Preparation of a new chiral acridino-18-crown-6 ether-based stationary phase for enantioseparation of racemic protonated primary aralkyl amines

Article abstract of DOI:10.1016/j.tet.2007.09.056

Starting from commercially available and relatively cheap chemicals first enantiopure dimethyl-substituted monoaza-18-crown-6 ether (R,R)-21 containing a diphenylamine unit was prepared, which was then transformed to dimethyl-substituted acridino-18-crown

Full text of DOI:10.1016/j.tet.2007.09.056

Available online at www.sciencedirect.com
Tetrahedron 64 (2008) 1012–1022
Preparation of a new chiral acridino-18-crown-6 ether-based
stationary phase for enantioseparation of racemic
protonated primary aralkyl amines
a
Szilvia Lakatos, Jozsef Fetter, Ferenc Bertha, Peter Huszthy,
a
a
a,b,
b
*
ꢀ
Viktor Farkas, Gyorgy Orosz and Miklos Hollosi
ꢀ
€
ꢀ
Tunde Toth,
c
d
e
€
ꢀ
ꢀ
aDepartment of Organic Chemistry and Technology, Budapest University of Technology and Economics,
H-1521 Budapest, PO Box 91, Hungary
bResearch Group for Alkaloid Chemistry of the Hungarian Academy of Sciences, H-1521 Budapest, PO Box 91, Hungary
cProtein Modelling Group of the Hungarian Academy of Sciences, Eo€tvo€s Loraꢀnd University,
H-1518 Budapest, 112, PO Box 32, Hungary
dReanal Fine Chemicals Co., H-1147 Budapest, Telepes u. 53, Hungary
eInstitute of Chemistry, Eo€tvo€s Loraꢀnd University, H-1518 Budapest, 112, PO Box 32, Hungary
Received 8 August 2007; revised 6 September 2007; accepted 20 September 2007
Available online 23 October 2007
ꢀ
Dedicated to Professor Csaba Szantay on the occasion of his 80th birthday
Abstract—Starting from commercially available and relatively cheap chemicals first enantiopure dimethyl-substituted monoaza-18-crown-6
ether (R,R)-21 containing a diphenylamine unit was prepared, which was then transformed to dimethyl-substituted acridino-18-crown-6
ligand (R,R)-19 having an N-allyl-carbamoyl linker by several steps. The terminal double bond of the latter made possible to attach (R,R)-
19 to g-mercaptopropyl-functionalized spherical HPLC quality silica gel obtaining a new chiral stationary phase (R,R)-CSP-37. Based on
electronic circular dichroism (ECD) studies the N-allyl-carbamoyl group attached to the acridine ring of the chiral host (R,R)-19 does weaken
exciton interaction between the host and guest molecules, but does not destroy the discriminating power of the chiral host. An HPLC column
filled with (R,R)-CSP-37 was tested for the enantioseparation of racemic 1-(1-naphthyl)- and 1-(2-naphthyl)ethylamine hydrogenperchlorates
using isocratic conditions.
Ó 2007 Elsevier Ltd. All rights reserved.
1. Introduction
Cram and co-workers were the first who prepared and uti-
lized CSPs containing chiral crown ethers for the resolution
of primary amines, amino acids and amino acid esters. They
immobilized optically active bis(1,10-binaphthyl)-22-crown-
6 ethers on silica gel4 and polystyrene resin.5
It is well known that the enantiomers of chiral biogen
amines, amino acids and their derivatives have different bi-
ological and physiological properties. Therefore, efficient
methods for their enantiomeric separation and determination
of their enantiomeric compositions are essential.
In 1987 Shinbo and co-workers were dynamically coating
(3,30-diphenyl-1,10-binaphthyl)-20-crown-6 ether6 and 5
years later (6,60-dioctyl-3,30-diphenyl-1,10-binaphthyl)-20-
crown-6 ether7 on octadecylsilica gel. The latter CSPs
were successfully applied for the resolution of various race-
mic compounds containing a primary amino group.
Among the various methods available, liquid chromato-
graphy (LC) on chiral stationary phases (CSPs) has been
proved to be the most precise and efficient means of separat-
ing the two enantiomers and of determining the enantiomer-
ic compositions.1 CSPs having crown ether selectors have
been most successfully utilized in enantiomeric separation
of primary amines, amino acids and their derivatives for
both analytical and preparative purposes.2,3
In 2001 Hyun and co-workers using suitable side chains at-
tached a (3,30-diphenyl-1,10-binaphthyl)-20-crown-6 deriva-
tive covalently to silica gel, and this CSP proved to be very
useful in the resolution of various racemic a-amino acids,8
non-cyclic and cyclic amines, amino alcohols9 and aryl
a-amino ketones.10
* Corresponding author. Tel.: +36 1 463 1071; fax: +36 1 463 3297; e-mail:
huszthy@mail.bme.hu
URL: http://www.och.bme.hu/org
Application of (+)-(18-crown-6)-2,3,11,12-tetracarboxylic
acid as a chiral selector for liquid chromatographic CSPs
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.09.056

S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
1013
was first reported by Machida11 and Hyun12 in 1998. The lat-
ter two research groups used the same chiral crown ether as
a selector, but with different modes of attachments to obtain
the CSPs.
CSP so obtained [(R,R)-CSP-3, see Fig. 1] separated well
the enantiomers of racemic 1-NEA, 1-phenylethylamine
hydrogenperchlorate (PEA), phenylalanine methyl ester hy-
drogenperchlorate (PAME) and phenylglycine methyl ester
hydrogenperchlorate (PGME) at atmospheric pressure.25,26
For the purpose of improving enantiomeric separation, an-
other optically active di-tert-butylpyridino-18-crown-6
ether derivative was attached to HPLC quality silica gel by
covalent bonds and this CSP [(R,R)-CSP-4, see Fig. 1] sep-
arated the enantiomers of racemic 1-NEA and PEA under
high pressure with great efficiency.27
Especially Hyun and co-workers have been very successful
in developing new CSPs containing (+)-(18-crown-6)-
2,3,11,12-tetracarboxylic acid for the resolution of a great
variety of racemic amino compounds and their deriva-
tives.13–18
Recently Hirose and co-workers reported the preparation
and the successful application of several CSPs where the op-
tically active pseudo-18-crown-6 ethers as selectors were co-
valently bound to silica gel.19–22 These CSPs separated the
enantiomers of racemic amines, amino alcohols, amino
acids and amino acid esters with great efficiency.
Very recently an optically active dimethylpyridino-18-
crown-6 ether derivative was immobilized on spherical
HPLC quality silica gel and the CSP so obtained [(S,S)-
CSP-5, see Fig. 1] separated very well the enantiomers of
racemic 1-NEA, 1-(2-naphthyl)ethylamine hydrogenper-
chlorate (2-NEA), aromatic a-amino acids and aliphatic
a-amino acids containing different aromatic side-chain pro-
tecting groups applying high pressure.28
In 1993 Bradshaw and co-workers reported the preparation
of the first CSP [(S,S)-CSP-1, see Fig. 1] containing an opti-
cally active pyridino-18-crown-6 ether as a chiral selector.
They attached an enantiopure dimethylpyridino-18-crown-
6 ether derivative to ordinary silica gel by covalent bonds,
and using first an acetone–methanol mixture as an eluent
they could not get a good enantiomeric separation of racemic
1-(1-naphthyl)ethylamine hydrogenperchlorate (1-NEA) at
atmospheric pressure.23 However, using the same CSP
[(S,S)-CSP-1], but applying pure methanol as an eluent
they obtained an almost baseline separation of 1-NEA at at-
mospheric pressure.24
It was reported a few years ago that optically active dimethy-
lacridino-18-crown-6 ether (R,R)-6 (see Fig. 2) showed
higher enantioselectivity towards 1-NEA and PEA than its
pyridino analogue (S,S)-7 (see Fig. 2).29
The higher enantioselectivity was rationalized by the
stronger p–p interaction of the extended p system of the ac-
ridine unit and the more rigid conformation of host molecule
(R,R)-6.29
The latter research group also attached a diphenylpyridino-
18-crown-6 ether to ordinary silica gel by covalent bonds,
but this CSP [(R,R)-CSP-2, see Fig. 1] was less efficient in
the enantioseparation of 1-NEA than (S,S)-CSP-1.24
This observation initiated and motivated us to prepare a CSP
based on an optically active dimethylacridino-18-crown-6
ether derivative as a chiral selector.
Studying the reports connected with our plan in the literature
we found that Charbonniere and Ziessel described the prep-
aration of the monoaza-18-crown-6 ligand 8 (see Fig. 3)
An enantiopure di-tert-butylpyridino-18-crown-6 ether de-
rivative was first attached to ordinary silica gel, and the
N
N
Me
O
O
O
O
Me
Me
O
O
O
O
Me
∗
∗
∗
∗
O
O
(R,R)-6
(S,S)-7
Figure 2. Structures of enantiopure dimethyl-substituted acridino- and
pyridino-18-crown-6 ligands.
X
N
O
O
H
O
O
O
8: X=H
9: X=Cl
Figure 1. Schematics of reported CSPs based on enantiopure pyridino-
18-crown-6 ethers as selectors.
Figure 3. Reported monoaza-18-crown-6 ether type ligands containing a
diphenylamine unit.

1014
S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
M
R1
R2
R1
R2
R1
R1
R2
R1 CO2 R1
O
O
1) (COCl)2
2) AlCl3
10% aq. MOH
N
H
N
N
R2
R2
R2
10: R1=R2=H
13: R1=R2=H
16: R1=R2=H
11: R1=H, R2=Me
12: R1=R2=Me
14: R1=H, R2=Me
15: R1=R2=Me
17: R1=H, R2=Me
18: R1=R2=Me
Note: M = Na or K
Scheme 1. Reported preparation of acridine-9-carboxylic acid alkali metal salts from diphenylamines.32–34
containing a diphenylamine unit.30 It should be noted here
some extent of racemization when carried out at higher tem-
peratures than 50 ꢀC, thus to secure total inversion of config-
uration the reaction temperature was always kept at
50 ꢀC.35–37
ꢀ
that Agai and co-workers earlier reported the synthesis of
the chloro-substituted derivative of the latter macrocycle,
i.e., 9 (see Fig. 3).31
It was also reported that some diphenylamines (10–12 for
example, see Scheme 1) had been transformed to N-arylisa-
tins (13–15) then to acridine-9-carboxylic acid alkali metal
salts (16–18) as outlined in Scheme 1.32–34
As we should carry out the reaction at 50 ꢀC for obtaining its
enantiopure analogue (R,R)-21 from diol 22 and ditosylate
(S,S)-24, the preparation of achiral ligand 8 was also per-
formed at this temperature in otherwise the same conditions
than getting the highest yield at 80 ꢀC. When we carried out
the reaction of diol 22 and ditosylate 23 at 50 ꢀC we obtained
achiral ligand 8 unfortunately only in 25% yield.
Based on these observations we prepared the enantiopure di-
methylacridino-18-crown-6 ether with a potassium carbox-
ylate unit at position 9 of the acridine moiety, transformed
it to an N-allyl-carbamoyl derivative, immobilized the latter
on spherical HPLC quality silica gel and tested the new CSP
so obtained for enantiomeric separation of racemic proton-
ated primary amines containing an aromatic moiety. The
synthesis of N-allyl-carbamoyl derivative of the parent achi-
ral acridino-18-crown-6 ether, the complexation of the latter
and its enantiopure dimethyl-substituted analogue with the
enantiomers of protonated primary amines using CD spec-
troscopy are also reported here.
Macrocyclization of diol 22 and ditosylate (S,S)-24 in the
presence of K2CO3 rendered the very stable potassium tosy-
late complex of (R,R)-21, which could not be decomplexed
by chromatography on silica gel. Decomplexation was per-
formed by chromatography on neutral alumina obtaining
the free ligand (R,R)-21 in 14% yield calculated for the
precursors 22 and (S,S)-24.
We note here that the very stable potassium tosylate complex
of dimethylacridino-18-crown-6 ligand (R,R)-6 (see Fig. 2)
was reported and chromatography on neutral alumina was
also used to obtain the free macrocycle.35
2. Results and discussion
2.1. Synthesis
ꢀ
Agai and co-workers reported that for the preparation of
ligand 9 (see Fig. 3) they treated N-(5-chloro-2-hydroxy-
phenyl)-N-(2-hydroxyphenyl)formamide with 1,11-dichloro-
3,6,9-trioxaundecane using K2CO3 as a base in cyclohexa-
none followed by removal of the formyl group from the
nitrogen. In these conditions Agai and co-workers reached
a 51% yield for ligand 9.31
The synthesis of enantiopure dimethylacridino-18-crown-6
ether derivative (R,R)-19 (see Scheme 2) containing an N-
allyl-carbamoyl side chain, which makes possible the cova-
lent attachment of the chiral selector to silica gel, and the
preparation of its achiral analogue 20 are shown in Scheme 2.
ꢀ
As for a model reaction, presumably useful in the case of the
preparation of its enantiopure dimethyl-substituted analogue
(R,R)-21 (see Scheme 2), first we studied the reaction lead-
ing to the reported30 macrocycle 8. Charbonniere and Ziessel
reported a 55% yield for ligand 8 obtained by macrocycliza-
tion of diol 22 and ditosylate 23 in the presence of Cs2CO3 in
acetonitrile at 80 ꢀC. Unfortunately no other reaction condi-
tions were given.30 We changed the other reaction
parameters in wide ranges, but could not reach better yields
at 80 ꢀC than 44%. We observed, however, that substitution
of Cs2CO3 with the much cheaper K2CO3 did not affect the
yield of macrocycle 8.
Although starting from diol 25 and carrying out the macro-
cyclization in the latter conditions we hardly reached 15%
yield for ligand 26, but by using ditosylate 23 instead of
the dichloro-oligoether and substituting cyclohexanone
with acetonitrile, we obtained 52% yield for macrocycle
26 (see Scheme 2). We obtained higher (78%) yield for enan-
tiopure ligand (R,R)-27 with total inversion of configuration
of ditosylate (S,S)-24 applying the latter reaction conditions.
Removal of the formyl group from the nitrogen of macro-
cycle 26 went excellently (100% yield) using potassium
hydroxide in aqueous ethanol. In the case of macrocycle
(R,R)-27, because of the very strong potassium cation com-
plexation tendency of ligand (R,R)-21 (see above), we did
not want to use any alkali metal hydroxide, so we removed
the formyl group from the nitrogen with hydrochloric acid
in aqueous methanol obtaining a good yield (88%). In the
Our research group reported that the reactions of different
phenols with enantiopure secondary tosylates in the pres-
ence of K2CO3 using different dipolar solvents can lead to

S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
1015
K2CO3 (50 °C, MeCN)
R
R
N
X
N
X
∗
∗
OH
OH
O
R
O
O
O
O
R
∗
∗
TsO
O
OTs
3
23: R=H
(S,S)-24: R=Me
O
22: X=H
25: X=
8: R=H, X=H (25%)
(R,R)-21: R=Me, X=H (14%)
26: R=H, X=
CH
CH O (52%)
(R,R)-27: R=Me, X=CH O (78%)
KOH (EtOH - H2O)
HCl (MeOH - H2O)
26
8 (100%)
(R,R)-27
(R,R)-21 (88%)
O
O
COO K
8
1) (COCl)2
10% aq. KOH
2) I2 (PhBr)
(R,R)-21
N
N
R
O
O
O
R
O
O
O
O
R
∗
∗
∗
∗
R
O
O
O
28: R=H (32%)
30: R=H (92%)
(R,R)-29: R=Me (53%)
(R,R)-31: R=Me (90%)
CONHCH2CH=CH2
30
1) SOCl2
2) CH2=CH-CH2-NH2 / Et3N
(R,R)-31
N
R
O
O
O
O
R
∗
∗
O
20: R=H (44%)
(R,R)-19: R=Me (58%)
Scheme 2. Preparation of enantiopure dimethylacridino-18-crown-6 ether derivative (R,R)-19 and its achiral analogue 20 containing an N-allyl-carbamoyl side
chain.
latter case we used aqueous tetramethylammonium hydrox-
ide solution for neutralization to obtain the free ligand
(R,R)-21.
potassium carboxylate unit in a 90% yield. Potassium salts
30 and (R,R)-31 were first treated with thionyl chloride to ob-
tain the relevant acid chlorides, then the latters were reacted
without isolation and purification with allylamine using
triethylamine as a base to gain ligands 20 and (R,R)-19 con-
taining an N-allyl-carbamoyl side chain in acceptable yields
(44% and 58%, respectively).
Achiral macrocycle 8 was first treated with oxalyl
chloride to obtain the monoamide monoacid chloride
intermediate {6,7,9,10,12,13,15,16-octahydro-22H-dibenzo-
[n,q][1,4,7,10,13,16]pentaoxazacyclooctadecine-22-yl]-
(oxo)acetyl chloride}, which was transformed to isatin
derivative 28 without isolation and purification. Using
aluminium chloride as a catalyst for this Friedel–Crafts
type acylation we obtained only a 7% yield of isatin deriva-
tive 28, but applying iodine instead of aluminium chloridewe
reached a 32% yield of it. Isatin derivative 28 was heated with
diluted aqueous potassium hydroxide to render the achiral li-
gand 30 containing a potassium carboxylate unit in a very
good yield (92%, see Scheme 2). Starting from chiral ligand
(R,R)-21 and following the procedure applied for its achiral
analogue 8, we obtained enantiopure isatin derivative
(R,R)-29 in a 53% yield and ligand (R,R)-31 containing the
The preparation of N,N-bis(2-hydroxyphenyl)formamide
(25) needed for the synthesis of macrocycles 20 and (R,R)-
19 is shown in Scheme 3. Formamide derivative 25 was ob-
tained by a modification of the procedure reported for the
preparation of N-(5-chloro-2-hydroxyphenyl)-N-(2-hydroxy-
phenyl)formamide.38 First N-(2-methoxyphenyl)formamide
(32)39 was treated with 2-bromoanisole (33) in DowthermÒ
A using copper powder and cuprous iodide as catalysts in
the presence of sodium carbonate then the unreported
formamide derivative 34 was demethylated with anhydrous
aluminium chloride in chlorobenzene to obtain N,N-bis(2-
hydroxyphenyl)formamide (25). It should be mentioned

1016
S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
here that the melting point of diol 25 is reported in a patent,40
but no more information of it is given.
allyl-carbamoyl derivatives of both the chiral and achiral
hosts [(R,R)-19 and 20, respectively] also showed an exciton
couplet in the Bb region of the spectra (Fig. 4). The (R)-
1
NEA complex of the achiral host 20 showed a mirror image
exciton spectrum with oppositely signed bands, while the
band intensities and the amplitude of the couplet in the spec-
trum of the homochiral (R)-NEA complex of the chiral host
(R,R)-19 were much smaller than those of the (S)-NEA com-
plex. As shown in Figure 4, the homochiral complex (R,R)-
19/(R)-NEA showed practically no exciton splitting.
Cu / CuI / Na2CO3
(Dowtherm®)
+
N
H
C
H
Br
N
C
O
MeO
OMe
MeO
OMe
H
O
34 (90%)
32
33
The following consequences can be drawn from the results
of the ECD experiments:
AlCl3
(chlorobenzene)
34
25 (95%)
(i) As reported earlier,42 the (R,R)-dimethylacridino-18-
crown-6 ether host (R,R)-6 has a significant enantiomer
discriminating potential on the basis of the amplitudes
of the exciton couplet in the spectra of heterochiral
and homochiral complexes.
Scheme 3. Preparation of N,N-bis(2-hydroxyphenyl)formamide (25).
The new CSP [(R,R)-CSP-37] was prepared by a modifica-
tion of the reported procedure described by Gasparrini and
co-workers for another compound containing a terminal
double bond,41 and is outlined in Scheme 4. First spherical
HPLC quality silica gel was heated with g-mercapto-
propyl-trimethoxysilane (35) then this g-mercaptopropyl-
functionalized silica gel (36) was reacted with chiral ligand
(R,R)-19 containing the terminal double bond in the pres-
ence of free radical initiator azoisobutyronitrile (AIBN) to
form (R,R)-CSP-37.
(ii) The presence of the N-allyl-carbamoyl linker group de-
creases the amplitude of the exciton couplet in the spec-
trum of the heterochiral complex [(R,R)-19/(S)-NEA].
In the spectrum of the homochiral complex [(R,R)-19/
(R)-NEA] the intensity of the oppositely signed bands
is much lower but the significant difference between
the measured spectrum of the homochiral complex
and the sum spectrum [(R,R)-19+(R)-NEA at 1:1 M ra-
tio] still reflects significant interaction. It should be em-
phasized here that there is no direct correlation between
the stability of the heterochiral and homochiral com-
plexes and the amplitude of the exciton couplet in their
ECD spectra. A complex with a large log K (K is the
stability constant) value may give rise to an exciton
couplet with small amplitude (and vice versa) if the
geometric factors (distance and relative orientation)
of the interacting chromophore systems are not favour-
able. The long wavelength shoulder atw280 nm in the
UV spectra of the heterochiral and homochiral com-
plexes of (R,R)-19 may be a sign of p–p interaction be-
tween the aromatic chromophores of the complex and
the double bond of the allyl group. (A definite shoulder
is not seen in the spectra of the complexes of the linker-
containing achiral host 20.)
2.2. Electronic circular dichroism (ECD) spectroscopy
In an attempt to probe the effect of a linker group on the
enantiomer discriminating potential of acridino-18-crown-
6 ligands, ECD spectra of the complexes with (R)- and
(S)-NEA of chiral (R,R)-6 and its achiral parent crown ether
{2,5,8,11,14-pentaoxa-26-azatetracyclo[13.9.3.0.19,27021,25]-
heptacosa-15,17,19,21,22,24(1),26-heptaene} as well as
their N-allyl-carbamoyl derivatives [(R,R)-19 and 20, re-
spectively] were measured in acetonitrile. The chiroptical
properties of acridino-18-crown-6 ligands and their (R)-
and (S)-NEA complexes were reported earlier.42 The ECD
spectra of the complexes reflected exciton interaction.42
The ECD spectra of the (S)-NEA complexes of the N-
spherical HPLC quality
OMe
O
silica gel
spherical HPLC
HS (CH2)3 Si OMe
HS (CH2)3 Si O
O
silica gel
toluene (reflux)
OMe
35
36
O
spherical HPLC
silica gel
CONH(CH2)3S(CH2)3Si
N
O
O
36
(R,R)-19
AIBN (CHCl3)
(reflux)
Me
O
O
O
Me
∗
∗
O
O
(R,R)-CSP-37
Scheme 4. Preparation of g-mercaptopropyl-functionalized spherical HPLC quality silica gel 36 and immobilization of chiral selector (R,R)-19 on it to form the
new CSP [(R,R)-CSP-37].

S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
1017
(a)
(a)
160
80
0
-80
1
200
220
240
260
λ/nm
280
300
320
0
1
2
3
4
5
(b)
160
80
min.
(b)
0
-80
-160
1
200
220
240
260
λ/nm
280
300
320
Figure 4. ECD spectra (top) and UV spectra (bottom) of (a) heterochiral
complex of (R,R)-19 (—) with (S)-1-NEA ( ) and 20 with (S)-1-NEA
(
) compared to (S)-1-NEA ( ); (b) homochiral complex of (R,R)-19
0
1
2
3
4
5
(—) with (R)-1-NEA ( ) and 20 with (R)-1-NEA ( ) compared to
(R)-1-NEA ( ).
min.
Figure 5. Chromatograms of enantioseparations of racemic (a) 1-NEA and
(b) 2-NEA using (R,R)-CSP-37. Isocratic condition: 20% 100 mM aqueous
HClO4 solution in methanol (flow rate: 1.2 mL/min).
(iii) The achiral acridino-18-crown-6 host not having
a linker group forms complexes with the (S)- and (R)-
NEA showing mirror image spectra.42 The amplitude
of the couplet of the (S)-NEA or (R)-NEA complex is
identical and comparable to that in the spectrum of
the (R,R)-19/(S)-NEA heterochiral complex, but it is
much larger than the amplitude of the couplet of the
homochiral (R,R)-19/(R)-NEA complex.
aqueous HClO4 solution as an acidic modifier (see Fig. 5).
The shorter retention time of 2-NEA is due to the unfavour-
able position of the naphthalene ring therefore smaller the
stability constant of the complex. The elution order of the
enantiomers was determined by injection of standard au-
thentic (R)-enantiomers. It was shown in both cases that
the (R)-enantiomer eluted with a shorter retention time
than that of its antipode. This demonstrates the generally ob-
served higher stability of heterochiral complexes [(R,R)-
crown ether-(S)-ammonium salt] compared to that of homo-
chiral complexes [(R,R)-crown ether-(R)-ammonium salt].
This behaviour is in full agreement with our observation
using the (S,S)-CSP-5 containing a dimethyl-substituted pyr-
idino crown ether derivative as a chiral selector attached to
spherical HPLC quality silica gel.28
Based on the ECD studies discussed above the N-allyl-
carbamoyl group attached to the acridine ring of the chiral
host [(R,R)-19] does weaken exciton interaction between
the host and guest molecules but does not destroy the
discriminating power of the chiral host.
2.3. High performance liquid chromatography
After successful filling with (R,R)-CSP-37, the HPLC
column was tested for the enantioseparation of racemic
1-NEA and 2-NEA using isocratic conditions. In our previ-
ous studies in the case of pyridino-18-crown-6 ether selector
the chromatographic enantioseparations were performed by
using acetic acid as an acidic modifier.28 Because of the rel-
atively low basicity and the stronger complex formation ten-
dency29 of the acridino ligand, acceptable analysis times and
effective resolutions were achieved only by using 100 mM
3. Conclusion
We have shown that using bis(2-hydroxyphenyl)formamide
(25) instead of bis(2-hydroxyphenyl)amine (22) for the mac-
rocyclization with tetraethyleneglycols 23 and (S,S)-24 raises
the yields appreciably. We found that crown ethers 8 and

1018
S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
(R,R)-21 containing a diphenylamine unit could be trans-
formed to isatin derivatives 28 and (R,R)-26 in significantly
better yields if we used iodine instead of aluminium chloride
as a catalyst, although the latter has been reported to be ap-
plied widely in similar reactions. We have demonstrated
that the terminal double bond of the N-allyl-carbamoyl linker
is a very useful unit for attaching a selector covalently to g-
mercaptopropyl-functionalized spherical HPLC quality silica
gel. The ECD studies have shown that the N-allyl-carbamoyl
group attached to the acridine ring of the chiral host (R,R)-19
weakens exciton interaction between the host and guest mol-
ecules, but it does not destroy enantioselectivity. We demon-
strated that the new chiral stationary phase (R,R)-CSP-37
separated the enantiomers of racemic 1- and 2-NEA effi-
ciently. Experiments are in progress to apply this new CSP
for the enantioseparation of selected racemic protonated pri-
mary amines, amino acids and their derivatives.
and kept at this temperature with vigorous stirring until
TLC analysis showed the total consumption of the starting
materials (2 days). The solvent was removed and the residue
was taken up in a mixture of water (80 mL) and CH2Cl2
(50 mL). The aqueous phase was extracted with CH2Cl2
(3ꢁ30 mL). The combined organic phase was shaken with
water (30 mL), dried over MgSO4, filtered and the solvent
was removed. The crude product was purified by chromato-
graphy on silica gel using CH2Cl2–EtOAc (10:1) mixture as
an eluent.
The white crystals took up a mole of water on standing.
Yield: 0.98 g (25%). Mp: 136–137 ꢀC (acetonitrile); Rf:
0.3 (silica gel TLC, toluene–EtOAc 1:1); IR (KBr) nmax
3536, 3520, 3496, 3400, 3232, 3040, 1648, 1584, 1536,
1520, 1496, 1456, 1336, 1288, 1248, 1208, 1116, 1088,
948, 744 cmꢂ1; 1H and 13C NMR spectra of our compound
1
8 were slightly different than the ones reported30 for 8. H
NMR (CDCl3) d 3.70 (br s, NH, 1H), 3.76 (s, 8H), 3.94–
3.96 (m, 4H), 4.21–4.23 (m, 4H), 4.36 (br s, complexed
H2O, 2H), 6.90–6.94 (m, 6H), 7.43–7.45 (m, 4H); 13C
NMR (CDCl3) d 68.09, 69.92, 70.94, 71.26, 111.94,
117.13, 121.15, 121.47, 131.93, 148.97. Anal. Calcd for
C20H25NO5$H2O: C, 63.64; H, 7.21; N, 3.71. Found: C,
63.65; H, 7.19; N, 3.57.
4. Experimental
4.1. General
Infrared spectra were recorded on a Zeiss Specord IR 75
spectrometer. Optical rotations were taken on a Perkin–
Elmer 241 polarimeter that was calibrated by measuring
1
the optical rotations of both enantiomers of menthol. H
4.2.1.1. Macrocycle 8 from N-formyl-derivative 26. To
a vigorously stirred boiling solution of N-formyl-derivative
26 (3.9 g, 10 mmol) in EtOH (120 mL) was added under
Ar a solution of KOH (3.4 g, 52 mmol) in water (13.5 mL)
dropwise. After addition of the aqueous KOH solution the
reaction mixture was stirred at boiling temperature for
30 min and then it was let to cool down to rt. The crystals
were filtered, washed with water (2ꢁ20 mL) then with
EtOH (2ꢁ20 mL) and air-dried to get macrocycle 8 as
a monohydrate (3.78 g, 100%), which was identical in every
respect to that prepared by the procedure described above.
(500 MHz) and 13C (125 MHz) NMR spectra were obtained
on a Bruker DRX-500 Avance spectrometer. Mass spectra
were recorded on a Finningan-MAT 95 XP MS instrument
(reference compound: heptacosafluorotributylamine) using
EI (70 eV) method. Elemental analyses were performed in
the Microanalytical Laboratory of the Department of
€ €
Organic Chemistry, Institute for Chemistry, L. Eotvos
University, Budapest, Hungary. Melting points were taken
on a Boetius micro-melting point apparatus and were uncor-
rected. Starting materials were purchased from Aldrich
Chemical Company unless otherwise noted. Silica gel 60
F254 (Merck) and aluminium oxide 60 F254 neutral type E
(Merck) plates were used for TLC. Aluminium oxide (neu-
tral, activated, Brockman I) and silica gel 60 (70–230
mesh, Merck) were used for column chromatography. Ratios
of solvents for the eluents are given in volumes (mL/mL).
Solvents were dried and purified according to well-
established43 methods. Evaporations were carried out under
reduced pressure unless otherwise stated.
4.2.2. (6R,16R)-6,16-Dimethyl-6,7,9,10,12,13,15,16-octa-
hydro-22H-dibenzo[n,q][1,4,7,10,13,16]pentaoxaza-
cyclooctadecine [(R,R)-21] from 22 and (S,S)-24. Diol
22 (1.06 g, 5.3 mmol), ditosylate (S,S)-2435,36 (2.8 g,
5.3 mmol) and finely powdered anhydrous K2CO3 (7.28 g,
5.3 mmol) were mixed with vigorous stirring in anhydrous
acetonitrile (100 mL) at rt under Ar. The flask was immersed
in an oil bath and the temperature of the reaction mixture was
raised to 50 ꢀC and kept at this temperature with vigorous
stirring until TLC analysis showed the total consumption
of the starting materials (50 h). After the reaction was com-
pleted the solvent was removed and the residue was taken up
in a mixture of water (80 mL) and CH2Cl2 (80 mL). The
aqueous phase was shaken with CH2Cl2 (3ꢁ40 mL). The
combined organic phase was dried over MgSO4, filtered
and the solvent was removed.
Electronic circular dichroism measurements were per-
formed on a Jasco Dichrograph J-810 at room temperature
in quartz cell with 0.02 cm path length. The spectra were av-
eraged of five scans in the region between 190 and 320 nm in
acetonitrile. The concentration of crown ethers was 0.5 mM.
4.2. Synthesis
4.2.1. 6,7,9,10,12,13,15,16-Octahydro-22H-dibenzo[n,q]-
[1,4,7,10,16]pentaoxazacyclooctadecine (8) from 22 and
23. Diol 22 (2.1 g, 10.4 mmol), ditosylate 23 (6.1 g,
12.1 mmol) and finely powdered anhydrous K2CO3
(13.1 g, 95 mmol) were mixed with vigorous stirring in an-
hydrous acetonitrile (200 mL) at room temperature (rt) un-
der Ar. The flask was immersed in an oil bath and the
temperature of the reaction mixture was raised to 50 ꢀC
The crude product was purified by flash-chromatography on
silica gel using CH2Cl2–EtOAc (20:1) mixture as an eluent to
obtain the potassium tosylate complex of (R,R)-21 (0.58 g,
18%); Rf: 0.45 (silica gel TLC, CH2Cl2–EtOAc 3:1); [a]D25
ꢂ40.8 (c 0.34, CH2Cl2); IR (neat) nmax 3416, 1600, 1528,
1
1496, 1352, 1244, 1176, 1120, 924, 744, 654 cmꢂ1; H
NMR (CDCl3) d 1.32 (d, J¼6.5 Hz, 6H), 2.45 (s, 3H),
3.69–3.71 (m, 4H), 3.74–3.84 (m, 8H), 4.71–4.73 (m, 2H),

S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
1019
6.85–6.94 (m, 6H), 7.10 (br s, NH, 1H), 7.33 (d, J¼8 Hz, 2H),
7.39 (d, J¼7.5 Hz, 2H), 7.81 (d, J¼8 Hz, 2H).
163.28. Anal. Calcd for C21H25NO6: C, 65.10; H, 6.50; N,
3.62. Found: C, 65.39; H, 6.46; N, 3.44.
The potassium tosylate complex of (R,R)-21 (0.58 g,
0.97 mmol) was subjected to chromatography on aluminium
oxide using toluene–EtOAc (100:1) mixture as an eluent to
obtain the free ligand (R,R)-21 (0.29 g, 77%). The overall
yield calculated for diol 22 and ditosylate (S,S)-24 was
14%. The white crystals took up a mole of water on standing
in open air. Mp: 71–74 ꢀC; Rf: 0.27 (aluminium oxide TLC,
toluene–EtOAc 100:1); [a]2D5 ꢂ72.49 (c 0.578, CH2Cl2); IR
(KBr) nmax 3512, 3392, 1640, 1600, 1592, 1584, 1524, 1504,
4.2.4. (6R,16R)-6,16-Dimethyl-6,7,9,10,12,13,15,16-octa-
hydro-22H-dibenzo[n,q][1,4,7,10,13]pentaoxazacyclooc-
tadecine-22-carbaldehyde [(R,R)-27]. Macrocycle (R,R)-27
was prepared in the same way as described above for 26 start-
ing from diol 25 (9.35 g, 40.8 mmol), ditosylate (S,S)-24
(23.77 g, 44.8 mmol), finely powdered anhydrous K2CO3
(47.0 g, 340 mmol) and anhydrous acetonitrile (640 mL). In
this case the reaction was complete in 7 days at 50 ꢀC. The
crude product was purified by chromatography on silica gel
using toluene–EtOAc (first 4:1 then 1:1) mixture as eluents
to obtain macrocycle (R,R)-27 (13.2 g, 78%) as a thick oil.
Rf: 0.18 (silica gel TLC, toluene–EtOAc 1:1); [a]2D5 +29.87
(c 0.76, CH2Cl2); IR (neat) nmax 1692, 1596, 1496, 1456,
1496, 1468, 1440, 1356, 1244, 1208, 1096, 948, 748 cmꢂ1
;
1H NMR (CDCl3) d 1.32 (d, J¼6 Hz, 6H), 2.01 (br s, com-
plexed H2O, 2H), 3.66–3.83 (m, 12H), 4.50–4.59 (m, 2H),
6.78 (br s, NH, 1H), 6.80–6.94 (m, 6H), 7.37 (d, J¼8 Hz,
2H); 13C NMR (CDCl3) d 16.31, 71.34, 71.63, 74.48,
74.76, 114.73, 116.27, 120.06, 121.42, 134.41, 147.10.
Anal. Calcd for C22H29NO5$H2O: C, 65.17; H, 7.71; N,
3.45. Found: C, 65.31; H, 7.65; N, 3.50.
1376, 1332, 1272, 1256, 1244, 1124, 1048, 752, 672 cmꢂ1
;
1H NMR (CDCl3) d 1.16 (d, J¼6 Hz, 3H), 1.26 (d,
J¼6 Hz, 3H), 3.22–3.63 (m, 12H), 4.50–4.62 (m, 2H),
6.86–7.37 (m, 8H), 8.5 (s, 1H); 13C NMR (CDCl3) d (all
the five of the different aliphatic and five of the six different
aromatic carbons give double signals in the spectrum) 17.00,
17.22, 70.81, 70.86, 71.38, 71.54, 73.20, 73.28, 74.61, 74.74,
114.42, 114.72, 120.37, 120.49, 127.75, 127.93, 128.33,
130.17, 131.62, 152.76, 152.99, 163.89. Anal. Calcd for
C23H29NO6: C, 66.49; H, 7.04; N, 3.37. Found: C, 66.40;
H, 6.81; N, 3.19.
4.2.2.1. Macrocycle (R,R)-21 from N-formyl-deriva-
tive (R,R)-27. To a stirred solution of N-formyl-derivative
(R,R)-27 (1 g, 2.4 mmol) in MeOH (10 mL) was added
30% aqueous HCl solution (3.0 mL, 28 mmol) and the re-
sulting reaction mixture was stirred at rt for 10 h. To the
stirred reaction mixture was added Me4NOH$5H2O (6 g,
33 mmol) dissolved in water (60 mL). Most of the MeOH
and about half amount of the water were distilled off, and
the aqueous solution was extracted with ether (1ꢁ100 and
2ꢁ50 mL). The combined organic phase was shaken with
water (30 mL), dried over MgSO4, filtered and the solvent
was removed. The crude product (R,R)-21 (0.86 g, 88%)
so obtained was identical in every respect to that prepared
by the procedure described above.
4.2.5. 9,10,12,13,15,16,18,19-Octahydroindolo[7,1-n,o]-
[1,4,7,10,13,16]benzopentaoxazacyclooctadecine-1,2-di-
one (28). To a vigorously stirred solution of macrocycle 8
(6.5 g, 18.08 mmol) in dry and pure toluene (20 mL) was
added dropwise at rt under Ar oxalyl chloride (3.44 mL,
5.0 g, 39 mmol). The reaction mixture was stirred at rt for
10 min and at 70 ꢀC for 1 h.
4.2.3. 6,7,9,10,12,13,15,16-Octahydro-22H-dibenzo[n,q]-
[1,4,7,10,13,16]pentaoxazacyclooctadecine-22-carbalde-
hyde (26). Diol 25 (4.65 g, 22.3 mmol) ditosylate 23 (11.2 g,
22.3 mmol) and finely powdered anhydrous K2CO3 (22.0 g,
159 mmol) were mixed in anhydrous acetonitrile (200 mL)
with vigorous stirring at rt under Ar. The temperature of
the reaction mixture was raised to 50 ꢀC and kept at this tem-
perature for 65 h. The cold reaction mixture was filtered and
the precipitate was washed with acetonitrile (3ꢁ20 mL).
The solvent was removed from the combined acetonitrile so-
lution and the residue was taken up in a mixture of water
(20 mL) and CH2Cl2 (60 mL). The phases were shaken
well and separated. The organic phase was dried over
MgSO4, filtered and the solvent was removed. The crude
product was purified by chromatography on silica gel using
toluene–EtOAc (first 1:1 then 1:2) mixtures as eluents to ob-
tain macrocycle 26 (4.05 g, 52%) as colourless crystals. Mp:
131 ꢀC; Rf: 0.31 (silica gel TLC, EtOAc); IR (KBr) nmax
1692, 1600, 1504, 1464, 1336, 1280, 1148, 1136, 1128,
The volatile materials were removed and the crude
6,7,9,10,12,13,15,16-octahydro-22H-dibenzo[n,q][1,4,7,10,-
13,16]pentaoxazacyclooctadecine-22-yl(oxo)acetyl chlo-
ride was dissolved in bromobenzene (80 mL). To the latter
solution was added iodine at rt with stirring under Ar
(0.4 g, 1.58 mmol). The reaction mixture was stirred at re-
flux temperature for 40 h and the solvent was removed
(1 mmHg). The crude product was purified by chromato-
graphy on silica gel using CH2Cl2–EtOAc (4:1) as an eluent
to give 28 (2.4 g, 32%) as red crystals. Mp: 116–118 ꢀC; Rf:
0.34 (silica gel TLC, EtOAc); IR (KBr) nmax 1736, 1608,
1
1500, 1446, 1368, 1308, 1280, 1112, 784, 760 cmꢂ1; H
NMR (CDCl3) d 3.09–3.13 (m, 1H), 3.25–3.56 (m, 8H),
3.65–3.73 (m, 4H), 4.03–4.07 (m, 1H), 4.14–4.18 (m, 1H),
4.29–4.34 (m, 1H), 6.98–7.06 (m, 2H), 7.16 (d, J¼8 Hz,
1H), 7.21–7.31 (m, 3H), 7.36 (t, J¼8 Hz, 1H); 13C NMR
(CDCl3) d 68.69, 69.07, 70.00, 70.13, 71.05, 71.21, 71.53,
71.61, 113.38, 117.78, 119.19, 120.44, 124.33 (very high,
probably two carbon 13 signals together), 125.11, 129.10,
129.88, 140.87, 146.01, 155.40, 158.53, 183.72. Anal. Calcd
for C22H23NO7: C, 63.91; H, 5.61; N, 3.39. Found: C, 63.72;
H, 5.31; N, 3.18.
1120, 1108, 1088, 1040, 760 cmꢂ1 1H NMR (CDCl3)
;
d 3.63–3.66 (m, 8H), 3.71–3.73 (m, 2H), 3.80–3.82 (m,
2H), 4.07–4.11 (m, 4H), 6.90–7.25 (m, 8H), 8.56 (s, 1H);
13C NMR (CDCl3) d (all the four of the different aliphatic
and five of the six different aromatic carbons give double
signals in the spectrum) 59.72, 68.60, 68.79, 69.60, 69.90,
70.87, 71.37, 71.59, 114.16, 114.89, 121.16, 126.99,
127.90, 128.65, 128.88, 129.41, 131.28, 153.54, 154.49,
4.2.6. (9R,19R)-9,19-Dimethyl-9,10,12,13,15,16,18,19-
octahydroindolo[7,1-n,o][1,4,7,10,13,16]benzopentaoxa-
zacyclooctadecine-1,2-dione [(R,R)-29]. Macrocycle
(R,R)-29 was prepared in the same way as described above

1020
S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
for 28 starting from (R,R)-21 (10.8 g, 27.87 mmol), dry and
pure toluene (100 mL) and oxalyl chloride (5.3 mL, 7.71 g,
60.8 mmol). Crude (6R,16R,)-6,16-dimethyl-6,7,9,10,12,
13,15,16-octahydro-22H-dibenzo[n,q][1,4,7,10,13]pentaox-
azacyclooctadecine-22-yl(oxo)acetyl chloride, bromobenz-
ene (123 mL) and iodine (0.6 g, 2.37 mmol) were reacted
as above. The crude product was purified by chromatography
on silica gel using CH2Cl2–EtOAc (7:1) as an eluent to give
(R,R)-29 (6.5 g, 53%) as red crystals. Mp: 137–139 ꢀC; Rf:
0.16 (silica gel TLC, CH2Cl2–EtOAc, 7:1); IR (KBr) nmax
1740, 1696, 1608, 1500, 1460, 1372, 1312, 1280, 1240,
48.9 g, 411 mmol) was added with vigorous stirring potas-
sium salt 30 (2.35 g, 5 mmol) in portions at 0 ꢀC. The suspen-
sion was stirred at 0 ꢀC for 5 min then the temperature of the
reaction was raised to reflux and the mixture was kept boiling
for 45 min. The excess thionyl chloride was removed by dis-
tillation then dry and pure toluene (10 mL) was added to the
residue. The toluenewas removed and the residuewas treated
with toluene (10 mL) as above to remove the traces of thionyl
chloride. The crude acid chloride was suspended in pure and
dry CH2Cl2 (40 mL) and to this suspension was added drop-
wise at 0 ꢀC a mixture of allylamine (5 mL, 3.81 g, 67 mmol)
and triethylamine (10 mL, 7.26 g, 72 mmol) dissolved in
pure and dry CH2Cl2 (20 mL) under Ar. The reaction mixture
was stirred at 0 ꢀC for 5 min then at rt for 1 h. The volatile
materials were removed and the solid residue was triturated
thoroughly with water (100 mL). The yellow crystals were
filtered off, washed with water (3ꢁ20 mL) and dried. The
crude product was recrystallized from MeOH to give 20
(1.04 g, 44%) as a monohydrate. Mp: 274–6 ꢀC (MeOH);
IR (KBr) nmax 3450, 3336,1656, 1624, 1568, 1536, 1472,
1136, 1096, 760 cmꢂ1 1H NMR (CDCl3) d 0.82 (d,
;
J¼6 Hz, 3H), 1.16 (d, J¼6 Hz, 3H), 3.04–3.15 (m, 2H),
3.34–3.65 (m, 10H), 4.33–4.36 (m, 1H), 4.57–4.60 (m,
1H), 6.98 (t, J¼8 Hz, 1H), 7.03 (t, J¼8 Hz, 1H), 7.09 (d,
J¼8 Hz, 1H), 7.25–7.28 (m, 3H), 7.35 (t, J¼8 Hz, 1H); 13C
NMR (CDCl3) d 16.81, 17.32, 70.88, 71.09, 71.19, 71.37,
74.47, 74.53, 74.82, 74.93, 113.72, 117.53, 119.24, 120.21,
124.25, 125.42, 126.01, 129.79, 129.85, 141.55, 145.20,
154.48, 158.73, 183.86. Anal. Calcd for C24H27NO7: C,
65.29; H, 6.16; N, 3.17. Found: C, 65.05; H, 5.95; N, 2.94.
1
1424, 1280, 1112, 924, 744 cmꢂ1; H NMR (DMSO-d6)
d 3.69–3.77 (m, 8H), 3.97–4.01 (m, 4H), 4.12–4.15 (m,
2H), 4.32–4.36 (m, 4H), 5.21–5.38 (m, 2H), 6.01–6.10 (m,
1H), 7.21 (d, J¼7.0 Hz, 2H), 7.48–7.60 (m, 4H), 9.20 (br s,
NH, 1H), (the proton signal of the complexed water molecule
merged into the big broad water singlet of DMSO-d6 centred
at 3.43 ppm); 13C NMR (DMSO-d6) d 41.46, 68.55, 68.71,
69.26, 70.41, 107.75, 116.07, 116.66, 123.00, 127.19,
134.91, 139.93, 141.01, 154.78, 166.26. MS Calcd for
C25H28N2O6: 452.1942. Found: 452.1921. Anal. Calcd for
C25H28N2O6$H2O: C, 63.82; H, 6.43; N, 5.95. Found: C,
63.45; H, 6.46; N, 5.71.
4.2.7. Potassium 6,7,9,10,12,13,15,16-octahydro-1,21-
methenobenzo[n,q][1,4,7,10,13,16]pentaoxazacycloocta-
decine-23-carboxylate (30). Isatin derivative 28 (2.4 g,
5.8 mmol) was suspended at rt in 10% aqueous KOH
(32 mL) with vigorous stirring and this suspension was
heated to reflux and then refluxed for 2 h. The reaction mix-
ture was stored at rt for 3 h and in a refrigerator for a day. The
yellow precipitate was filtered off and recrystallized from
boiling water to give 30 (2.5 g, 92%) as a bright yellow
monohydrate. Mp: >340 ꢀC; IR (KBr) nmax 3496, 1608,
1564, 1504, 1472, 1456, 1424, 1376, 1324, 1280, 1256,
1
1160, 1104, 956, 744, 616, 512 cmꢂ1; H NMR (DMSO-
4.2.10. (6R,16R)-N-Allyl-6,16-dimethyl-6,7,9,10,
12,13,15,16-octahydro-1,21-methenodibenzo[n,q][1,4,7,
10,13,16]pentaoxazacyclooctadecine-23-carboxamide
[(R,R)-19)]. Optically active ligand (R,R)-19 was prepared
in the same way as described above for its achiral analogue
20 starting from potassium salt (R,R)-31 (2.49 g, 5 mmol)
and thionyl chloride (30 mL, 48.9 g, 411 mmol). Yellow
crystals as monohydrates; yield: 1.45 g (58%). Mp: 178 ꢀC
(MeOH); [a]2D5 ꢂ70.8 (c 1.97, CH2Cl2); IR (KBr) nmax
3450, 3248, 1656, 1568, 1464, 1376, 1276, 1160, 1144,
d6) d 3.67–3.70 (m, 4H), 3.75–3.78 (m, 4H), 4.03–4.06 (m,
4H), 4.36–4.39 (m, 4H), 7.20 (d, J¼8 Hz, 2H), 7.40 (t,
J¼8 Hz, 2H), 7.70 (d, J¼8 Hz, 2H); 13C NMR (DMSO-d6)
d 67.80, 67.94, 68.56, 70.33, 107.89, 120.67, 122.25,
124.22, 140.05, 153.28, 156.24, 169.78. Anal. Calcd for
C22H22NO7K$H2O: C, 56.28; H, 5.15; N, 2.98. Found: C,
56.07; H, 5.38; N, 2.81.
4.2.8. Potassium (6R,16R)-6,16-dimethyl-6,7,9,10,
12,13,15,16-octahydro-1,21-methenodibenzo[n,q][1,4,7,
10,13,16]pentaoxazacyclooctadecine-23-carboxylate
[(R,R)-31]. Potassium salt (R,R)-31 was prepared in the same
way as described above for its achiral analogue 30 starting
from (R,R)-29 (6.5 g, 14.72 mmol) and 10% aqueous KOH
(95 mL). Bright yellow crystals as monohydrates; yield:
6.6 g (90%). Mp: >340 ꢀC; [a]2D5 +8.29 (c 0.916, MeOH);
IR (KBr) nmax 3440, 1596, 1576, 1560, 1472, 1376, 1320,
1
1120, 1088, 1080, 940, 752 cmꢂ1; H NMR (CDCl3) d 2.7
(br s, complexed H2O, 2H), 1.45 (d, J¼6.5 Hz, 6H), 3.71–
4.01 (m, 12H), 4.28–4.31 (m, 2H), 5.11–5.15 (m, 2H),
5.25–5.38 (m, 2H), 6.02–6.10 (m, 1H), 6.60 (br s, NH, 1H),
7.09 (d, J¼7 Hz, 2H), 7.41 (t, J¼8 Hz, 2H), 7.56 (d,
J¼8.5 Hz, 2H); 13C NMR (CDCl3) d 16.11, 42.41, 71.35
(very high, probably two carbon 13 signals together),
74.58, 75.30, 112.71, 117.34, 117.50, 123.57, 127.14,
133.68, 140.15, 141.50, 153.80, 167.19. MS Calcd for
C27H32N2O6: 480.2255. Found: 480.2246. Anal. Calcd for
C27H32N2O6$H2O: C, 65.04; H, 6.87; N, 5.62. Found: C,
64.97; H, 6.89; N, 5.43.
1272, 1152, 1128, 1100, 956, 748 cmꢂ1
;
1H NMR
(DMSO-d6) d 1.42 (d, J¼6 Hz, 6H), 3.62–3.86 (m, 10H),
4.09 (d, J¼10 Hz, 2H), 5.05–5.08 (m, 2H), 7.25 (d,
J¼8 Hz, 2H), 7.38 (t, J¼8 Hz, 2H), 7.67 (J¼8 Hz); 13C
NMR (DMSO-d6) d 14.50, 68.77, 70.01, 72.73, 72.81,
109.26, 120.35, 122.40, 124.26, 140.72, 151.81, 152.64,
168.38. Anal. Calcd for C24H26NO7K$H2O: C, 57.93; H,
5.67; N, 2.81. Found: C, 58.22; H, 5.58; N, 2.62.
4.2.11. N,N-Bis(2-methoxyphenyl)formamide (34). A
mixture
of
N-(2-methoxyphenyl)formamide
(32)
(23.5 g, 0.155 mol), 2-bromoanisole (33) (21.5 mL, 32.1 g,
0.172 mol), finely powdered anhydrous Na2CO3 (62 g,
0.585 mol), CuI (1.9 g, 0.01 mol) and finely powdered cop-
per (0.8 g, 0.013 mol) was suspended in DowthermÒ
(120 mL) with vigorous stirring at rt under Ar. The
4.2.9. N-Allyl-6,7,9,10,12,13,15,16-octahydro-1,21-meth-
enodibenzo[n,q][1,4,7,10,13,16]pentaoxazacyclooctade-
cine-23-carboxamide (20). To thionyl chloride (30 mL,

S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
1021
temperature of the reaction mixturewas raised to 220–225 ꢀC
and kept in this range for 8 h. During the reaction, water con-
densed on the surface of the condenser was removed in every
couple of hours. After 8 h 2-bromoanisole (33) (5 mL, 7.5 g,
0.04 mol), finely powdered anhydrous Na2CO3 (20 g,
0.189 mol), CuI (0.9 g, 0.005 mol) and finely powdered cop-
per (0.5 g, 0.008 mol) were added to the reaction mixture and
it was stirred vigorously for 4 h at 220–225 ꢀC under Ar. The
brown reaction mixture was cooled to 90 ꢀC, the insoluble
materials were filtered off and washed with EtOAc
(3ꢁ80 mL). Filtrate and washings were combined and all
the volatile materials were removed by distillation (first at
5 mmHg then at 0.01 mmHg, in the end using an oil bath
of 110 ꢀC). The residue was dissolved in EtOAc (400 mL)
and this solution was shaken successively with water
(130 mL), 10% aqueous NaHSO3 solution (50 mL) and brine
(80 mL). The organic phase was dried over MgSO4, fil-
tered, heated with charcoal for 10 min, filtered again
and the solvent was removed. The crude product was trit-
urated with ether and the ethereal solution containing lot
of precipitate was stored in a refrigerator overnight. The
almost colourless crystals were filtered and recrystallized
from a MeOH–ether mixture to give pure 34 (36 g,
90%). Mp: 106–107 ꢀC (MeOH–ether); Rf: 0.47 (silica
gel TLC, CH2Cl2–acetone, 20:1); IR (KBr) nmax 1696,
1592, 1534, 1450, 1336, 1280, 1240, 1120, 1028,
(3.0 g) in dry and pure toluene (60 mL) was heated with
mechanical stirring at reflux temperature under Ar for
24 h, during which time the water formed was removed
azeotropically using a Dean–Stark adapter. After addition
of (g-mercaptopropyl)-trimethoxysilane (0.8 mL, 0.85 g,
4.3 mmol) the mixture was stirred in the above conditions
for 2 days. The mixture was cooled down to rt and the mod-
ified silica gel was collected by filtration, washed sequen-
tially with toluene (50 mL), CH2Cl2 (50 mL), MeOH
(50 mL), CH2Cl2 (50 mL) and dried at 80 ꢀC under reduced
pressure for 14 h. A sample of blank silica gel was dried in
the same way and it gave a combustion analysis of C,
1.04; H, 1.14; N, 0.00; S, 0.00. The combustion analysis of
modified silica gel 36 gave C, 4.26; H, 1.76; N, 0.00; S,
2.63. This result shows that each gram of 36 contained
0.89 mmol (by C%), 0.86 mmol (by H%) and 0.82 mmol
(by S%) of g-mercaptopropyl-silyl groups.
4.2.14. Chiral stationary phase (R,R)-CSP-37. A slurry of
g-mercaptopropyl-functionalized silica gel 36 (2.7 g), mac-
rocycle (R,R)-19 (389 mg, 0.78 mmol) and 2,20-azobis(2-
methylpropionitrile) (AIBN) (42 mg, 0.26 mmol) in freshly
distilled pure and dry CHCl3 (25 mL) was heated with me-
chanical stirring at reflux temperature under Ar for 2 days;
two additional portions of AIBN (42 mg each) were added
after 2 and 4 days. After cooling to rt the modified silica
gel [(R,R)-CSP-37] was collected by filtration, washed se-
quentially with CH2Cl2 (60 mL), MeOH (60 mL), CH2Cl2
(60 mL) and dried at 80 ꢀC under reduced pressure for
14 h. The combustion analysis of (R,R)-CSP-37 gave C,
8.51; H, 2.24; N, 0.42; S, 2.28. Loading density of macro-
cycle, based on N: 0.15 mmol/g.
740 cmꢂ1 1H NMR (CDCl3) d 3.83 (s, 3H), 3.84 (s,
;
3H), 6.90–6.99 (m, 4H), 7.22–7.29 (m, 4H), 8.40 (s,
1H); 13C NMR (CDCl3) d (five aromatic carbons and
the methyl carbon give double signals in the spectrum)
56.01, 56.07, 112.41, 112.55, 121.08, 128.73, 128.78,
129.05, 129.20, 129.67, 130.81, 155.07, 155.34, 163.54.
Anal. Calcd for C15H15NO3: C, 70.02; H, 5.88; N, 5.44.
Found: C, 70.15; H, 5.80; N, 5.38.
4.3. Chromatography
4.2.12. N,N-Bis(2-hydroxyphenyl)formamide (25). To
a vigorously stirred suspension of anhydrous AlCl3 (107 g,
0.8 mol) in dry and pure chlorobenzene (500 mL) was added
N,N-bis(2-methoxyphenyl)formamide (34) (51.5 g, 0.2 mol)
in portions at rt under Ar. The temperature of the reaction
mixture was raised to 70 ꢀC and kept at this temperature
for 4 h. The reaction mixture was cooled down to rt and it
was poured into a vigorously stirred mixture of 37% aqueous
HCl (22 mL) and ice-water (1.5 L). Stirring was continued
for another hour, the pale yellow crystals were filtered off
and washed with water (3ꢁ150 mL). The crude product
was dried and recrystallized from a mixture of EtOAc and
hexane to give pure 25 (43.6 g, 95%). Mp: 159–159.5 ꢀC
(EtOAc–hexane). Reported mp: 152–153 ꢀC;40 Rf: 0.36 (sil-
ica gel TLC, CH2Cl2–acetone, 10:1); IR (KBr) nmax 3550–
2300 (br), 3288, 1654, 1592, 1512, 1492, 1454, 1352,
The chiral column was prepared by packing (R,R)-CSP-37
into a 150ꢁ 4.6 mm stainless steel empty HPLC column us-
ing a slurry packing method. The packing was performed by
using a Haskel-pump at 350 bar.
Chromatography was performed on a Jasco-HPLC instru-
ment. Chromatograms were obtained by using (A) 100 mM
aqueous HClO4 solution and (B) methanol as eluents.
Acknowledgements
This work was supported by the Hungarian Scientific
Research Fund (OTKA K 62654 to P.H. and T 49792, NI
68466 to M.H.) and by the Ministry of Education of Hungary
(Postdoctoral Fellowship PAL 11/2003 to T.T.).
1
1272, 1240, 1152, 824, 752 cmꢂ1; H NMR (DMSO-d6)
d 6.78–6.81 (m, 2H), 6.92–6.99 (m, 2H), 7.11–7.18 (m,
4H), 8.29 (s, 1H), 9.74 (br s, OH, 2H); 13C NMR (DMSO-
d6) d (all the aromatic carbons give double signals in the
spectrum) 116.56, 116.65, 119.20, 119.50, 127.71, 128.31,
128.54, 128.58, 129.10, 129.42, 152.78, 153.20, 162.82.
Anal. Calcd for C13H11NO3: C, 68.11; H, 4.84; N, 6.11.
Found: C, 68.32; H, 4.58; N, 6.00.
References and notes
1. Subramanian, G. Chiral Separation Techniques: A Practical
Approach, Third completely revised and updated edition;
Wiley-VCH: Weinheim, 2006.
2. Hyun, M. H. J. Sep. Sci. 2003, 26, 242–250.
4.2.13. Preparation of g-mercaptopropyl-functionalized
spherical HPLC quality silica gel (36). A slurry of spheri-
cal HPLC quality silica gel (SuperspherÒ Si 60, Merck)
3. Hyun, M. H. Bull. Korean Chem. Soc. 2005, 26, 1153–1163.
4. Sousa, L. R.; Sogah, G. D. Y.; Hoffman, D. H.; Cram, D. J.
J. Am. Chem. Soc. 1978, 100, 4569–4576.

1022
S. Lakatos et al. / Tetrahedron 64 (2008) 1012–1022
€ €
25. Kontos, Z.; Huszthy, P.; Bradshaw, J. S.; Izatt, R. M.
Tetrahedron: Asymmetry 1999, 10, 2087–2099.
5. Sogah, G. D. Y.; Cram, D. J. J. Am. Chem. Soc. 1979, 101,
3035–3042.
€ €
26. Kontos, Z.; Huszthy, P.; Bradshaw, J. S.; Izatt, R. M.
Enantiomer 2000, 5, 561–566.
6. Shinbo, T.; Yamaguchi, T.; Nishimura, K.; Sugiura, M.
J. Chromatogr. 1987, 405, 145–153.
ꢀ
ꢀ ꢀ
7. Shinbo, T.; Yamaguchi, T.; Yanagishita, H.; Kitamoto, D.;
Sakaki, K.; Sugiura, M. J. Chromatogr. 1992, 625, 101–
108.
27. Horvath, G.; Huszthy, P.; Szarvas, S.; Szokan, G.; Redd, J. T.;
Bradshaw, J. S.; Izatt, R. M. Ind. Eng. Chem. Res. 2000, 39,
3576–3581.
ꢀ
ꢀ
28. Farkas, V.; Toth, T.; Orosz, G.; Huszthy, P.; Hollosi, M.
8. Hyun, M. H.; Han, S. C.; Lipshutz, B. H.; Shin, Y. J.; Welch,
C. J. J. Chromatogr., A 2001, 910, 359–365.
9. Hyun, M. H.; Han, S. C.; Lipshutz, B. H.; Shin, Y. J.; Welch,
C. J. J. Chromatogr., A 2002, 959, 75–83.
10. Hyun, M. H.; Tan, G.; Cho, Y. J. Biomed. Chromatogr. 2005,
19, 208–213.
Tetrahedron: Asymmetry 2006, 17, 1883–1889.
29. Prodi, L.; Bolletta, F.; Montalti, M.; Zaccheroni, N.; Huszthy,
P.; Samu, E.; Vermes, B. New J. Chem. 2000, 24, 781–785.
30. Charbonniere, L. J.; Ziessel, R. F. Tetrahedron Lett. 2000, 41,
2373–2376.
ꢀ
ꢀ
€
11. Machida, Y.; Nishi, H.; Nakamura, K.; Nakai, H.; Sato, T.
J. Chromatogr., A 1998, 805, 85–92.
31. Agai, B.; Nemeth, V.; Bocskei, Z.; Simon, K.; Bitter, I.; TTke,
L. Tetrahedron 1996, 52, 6713–6724.
ꢀ
12. Hyun, M. H.; Jin, J. S.; Lee, W. J. J. Chromatogr., A 1998, 822,
155–161.
32. Stolle, R. Chem. Ber. 1913, 46, 3915–3916.
33. Friedlander, P.; Kunz, K. Chem. Ber. 1922, 55, 1597–1607.
34. Newman, M. S.; Peowell, W. H. J. Org. Chem. 1961, 26, 812–
815.
ꢀ
13. Berkecz, R.; Sztojkov-Ivanov, A.; Ilisz, I.; Forro, E.; Fulop, F.;
€ €
ꢀ
Hyun, M. H.; Peter, A. J. Chromatogr., A 2006, 1125, 138–143.
ꢀ
35. Huszthy, P.; Samu, E.; Vermes, B.; Mezey-Vandor, G.;
14. Hyun, M. H.; Song, Y.; Cho, Y. J.; Kim, D. H. J. Chromatogr., A
2006, 1108, 208–217.
15. Hyun, M. H. J. Sep. Sci. 2006, 29, 750–761.
16. Tan, G. H.; Xue, J. Y.; Hyun, M. H. J. Sep. Sci. 2006, 29, 1407–
1411.
ꢀ
Nogradi, M.; Bradshaw, J. S.; Izatt, R. M. Tetrahedron 1999,
55, 1491–1504.
ꢀ
ꢀ
36. Samu, E.; Huszthy, P.; Somogyi, L.; Hollosi, M. Tetrahedron:
Asymmetry 1999, 10, 2775–2795.
ꢀ ꢀ ꢀ ꢀ
37. Szalay, L.; Farkas, V.; Vass, E.; Hollosi, M.; Moczar, I.; Pinter,
17. Hyun, M. H.; Choi, H. J.; Kang, B. S.; Tan, G.; Cho, Y. J. Bull.
Korean Chem. Soc. 2006, 27, 1775–1779.
ꢀ
A.; Huszthy, P. Tetrahedron: Asymmetry 2004, 15, 1487–1493.
ꢀ
38. Farkas, M.; Lang, G.; Dinya, Z.; Sohar, P.; Rozsa, L.; Budai, Z.
18. Hyun, M. H.; Cho, Y. J.; Song, Y.; Choi, H. J.; Kang, B. S.
Chirality 2007, 19, 74–81.
19. Hirose, K.; Nakamura, T.; Nishioka, R.; Ueshige, T.; Tobe, Y.
Tetrahedron Lett. 2003, 44, 1549–1551.
J. Mol. Struct. 1985, 131, 131–140.
39. Werner, W. Ger. Offen. DE 2,917,087, 6 November 1980; p 15;
Chem. Abstr. 1980, 94, 120892e.
ꢀ
20. Hirose, K.; Jin, Y. Z.; Nakamura, T.; Nishioka, R.; Ueshige, T.;
Tobe, Y. Chirality 2005, 17, 142–148.
40. Rozsa, L.; PetTcz, L.; Grasser, K.; Kosoczky, I.; Kiszelly, E.;
Nagy, J. Ger. Offen. DE 2,833,892, 22 February 1979, p 46;
21. Hirose, K.; Jin, Y. Z.; Nakamura, T.; Nishioka, R.; Ueshige, T.;
Tobe, Y. J. Chromatogr., A 2005, 1078, 35–41.
Chem. Abstr. 1979, 90, 187009h.
41. Gasparrini, F.; Misiti, D.; Villani, C.; Borchardt, A.; Burger,
M. T.; Still, W. C. J. Org. Chem. 1995, 60, 4314–4315.
ꢀ
42. Szarvas, S.; Majer, Z.; Huszthy, P.; Vermes, B.; Hollosi, M.
Enantiomer 2002, 7, 241–249.
43. Riddick, J. A.; Burger, W. B. Organic Solvents; Weissberger,
A., Ed.; Techniques of Organic Chemistry; Wiley-
Interscience: New York, NY, 1970; Vol. II.
22. Jin, Y. Z.; Hirose, K.; Nakamura, T.; Nishioka, R.; Ueshige, T.;
Tobe, Y. J. Chromatogr., A 2006, 1129, 201–207.
23. Bradshaw, J.S.;Huszthy, P.;Wang, T. M.;Zhu,C. Y.;Nazarenko,
A. Y.; Izatt, R. M. Supramol. Chem. 1993, 1, 267–275.
24. Huszthy, P.; Bradshaw, J. S.; Bordunov, A. V.; Izatt, R. M. Acta
Chim. Hung.—Models Chem. 1994, 131, 445–454.