therefore, it is desired to
develop CSPs which can be used with a normal mobile
phase as an eluent. In addition to this requirement, the
usage of lipophilic neutral amines rather than their
ammonium salts as substrate is required because of
their better solubility in a normal mobile phase.
To meet these requirements and achieve good chiral
separation, we designed OH type CSP, (S, S)-1, where
a selector is immobilised on silica gel covalently and
where a chiral phenolic crown ether is adopted as a
selector expecting stable salt-complex formation with a
neutral amine. As a selector, chiral pseudo-18-crown-6
meta-cyclophane frameworks are adopted, because high
enantiomer-selective complexations toward primary
amines and ammonium salts are reported.6 In addition
to (S, S)-1, OMe type CSP (S, S)-2 with the same
meta-cyclophane framework was prepared. An amide
linkage is known to be sufficiently stable for the chro-
matography. Therefore, CSPs 1 and 2 should be stable,
and both the reversed and the normal phases can be
used as eluent. For comparison, amides selectors (S,
S)-3 and (S, S)-4 were prepared (Fig. 1).
Keywords: chiral stationary phase; crown ether; chiral recognition.
* Corresponding author. Tel.: +81 6 6850 6228; fax: +81 6 6850 6229;
e-mail: hirose@chem.es.osaka-u.ac.jp
Figure 1. The structures of CSPs 1, 2 and selectors 3, 4.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(03)00020-0

1550
K. Hirose et al. / Tetrahedron Letters 44 (2003) 1549–1551
chromatography on CSP 1, the acid additive was not
necessary. The phenolic OH group placed in the bind-
ing site might play the same role as an acid derivative in
which the amine could be complexed with phenolic
crown ether to form ammonium phenolate coined
‘saltex’.8 On the contrary, a base, e.g. triethylamine
(TEA), could be used for this CSP and was effective9 to
improve the separation. The chromatograms on CSP 1
are shown in Figure 2 (a)–(c). The separation factors
(h) are 2.02, 2.12 and 5.26 using 11, 12, 13, respectively.
The corresponding h values on CSP 2 are 1.00, 1.33,
1.17, respectively. The chromatograms on CSP 2 are
shown in Figure 2 (d)–(f).10 The OH type CSP 1
exhibited excellent enantiomer-separation of all amines
employed here using the normal mobile phase without
addition of an acid additive.
Next, the enantiomer-selectivities of selectors 3 and 4 in
solution were examined. The binding constants of 3
with chiral amines, 11–13 were determined in chloro-
form-d by 1H NMR spectroscopic method.11 Those of 4
Scheme 1. Reagents and conditions: i, (1) 4-bromo-2,6-bis(bro-
momethyl)anisole, NaH, (2) EtOH, pyridinium p-toluenesul-
fonate, 80%; ii, diethylene glycol ditosylate, NaH, 53%; iii, (1)
n-BuLi, (2) CO2 gas, (3) HCl, 87%; iv, EtSNa, 69%; v, for 2,
(1)
1-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinoline,
(2)
Ac2O, 30%*, for 1, (1) 1-ethyl-3-(3-dimethylamino-propyl)-
carbodiimide hydrochloride, (2) Ac2O, 28%*; vi, 1-ethoxycar-
bonyl-2-ethoxy-1,2-dihydroquinoline, 4 80%, 3 72%. * The
content of selector bound aminopropyl group over total
amount of grafted aminopropyl group on silica gel.
The preparation of CSPs 1, 2 and selectors 3, 4 are
shown in Scheme 1. The chiral monoprotected diol
(S)-5 was synthesised from commercially available
mandelic acid according to the literature.7 The alcohol
(S)-5 was converted to crown ether (S, S)-8 via podand
(S, S)-7. The crown ether (S, S)-9, which was obtained
by carboxylation of (S, S)-8 is the intermediate for
both selectors and both CSPs. For the syntheses of
CSPs 1 and selectors 3, (S, S)-10 was obtained by the
demethylation of (S, S)-9 with sodium ethanthiolate.
The CSPs 1 and 2 were obtained by the amidation of
(S, S)-10 or (S, S)-9 with 3-aminopropylsilica gel,
followed by treatment of the resulting silica gel with
acetic anhydride. The amounts of selectors bound to
the silica gel were calculated based on the elemental
analyses of CSPs 1 and 2. CSP 1 and 2 were packed
into 250×4.6 mm I.D. stainless steel columns using a
conventional slurry packing method. The selectors (S,
S)-3 and (S, S)-4 were obtained by the amidation of (S,
S)-9 and (S, S)-10, respectively, with n-propylamine.
First, the separation of racemic organic amines on
CSPs 1 and 2 using the normal mobile phase was
examined. The amino compounds examined here were
selected from commonly used amines, 1-phenylethyl-
amine (11), 1-(1-naphthyl)ethylamine (12), and phenyl-
glycinol (13). As the mobile phase, an ordinarily used
normal phase, hexane–ethanol, was employed. In the
case of chromatography on CSP 2, trifluoroacetic acid
(TFA) was used as an indispensable acid additive for
the protonation of the amino group. In the case of
Figure 2. Chromatograms for the resolution of 11–13 on CSP
1 and on CSP 2 using normal mobile phase. Chromato-
graphic conditions: flow rate=0.7 ml/min, 25°C, hexane/
EtOH/TEA=70/30/0.2
for
11,
12
on
CSP
1,
hexane/EtOH/TEA=60/40/0.5 for 13 on CSP 1, hexane/
EtOH/TFA/H2O=70/30/0.5/0.2 for 11–13 on CSP 2. The
scale of the abscissa is presented in min.

K. Hirose et al. / Tetrahedron Letters 44 (2003) 1549–1551
Table 1. Binding constants and enantiomer-selectivities of
1551
Acknowledgements
selectors 3, 4 at 303 K and separation factors (h) of cor-
responding CSPs 1, 2
This work was partially supported by SUNBOR foun-
dation from Suntory Institute for Bioorganic Research
and a Grant-in-Aid for Scientific Research from the
Ministry of Education, Science, Sports and Culture of
Japan.
Selector
Substrate
KR/M−1 KS/M−1 KR/KS hc
11
12
1.1d
4.8d
2.8d
7.7
0.23
0.20
9.3
2.02
2.12
5.26
1.00
1.33
1.17
3a
0.54d
13
72
11·HCl
12·HCl
13·HCl
25
38
16
25
27
15
1.0
1.4
1.0
4b
References
1. Sousa, L. R.; Sogah, G. D. Y.; Hofmann, D. H.; Cram,
D. J. J. Am. Chem. Soc. 1978, 100, 4569–4576.
2. Sogah, G. D. Y.; Cram, D. J. J. Am. Chem. Soc. 1979,
101, 3035–3042.
3. (a) Shinbo, T.; Yamaguchi, T.; Nishimura, K.; Sugiura,
M. J. Chromatogr. 1987, 405, 145–153; (b) Shinbo, T.;
Yamaguchi, T.; Yanagishita, H.; Kitamono, D.; Sakaki,
K.; Sugiura, M. J. Chromatogr. 1992, 625, 101–108.
4. Instruction Manual for CROWNPAK CR (1), Daicel
Chemical Industries.
5. (a) Hyun, M. H.; Jin, J. S.; Lee, W. Bull. Korean Chem.
Soc. 1998, 19, 819–821; (b) Machida, Y.; Nishi, H.;
Nakamura, K.; Nakai, H.; Sato, T. J. Chromatogr. A
1998, 805, 85–92; (c) Hyun, M. H.; Jin, J. S.; Koo, H. J.;
Lee, W. J. Chromatogr. A 1999, 837, 75–82; (d) Kontos,
Z.; Huszthy, P.; Bradshaw, J. S.; Izatt, R. M. Tetra-
hedron: Asymmetry 1999, 10, 2087–2099; (e) Horva´th, G.;
Huszthy, P.; Szarvas, S.; Szo´ka´n, G.; Redd, J. T.; Brad-
shaw, J. S.; Izatt, R. M. Ind. Eng. Chem. Res. 2000, 39,
3576–3581; (f) Hyun, M. H.; Han, S. C.; Lipshutz, B. H.;
Shin, Y. J.; Welch, C. J. J. Chromatogr. A 2002, 959,
75–83.
6. (a) Hirose, K.; Ogasahara, K.; Nishioka, K.; Tobe, Y.;
Naemura, K. J. Chem. Soc., Perkin Trans. 2 2000, 1984–
1993; (b) Naemura, K.; Matsunaga, K.; Fuji, J.; Ogasa-
hara, K.; Nishikawa, Y.; Hirose, K.; Tobe, Y. Anal. Sci.
1998, 14, 175–182; (c) Hirose, K.; Fuji, J.; Kamada, K.;
Tobe, Y.; Naemura, K. J. Chem. Soc., Perkin Trans. 2
1997, 1649–1657.
a In CDCl3.
b In CDCl3/CD3OD=7/3.
c From chromatography for the corresponding substrate, see Figure
2.
d Obtained by an extrapolation according to the van’t Hoff relation.
were determined with the corresponding ammonium
hydrochloride in methanol-d4–chloroform-d (v/v=3/7)
mixed solvent. The results are listed in Table 1 where
the enantiomer-selectivity of the selectors are shown by
the ratio of binding constants (KR/KS) with the separa-
tion factor (h) of corresponding CSP. The binding
constants of amines 11–13 with selector 3 range from
0.54 to 72 M−1. The enantiomer selectivities are large
(KR/KS=0.23, 0.20, 9.3). The enantiomer-selectivities of
4 were smaller than those of 3. Clear selectivity was
observed when 12·HCl was employed (KR/KS=1.4).
The chromatogram for the resolution of 12 on CSP 2
using an acid additive showed a separation factor h=
1.33 where the baseline separation was performed (Fig.
2 (e)). In the case of 13·HCl, the enantiomer-selectivity
(KR/KS) was small with selector 4 in solution, but the
enantiomer separation using the corresponding CSP 2
was clearly observed (Fig. 2 (f), h=1.17). In general,
the enantiomer which has larger retention time in chro-
matogram has larger binding constant in complexation
of the corresponding model compounds. Since there
exists a correlation between the enantiomer-selectivities
in chiral chromatography and that of the corresponding
model compounds of the selectors in solution, it is
deduced that the chiral separation arose from chiral
recognition in host–guest interaction.
7. Huszthy, P.; Bradshaw, J. S.; Zhu, C. Y.; Izatt, R. M.;
Lifson, S. J. Org. Chem. 1991, 3330–3336.
8. Kaneda, T.; Ishizaki, Y.; Misumi, S.; Kai, Y.; Hirao, G.;
Kasai, N. J. Am. Chem. Soc. 1988, 110, 2970–2972.
9. The addition of TEA effects to improve sharpness of
peaks. The reason for this additive effect is not clear.
10. The chromatographic conditions of CSPs 1 and 2 for
each amine shown here were not optimised.
In conclusion, we have prepared OH type CSP 1, and
OMe type CSP 2 in which selectors were immobilised
on silica gel covalently. On both CSPs 1 and 2, clear
chromatographic enantiomer-separations of chiral
amines were observed using normal mobile phases.12
Especially, the chromatography on CSP 1 exhibited
excellent enantiomer-separations using a normal mobile
phase without acid additives. A correlation between the
enantiomer-selectivities in chiral chromatography and
that of the corresponding model compounds of the
selectors in solution was observed, implying that the
chiral separation arose from chiral recognition in host–
guest interaction. Currently, we are investigating the
enantiomer separations of a wide range of amino com-
pounds in order to clarify the characteristics of sub-
strates which can be separated using these CSPs.
11. Hirose, K. J. Incl. Phenom. Macrocyclic Chem. 2001, 39,
193–209.
12. The CSPs 1 and 2 can be used for the separation of other
primary amines and amino acids, such as norephedrine,
phenylalanine, and alaninol, which exhibit separation
factors (h) 1.32, 1.24, 1.07 on CSP 1 and 1.47, 1.45, 1.18
on CSP 2, respectively. Chromatographic conditions:
flow rate=0.7 ml/min, 25°C, hexane/iso-PrOH/MeOH/
TFA=80/15/5/0.5 for the separation of alaninol, hexane/
EtOH/TFA=70/30/0.5 for the separation of other
amines.