Selective synthesis of three conformational isomers of tetrakis[(ethoxycarbonyl)methoxy]thiacalix[4]arene and their complexation properties towards alkali metal ions

Article abstract of DOI:10.1039/a803734e

5,11,17,23-Tetra-tert-butyl-2,8,14,20-tetrathiacalix[4]arene-25,26,27,28- tetrol (TCA) underwent facile tetra-O-alkylation by treatment with ethyl bromoacetate in the presence of an alkali carbonate as base catalyst in DMF or acetone to provide a mixture of conformational isomers (cone, partial cone, and 1,3-alternate) of 5,11,17,23-tetra-tert-butyl-25,26,27,28-tetrakis[(ethoxycarbonyl)methoxy]-2,8, 14,20-tetrathiacalix[4]arene(1), the stereochemistries of which were unambiguously assigned by ¹H NMR and X-ray analysis. The isomer distribution depended significantly on the base used, thus providing a facile route for the preparation of a particular conformer; Na₂CO₃, K₂CO₃, and Cs₂CO₃ gave cone- (77% yield), partial-cone- (58% yield), and 1,3-alternate-1 (78% yield) in acetone, respectively. Cone- and partial-cone-1, in turn, showed preference for Na⁺ and K⁺, respectively, in an ion-pair extraction study, while 1,3-alternate-1 preferred most Rb⁺ ion, followed by K⁺ and then Cs⁺. These results imply that the size of the cavities provided by the (ethoxycarbonyl)methoxy groups arranged on the periphery of the thiacalix[4]arene skeleton is in the order cone- < partial-cone- < 1,3-alternate-1. The ion selectivity of cone-1 was rather better than that of the methylene-bridged counterpart. 5,11,17,23-tetra-tert-butyl-25,26,27,28-tetrakis[(ethoxycarbonyl)methoxy] calix[4]arene (2). The stoichiometry of the complex of cone-1 with Na⁺ ion was determined to be 1:1 with the stability constant of 10^2.85 mol^-1 dm³ in 50 (v/v)% CDCl₃-CD₃OD.

Full text of DOI:10.1039/a803734e

View Article Online / Journal Homepage / Table of Contents for this issue
Selective synthesis of three conformational isomers of
tetrakis[(ethoxycarbonyl)methoxy]thiacalix[4]arene and their
complexation properties towards alkali metal ions†
Nobuhiko Iki,* Fumitaka Narumi, Toyohisa Fujimoto, Naoya Morohashi and Sotaro Miyano*
Department of Biomolecular Engineering, Graduate School of Engineering, Tohoku University,
Aramaki-Aoba 07, Aoba-ku, Sendai 980-8579, Japan
Received (in Cambridge) 18th May 1998, Accepted 21st September 1998
5,11,17,23-Tetra-tert-butyl-2,8,14,20-tetrathiacalix[4]arene-25,26,27,28-tetrol (TCA) underwent facile tetra-O-
alkylation by treatment with ethyl bromoacetate in the presence of an alkali carbonate as base catalyst in DMF or
acetone to provide a mixture of conformational isomers (cone, partial cone, and 1,3-alternate) of 5,11,17,23-tetra-
tert-butyl-25,26,27,28-tetrakis[(ethoxycarbonyl)methoxy]-2,8,14,20-tetrathiacalix[4]arene (1), the stereochemistries
of which were unambiguously assigned by 1H NMR and X-ray analysis. The isomer distribution depended
significantly on the base used, thus providing a facile route for the preparation of a particular conformer; Na2CO3,
K2CO3, and Cs2CO3 gave cone- (77% yield), partial-cone- (58% yield), and 1,3-alternate-1 (78% yield) in acetone,
respectively. Cone- and partial-cone-1, in turn, showed preference for Naϩ and Kϩ, respectively, in an ion-pair
extraction study, while 1,3-alternate-1 preferred most Rbϩ ion, followed by Kϩ and then Csϩ. These results imply
that the size of the cavities provided by the (ethoxycarbonyl)methoxy groups arranged on the periphery of the
thiacalix[4]arene skeleton is in the order cone- < partial-cone- < 1,3-alternate-1. The ion selectivity of cone-1
was rather better than that of the methylene-bridged counterpart, 5,11,17,23-tetra-tert-butyl-25,26,27,28-
tetrakis[(ethoxycarbonyl)methoxy]calix[4]arene (2). The stoichiometry of the complex of cone-1 with Naϩ ion
was determined to be 1:1 with the stability constant of 102.85 molϪ1 dm3 in 50 (v/v)% CDCl3–CD3OD.
isomers of 5,11,17,23-tetra-tert-butyl-25,26,27,28-tetrakis[(eth-
Introduction
oxycarbonyl)methoxy]calix[4]arene (2) which were found to be
able to recognize particular alkali metal ions selectively.6 Also
Reinhoudt et al. studied the effect of solvent, temperature, and
base catalysts on the conformational distribution in the form-
ation of ethers in calix[4]arenes.7
In the last decade, calixarenes have been drawing much atten-
tion as key compounds in molecular recognition chemistry and
supramolecular chemistry.1 Almost all the calixarenes have the
metacyclophane skeleton bridged by methylene groups due to
their ease of synthesis via condensation of p-alkylphenols and
formaldehyde under alkaline conditions. The general strategy
used to develop the important abilities of these molecules such
as sensing2 and separation3 of metal ions and organic mole-
cules has relied on the modification of the upper4 and/or
lower5 rims of the parent calixarenes. For example, Shinkai et
al. reported the introduction of (ethoxycarbonyl)methyl groups
into 5,11,17,23-tetra-tert-butylcalix[4]arene-25,26,27,28-tetrol
(CA) by direct O-alkylation. They obtained the conformational
By contrast, development of new functions via replacement
of the bridge methylene groups by hetero atoms remains to be
exploited. Sone et al. first reported the synthesis of 5,11,17,23-
tetra-tert-butyl-2,8,14,20-tetrathiacalix[4]arene-25,26,27,28-
tetrol (TCA) by acid catalysed cyclization of an acyclic
tetramer.8 Although the synthesis allowed them to carry out 1H
NMR studies of TCA,9 little is known about the other proper-
ties and applications of this cyclic entity because of the dif-
ficulty in obtaining it in substantial quantities. Recently, we
reported a convenient and facile synthesis of TCA in a satisfac-
tory yield by simply heating p-tert-butylphenol with elemental
sulfur in the presence of a base.10 It has been found that TCA
has a very high ability to bind transition metal ions,11 which had
been quite unexpected from the poor binding ability of CA.
However, neither TCA or CA could form complexes with alkali
metal ions in solvent extraction. As a part of our efforts to
develop the potential of TCA, herein we report the stereo-
selective etherification of TCA to give the conformational
isomers of 5,11,17,23-tetra-tert-butyl-25,26,27,28-tetrakis[(eth-
oxycarbonyl)methoxy]-2,8,14,20-tetrathiacalix[4]arene (1). Also
reported are the results of selective alkali metal ion extraction
with the conformational isomers of 1.
X
X
X
X
Compd.
X
R
OR
RO
OR
OR
CA
TCA
1
CH2
S
H
H
S
CH2COOEt
CH2COOEt
2
CH2
† Supplementary data (SUPPL. NO. 57448, pp. 6) is available from the
British Library. For details of the Supplementary Publications Scheme,
see ‘Instructions for Authors’, J. Chem. Soc., Perkin Trans. 2, available
via the RSC Web page (http://www.rsc.org/authors). The supplemen-
tary data is also available on the RSC’s web server (http://www.rsc.org/
suppdata/perkin2/1998/2745/).
Results and discussion
Conformation selective etherification of TCA
It is known that TCA exists in a cone conformation but the
energy barrier of the ring inversion is somewhat lower than that
J. Chem. Soc., Perkin Trans. 2, 1998, 2745–2750
2745

View Article Online
Fig. 1 Schematic representation of four possible conformers of 1.
Table 1 Conformer distribution for the reaction of TCA with ethyl bromoacetate
Conformer distribution (%)
Run
Base
Solvent
Temp/ЊC
Time/h
Yield (%)
Cone
Partial cone
1,3-Alternate
1 a
2
3
Li2CO3
Na2CO3
K2CO3
Cs2CO3
Li2CO3
Na2CO3
K2CO3
Cs2CO3
DMF
DMF
DMF
DMF
Acetone
Acetone
Acetone
Acetone
60
60
60
60
Reflux
Reflux
Reflux
Reflux
34
16
6
0
81
95
85
0
85
95
96
0
26
13
6
0
65
63
34
0
9
61
16
0
9
24
60
0
0
15
81
4
2
5 a
6
46
36
28
4
0
91
24
3
7
8
a A mixture of di- and tri-O-substituted TCA was obtained (ca. 20% yield each).
of CA due to the weaker intramolecular hydrogen bonding
with the ring size slightly enlarged by replacing the methylene
linkage with a sulfide bond.10 As molecular models suggest,
however, the (ethoxycarbonyl)methoxy group seems large
enough to prevent the so-called oxygen through the annulus
rotation12 even in the case of TCA. Therefore, complete O-
alkylation of the OH groups of TCA may produce all the four
possible isomers at most, each of which should be conform-
ationally stable in the cone, 1,2-alternate, partial cone, and 1,3-
alternate conformations (Fig. 1). Thus, TCA was treated with
ethyl bromoacetate in the presence of an alkali carbonate in
dimethylformamide (DMF) or acetone according to Shinkai’s
procedure6 used for the etherification of CA (Table 1). It was
found that direct alkylation of TCA affords three conform-
ational isomers out of the possible four, i.e. cone-, partial-cone-,
Na2CO3 and K2CO3 gave cone-2 selectivity in either DMF or
acetone but Cs2CO3 gave partial-cone-2.6 They needed the
protect–deprotection technique, using a benzyl group, to obtain
1,3-alternate-2. By contrast, TCA could conveniently be deriv-
atized directly into 1,3-alternate-1 in the presence of Cs2CO3.
This might be attributed to the intermediate being large enough
to accommodate the Csϩ ion since the TCA skeleton is so much
larger than the CA skeleton as discussed below.
Structure analysis by 1H NMR and X-ray
The conformations of usual calix[4]arene derivatives can con-
veniently be assigned by use of the 1H NMR resonance pattern
of the methylene groups linking the phenol nuclei.13 However,
this tactic cannot be applied to 1 because of the bridging
epithio groups. So we relied on the chemical shifts and reson-
ance patterns of the other protons. Theoretically, only partial-
cone-1 among the possible four conformers has three non-
equivalent p-tert-butylphenol units, which will give a 2:1:1
resonance pattern in the NMR spectrum. Therefore, the isomer
which exhibited But peaks at 1.05 (18H, s), 1.31 (9H, s), and
1.43 (9H, s) was safely assigned as partial cone. On the other
hand, in cone- and 1,3-alternate-1, all the phenyl units are
equivalent so that both of the conformers should give the same
resonance pattern for the But protons (36H, s), -OCH2CO- (8H,
s), and Ar-H (8H, s) (see the Experimental section). However, it
was possible to distinguish between these two isomers by con-
sidering their chemical shifts. In 1,3-alternate-1, the But and
aromatic protons are in the deshielding zone of not only the
attaching phenyl ring but also the two adjacent phenyl units,
while those of cone-1 are in the shielding zone of the adjacent
phenyl nuclei. Therefore, it is expected that the But and
aromatic protons of the 1,3-alternate should give resonances at
lower field than those of the cone conformer. Furthermore,
the -OCH2CO- moiety of 1,3-alternate-1 is shielded by the
adjacent phenyl rings so that its methylene protons should res-
onate at higher field than those of the cone conformer. Thus,
the isomer which exhibited But and CH2 protons at δ 1.09 and
5.18 ppm, respectively, was assigned to cone-1, while that which
exhibited protons at δ 1.25 and 4.60 ppm was assigned to 1,3-
alternate-1.
1
and 1,3-alternate-1, as unambiguously assigned by H NMR
and X-ray structure analysis (vide infra). Available evidence,
including detailed TLC and 1H NMR scrutiny, did not indicate
the formation of 1,2-alternate-1.
Table 1 lists the combined yields of isolated 1 and the isomer
distributions. It is interesting to note that the alkali metal
carbonates showed a distinct influence on the distribution of
conformers, suggesting that the template effect may be the main
controlling factor as in the case of tetra-O-alkylation of
CA to 5,11,17,23-tetra-tert-butyl-25,26,27,28-tetrakis[(ethoxy-
carbonyl)methoxy]calix[4]arene (2) performed by Shinkai et al.6
In DMF, base catalysts Na2CO3 and K2CO3 gave the partial
cone quite selectively whereas Cs2CO3 gave the 1,3-alternate.
The selectivity was rather conspicuous in acetone; Na2CO3,
K2CO3, and Cs2CO3 produced the cone, partial cone, and
1,3-alternate in the isolated yields of 77, 58, and 78%, respect-
ively. One possible explanation for the enhanced template
effect in acetone might be that the precursor phenolate inter-
mediates coordinate more tightly to the template metal ions
(Fig. 2). On the other hand, Li2CO3 gave no 1, giving a mix-
ture of di- and tri-O-alkylated TCA only in a poor yield
(ca. 20% each) as evidenced by TLC and 1H NMR. The distri-
bution of the conformers in Table 1 suggests that the size of
the cavity consisting of the phenyl and ester groups in each
conformer is in the order of cone < partial cone < 1,3-alternate
which may explain the results of a solvent extraction study (vide
infra).
1
The H NMR assignments were eventually confirmed by
Shinkai’s results of the direct tetra-etherification show that
X-ray structure analysis of the conformation of cone-1. The
2746
J. Chem. Soc., Perkin Trans. 2, 1998, 2745–2750

View Article Online
Fig. 2 Possible tri-O-alkylated intermediates coordinating to the template metal ions to yield cone-, partial-cone-, and 1,3-alternate-1. Small
ellipses denote 4-tert-butylphenyl moiety.
Table 2 The mean bond lengths of interest for cone-1 and -2a
Bond
Cone-1 (X = S)
Cone-2 (X = CH2)b
Ar(C–C)
Ar–X
1.39(1)
1.79(1)
1.52(1)
1.48(1)
1.39(1)
1.42(1)
1.19(1)
1.29(1)
1.39(1)
1.55(1)
1.52(1)
1.49(1)
1.41(1)
1.46(1)
1.19(1)
1.31(1)
But(C–C)
CH2–CO
Ar–O
CH2–O
C᎐O
᎐
O–CO
a Bond lengths are in Å (1 Å = 0.1 nm). b See reference 15.
Fig. 3 X-Ray structure of cone-1. Cell constants; monoclinic cell with
a = 16.390 Å, b = 21.518 Å, c = 16.778 Å, β = 93.92Њ, V = 5903 Å3. For
Z = 4 and FW = 1065.42, the calculated density is 1.2 g cmϪ3. Space
group: P21/n. R = 0.052, Rw = 0.065.
ORTEP diagram (Fig. 3) of cone-1 clearly indicates a typical
flattened cone structure, which is frequently observed in tetra-
O-alkylated calix[4]arenes with cone conformations in the solid
state.14 The (ethoxycarbonyl)methoxy groups seemed to be
rather randomly arranged in a unit cell. Its terminal Me group
showed a large anisotropic displacement, which implies large
thermal motion and/or disorder. Table 2 summarizes the aver-
age bond lengths of interest for cone-1 as well as the reported
values for cone-2.15 Only Ar–X, the bond length between the
benzene ring and the bridging group, in cone-1 showed a 15%
larger value than Ar–X in cone-2. Consequently, it is expected
that 1 has a somewhat larger cavity than 2 so that it can
accommodate the larger metal ions. This may explain, at least
in part, why TCA can undergo direct O-alkylation to the 1,3-
alternate using Cs2CO3 as the template.
Fig. 4 The percent extraction (E%) by three conformers of 1. Organic
phase: [L (=1)]org,init = 2.5 × 10Ϫ3
M in CH2Cl2. Aqueous phase:
[MOH]aq,init = 0.1 M, [HPic]aq,init = 2.5 × 10Ϫ4 M.
means that the species exists in the organic phase. As shown, the
formed metal complexes MLϩ are extracted as an ion-pair with
the picrate anion into the organic phase. The concentration‡
of MLϩPicϪ in the organic phase is given by eqn. (2), where
[MLϩPicϪ]org = [Pic]aq,init Ϫ [Pic]aq
(2)
[Pic]aq,init and [Pic]aq are the initial and final concentrations of
the picrate ion in the aqueous phase. The percent extraction,
E%, was calculated by eqn. (3).
Solvent extraction study
In order to evaluate the ability of the three conformers of 1 to
recognize alkali metal ions, ion-pair extraction was carried out.
The extraction reaction is given by eqn. (1), where Mϩ, L, PicϪ,
E% = [MLϩPicϪ]org/[Pic]aq,init × 100%
(3)
Fig. 4 shows the E% of the conformers versus alkali metal
ions. Cone-1 showed the highest E% for the Naϩ ion compared
to other metal ions using CH2Cl2 as the organic phase. By
Mϩ ϩ Lorg ϩ PicϪ
MLϩPicϪ
(1)
org
MLϩ, and MLϩPicϪ denote the metal ion, ligand (here, 1 or 2),
picrate anion, metal complex, and ion-pair. Subscript org
Ϫ3
᎐
‡ Hereafter, 1 M᎐1 mol dm
.
᎐
J. Chem. Soc., Perkin Trans. 2, 1998, 2745–2750
2747

View Article Online
ation ability [i.e. eqn. (5)], which tempted us to study the
thermodynamic stability of the metal complexes of 1 and 2.
However, the poor solubility of the complexes in the usual
1
NMR solvents, in addition to the subtle changes in H NMR
signals upon complexation with metal ions, restricted the
measurements to those of Naϩ complexes with cone-1 and -2
in 50 (v/v)% CDCl3–CD3OD.
The stoichiometry of the cone-1 complex with Naϩ in the
mixed solvent was determined to be 1:1 by the continuous vari-
ation method, while that of cone-2 was also known to be 1:1.6
Therefore the reaction to be considered is the same as eqn. (5).
In this case, Mϩ is Naϩ, L is cone-1 or cone-2, and MLϩ is
cone-Na1ϩ or cone-Na2ϩ. The equilibrium constant, K, is
then defined by eqn. (8). In the case of L = cone-1, since the
[MLϩ]
[Mϩ][L]
(8)
K =
Fig. 5 The percent extraction (E%) by cone conformers of 1 and 2.
Organic phase: [L (=1 or 2)]org,init = 2.5 × 10Ϫ3 M in CH2Cl2 or in nitro-
benzene (NB). Aqueous phase: [MOH]aq,init = 0.1 M, [HPic]aq,init
= 2.5 × 10Ϫ4 M. Data for cone-2 in CH2Cl2 are cited from ref. 6.
Circle: cone-1, square: cone-2, closed symbol: in CH2Cl2, open symbol:
in nitrobenzene.
intermolecular exchange of L between free L and MLϩ in eqn.
(5) was fast on the H NMR timescale, the chemical shift
1
change of L upon titration by Naϩ ion was analysed as usual.
By contrast in the case of L = cone-2, two distinct signals for
free L and MLϩ were observed, suggesting the inertness of
cone-Na2ϩ. Then each peak area was used to evaluate the K (for
detail, see the Experimental section). As a result, the K values
for cone-Na1ϩ and cone-Na2ϩ at 25 ЊC were estimated to be log
(K/MϪ1) = 2.85 and 4.03, respectively.
contrast, 1,3-alternate-1 extracted larger metal ions such as Kϩ
and Rbϩ better than the Naϩ ion. Partial-cone-1 showed a small
E% for all metal ions, among which it showed a preference for
the Kϩ ion. Although the aquation–deaquation energy of these
metal ions must be taken into account, the metal ion selectivity
seems to indicate that the cavity size provided by the ligating
groups of the conformational isomers of 1 is in the sequence
cone < partial cone < 1,3-alternate. This sequence may be said
to correspond to the conformational selectivity observed in
O-alkylation by these metal ions.
Considering the results reported by Shinkai et al., con-
formers of 2 generally exhibited better E% for alkali metal ions
than those of 1 under the same extraction conditions.6 Among
these conformers of 1 and 2, however, cone-1 exhibited the
highest Naϩ ion selectivity in CH2Cl2, which may make it a
candidate for the ionophore in an ion-selective electrode.
It has been known that the extractability in ion-pair extrac-
tion strongly depends on the relative permittivity of the organic
phase.16 Thus, solvents with high relative permittivity should
enhance ion-pair migration from the aqueous to the organic
phase. Therefore, nitrobenzene was used as the solvent instead
of CH2Cl2 for alkali metal ion extraction by cone-1 and -2. In
fact, the E% values for cone-1 and -2 in nitrobenzene generally
exceeded those in dichloromethane as indicated by the broken
lines in Fig. 5. However, the extractability of cone-1 did not
equal that of cone-2 except in the case of Csϩ. The ion-pair
extraction given by eqn. (1) is composed of reactions (4)–(7).
Why is the stability of cone-Na1ϩ lower than that of cone-
Na2ϩ? One possible explanation may be the fluxional property
of the intrinsic framework of the thiacalix[4]arene in 1, as
deduced from the longitudinal relaxation time, T1, of the parent
TCA.9 Thus, the basic framework as well as the (ethoxycarb-
onyl)methoxy groups of cone-1 may show faster thermal motion
than cone-2, resulting in the looser fixation of the Naϩ ion by
cone-1. We believe this is not a serious drawback of 1, because
the absolute value of log K is not as important as the selectivity,
namely the difference in log K among metal ions, when one
intends to use 1 as an ionophore in ion selective sensors.
Conclusion
In this study, we have reported that the conformational stereo-
selectivity of the tetra-O-alkylation of TCA was readily con-
trolled by choosing a suitable alkali carbonate as base catalyst.
Although the ability of 1 for alkali metal extraction is lower
than that of 2, the higher ion selectivity of 1 over 2 is an attract-
ive feature of 1 as an ionophore in sensory systems2 such as ion-
selective electrodes and chemically modified field effect
transistors.
Experimental
General remarks
Lorg
Mϩ ϩ L
L
(4)
(5)
(6)
(7)
MLϩ
Mps were taken using a Yamato MP-21 apparatus and are
corrected. Microanalyses were carried out in the Institute of
Chemical Reaction Science, Tohoku University. Merck silica gel
60GF254 was used for TLC. Silica gel columns were prepared
by the use of Merck silica gel 60 (63–200 µm). 1H NMR
was measured on a Bruker DPX-400 spectrometer operated at
400 MHz. The δ values are given in ppm and J values in Hz.
UV-VIS and IR spectra were obtained with a Shimadzu
UV-160 and IR-460, respectively. RP-HPLC, consisting of a
Shimadzu LC-10AD pump unit, a SPD-6A UV-VIS detector,
and a LiChrosphere RP-18 column (125 mm × 5 mm, 5 µm,
MERCK), was used to determine [Pic]aq in the aqueous–
nitrobenzene two-phase extraction system.
MLϩ ϩ PicϪ
[MLϩPicϪ]
[MLϩPicϪ]
[MLϩPicϪ]org
Due to the similarity in the molecular size and shape between
cone-1 and -2, it is assumed that free energies for the partition
of the ligand [eqn. (4)], ion-pairing of the complex [eqn. (6)],
and the partition [eqn. (7)] are similar in the cases of L = cone-1
and cone-2. Consequently, it may be concluded that the smaller
extractability of cone-1 is due to its complexing ability [eqn.
(5)].
Stability constants of Na؉ complexes of cone-1 and cone-2
Materials
The results of the above solvent extraction suggested that the
lower E% of cone-1 as compared to that of cone-2 seemingly
originated from the difference in the metal–ligand complex-
Samples of 5,11,17,23-tetra-tert-butyl-2,8,14,20-tetrathiacalix-
[4]arene-25,26,27,28-tetrol (TCA) were generously supplied
2748
J. Chem. Soc., Perkin Trans. 2, 1998, 2745–2750

View Article Online
by Cosmo Research Institute or synthesised as described
before.10 The cone conformer of 5,11,17,23-tetra-tert-butyl-
25,26,27,28-tetrakis[(ethoxycarbonyl)methoxy]calix[4]arene (2)
was synthesised as described by Shinkai et al.6 Standard solu-
tions (0.1 M) of alkali metal ions (Liϩ, Naϩ, Kϩ, Rbϩ, and Csϩ)
were prepared by dissolving the hydroxides in distilled water.
The accurate concentrations were determined by the usual
acid–base titration method. Picric acid was stocked as 0.01 M
aqueous solution. All the reagents purchased were of guaran-
teed reagent grade and used as purchased unless otherwise
noted. Acetone and DMF were distilled from CaH2 before use.
V = 5903(2) Å3 (obtained from a least squares refinement using
the setting angles of 20 carefully centered reflections in the
range 27.88Њ < 2θ < 29.81Њ, graphite monochromatized Mo-Kα
radiation, λ = 0.71069 Å), space group P21/n (No. 14), Z = 4,
Dcalc = 1.199 g cmϪ3. Colourless, prismatic shape. Crystal
dimensions 0.40 × 0.30 × 0.20 mm, µ(Mo-Kα) = 2.17 cmϪ1.
Data collection and processing. Rigaku AFC7R diffracto-
meter, ω/2θ mode with ω scan width = (1.26 ϩ 0.30tanθ)Њ,
ω scan speed 16.0Њ minϪ1; 14 396 reflections measured, 13 921
unique, giving 3294 with I > 3σ(I). An empirical absorption
correction based on azimuthal scans of several reflections
(transmission factors 0.9334–1.0000).
General procedure for the preparation of 5,11,17,23-tetra-tert-
butyl-25,26,27,28-tetrakis[(ethoxycarbonyl)methoxy]-2,8,14,20-
tetrathiacalix[4]arene (1)
Structure analysis and refinement. The structure was solved
by heavy-atom Patterson methods (PATTY)17 and expanded
using Fourier techniques (DIRDIF94).18 The non-hydrogen
atoms were refined anisotropically. Hydrogen atoms were
included but not refined. The final cycle of full-matrix
least-squares refinement was implemented by minimizing
Σw(|Fo| Ϫ |Fc|)2 for 3294 observed reflections [I > 3.00σ(I)] and
650 variable parameters and converged (largest parameter shift
was 0.09 times its esd) with unweighted and weighted agree-
ment factors of R = Σ |Fo| Ϫ |Fc| /Σ|Fo| = 0.052 and Rw =
{Σw(|Fo| Ϫ |Fc|)2/ΣwFc2}0.5 = 0.065, where w = 1/σ2(Fo) = {σc2(Fo)
TCA was suspended in dry acetone or DMF (ca. 50 mM solu-
tion) containing a 6-fold excess of an anhydrous alkali carbon-
ate and an 8-fold excess of ethyl bromoacetate. The mixture was
heated under nitrogen for 4–46 h (see Table 1). The reaction
was monitored by TLC. After cooling, the solid residue was
removed by filtration. The solution was evaporated to dryness
and maintained at 1 mmHg and 60 ЊC for several hours to
ensure the removal of unchanged ethyl bromoacetate. The
mixture of conformational isomers was separated by column
chromatography using silica gel. The eluents used were n-
hexane–benzene and n-hexane–ethyl acetate mixed solvents.
The cone-1 and partial-cone-1 were recrystallized from
ethanolic solution whereas 1,3-alternate-1 was recrystallized
from benzene solution. The yields for different base/solvent
systems are listed in Table 1.
2
ϩ p2Fo /4}Ϫ1, σc2(Fo) = esd based on counting statistics, and
p = p-factor. The standard deviation of an observation of unit
weight given by {Σw(|Fo| Ϫ |Fc|)2/(No Ϫ Nv)}0.5 was 1.14, where
No and Nv are the number of observations and variables,
respectively. The weighting scheme was based on counting
statistics and included a factor (p = 0.075) to downweight the
intense reflections. Plots of Σw(|Fo| Ϫ |Fc|)2 versus |Fo|, reflection
order in data collection, sinθ/λ and various classes of indices
showed no unusual trends. The maximum and minimum peaks
on the final difference Fourier map corresponded to 0.25 and
Ϫ0.21 eϪ ÅϪ3, respectively. Neutral atom scattering factors were
taken from Cromer and Waber.19 Anomalous dispersion effects
Cone-1. Mp 180.0–180.6 ЊC (Found: C, 63.18; H, 6.62; S,
12.22. C56H72O12S4 requires C, 63.13; H, 6.81; S, 12.04%); νmax
/
cmϪ1 2965 (CH), 1757 and 1732 (CO); δ (400 MHz, CDCl3) 1.09
(36H, s, But), 1.28 (12H, t, J 7.2, CH3), 4.21 (8H, q, J 7.2,
COOCH2), 5.18 (8H, s, OCH2CO), 7.29 (8H, s, ArH).
were included in Fcalc;
20 the values for δf Ј and δf Љ were those of
Partial-cone-1. Mp 212.0–212.8 ЊC (Found: C, 63.00; H, 6.66;
S, 12.21. C56H72O12S4 requires C, 63.13; H, 6.81; S, 12.04%);
νmax/cmϪ1 2960 (CH), 1761 and 1735 (CO); δ (400 MHz, CDCl3)
1.05 (18H, s, But), 1.31 (9H, s, But), 1.43 (9H, s, But), 1.13 (3H,
t, J 7.2, CH3), 1.27 (6H, t, J 7.2, CH3), 1.32 (3H, t, J 7.2, CH3),
3.98 (2H, q, J 7.2, COOCH2), 4.14–4.29 (6H, m, COOCH2),
4.65 (2H, s, OCH2CO), 4.71 (2H, d, J 15.2, OCH2CO), 4.74
(2H, s, OCH2CO), 4.78 (2H, d, J 15.2, OCH2CO), 7.01 (2H, d,
J 2.5, ArH), 7.52 (2H, d, J 2.5, ArH), 7.54 (2H, s, ArH), 7.86
(2H, s, ArH).
Creagh and McAuley.21 The values for the mass attenuation
coefficients are those of Creagh and Hubbell.22 All calculations
were performed using the teXsan23 crystallographic software
package of the Molecular Structure Corporation.
Solvent extraction
A typical procedure for the extraction study is as follows. To a
30 cm3 vial tube were pipetted 5 cm3 of 0.1 M alkali metal
hydroxide solution, 0.125 cm3 of 0.01 M picric acid and 5 cm3
of 1 or 2 solution (2.5 × 10Ϫ3 M) in CH2Cl2 or nitrobenzene,
and the mixture was then shaken at 300 strokes per min for 30
min. The concentration of picrate ion remaining in the aqueous
phase, [Pic]aq, was determined by measuring the absorbance at
360 nm or HPLC analysis [eluent: 40 (v/v)% CH3OH in water
containing 0.01 M tetrabutylammonium bromide, flow rate: 1.0
cm3 minϪ1, detection wavelength: 254 nm].
1,3-Alternate-1. Mp 329.5–331.0 ЊC (decomp.) (Found: C,
63.12; H, 6.72; S, 12.11. C56H72O12S4 requires C, 63.13; H, 6.81;
S, 12.04%); νmax/cmϪ1 2960 (CH), 1764 and 1736 (CO); δ (400
MHz, CDCl3) 1.25 (36H, s, But), 1.28 (12H, t, J 7.2, CH3), 4.22
(8H, q, J 7.2, COOCH2), 4.60 (8H, s, OCH2CO), 7.51 (8H, s,
ArH).
X-Ray structure analysis
Determination of stoichiometry of cone-Na1؉ complex
Singlecrystalsofcone-1§werepreparedbyrecrystallizationfrom
The stoichiometry was determined by the continuous variation
method.24 The ligand cone-1 and NaClO4 were first dissolved in
50 (v/v)% CDCl3–CD3OD solution separately at appropriate
concentrations, which were then mixed and made up with the
solvent to control the concentrations. The sum of the total
concentration of the ligand {[L]T, eqn. (9)} and the metal ion
{[M]T, eqn. (10)}, namely [L]T ϩ [M]T, was maintained to be 4.0
ethanol–CH2Cl2 solution and sealed in capillaries.
Crystal data. C56H72O12S4, M = 1065.42. Monoclinic,
a = 16.390(5) Å, b = 21.518(4) Å, c = 16.778(6) Å, β = 93.92(3)Њ,
§ Full crystallographic details, excluding structure factor tables, have
been deposited at the Cambridge Crystallographic Data Centre
(CCDC). For details of the deposition scheme, see ‘Instructions for
Authors’, J. Chem. Soc., Perkin Trans. 2, available via the RSC Web
page (http://www.rsc.org/authors). Any request to the CCDC for this
material should quote the full literature citation and the reference
number 188/143.
[L]T = [L] ϩ [MLϩ]
(9)
[M]T = [Mϩ] ϩ [MLϩ]
(10)
J. Chem. Soc., Perkin Trans. 2, 1998, 2745–2750
2749

View Article Online
F. Arnaud-Neu, V. Böhmer, J.-F. Dozol, C. Grüttner, R. A. Jakobi,
mM in 50 (v/v)% CDCl3–CD3OD. At the same time, the ratio
[L]T/[M]T was varied from 3:1 to 2:1, 1:1, 1:2, and 1:3. After
measuring the chemical shift, δ, of the -OCH2CO- protons, it
was then converted to the chemical shift change, ∆ by eqn. (11),
D. Kraft, O. Mauprivez, H. Rouquette, M.-J. Schwing-Weill,
N. Simon and W. Vogt, J. Chem. Soc., Perkin Trans. 2, 1997, 575.
5 See e.g., F. Arnaud-Neu, E. M. Collins, M. Deasy, G. Ferguson, S. J.
Harris, B. Kaitner, A. J. Lough, M. A. McKervey, E. Marques, B. L.
Ruhl, M. J. Schwing-Weill and E. M. Seward, J. Am. Chem. Soc.,
1989, 111, 8681; A. Arduini, A. Casnati, L. Dodi, A. Pochini
and R. Ungaro, J. Chem. Soc., Chem. Commun., 1990, 1597;
T. Nagasaki and S. Shinkai, Bull. Chem. Soc. Jpn., 1992, 65, 471;
M. Ogata, K. Fujimoto and S. Shinkai, J. Am. Chem. Soc., 1994,
116, 4505; J. M. Harrowfield, M. Mocerino, B. J. Peachey, B. W.
Skelton and A. H. White, J. Chem. Soc., Dalton Trans., 1996, 1687;
F. Arnaud-Neu, G. Barrett, D. Corry, S. Cremin, G. Ferguson, J. F.
Gallagher, S. J. Harris, M. A. KcKervey and M.-J. Schwing-Weill,
J. Chem. Soc., Perkin Trans. 2, 1997, 575.
6 K. Iwamoto and S. Shinkai, J. Org. Chem., 1992, 57, 7066.
7 See e.g., L. C. Groenen, B. H. M. Ruël, A. Casnati, P. Timmerman,
W. Verboom, S. Harkema, A. Pochini, R. Ungaro and D. N.
Reinhoudt, Tetrahedron Lett., 1991, 32, 2675; W. Verboom,
S. Datta, Z. Asfari, S. Harkema and D. N. Reinhoudt, J. Org. Chem.,
1992, 57, 5394.
8 T. Sone, Y. Ohba, K. Moriya and H. Kumada, Book of Abstracts
for Workshop on Calixarenes and Related Compounds, p. PS/B-36,
2–4 June, 1993, Fukuoka, Japan.
9 T. Sone, Y. Ohba, K. Moriya, H. Kumada and K. Ito, Tetrahedron,
1997, 53, 10 689.
10 H. Kumagai, M. Hasegawa, S. Miyanari, Y. Sugawa, Y. Sato,
T. Hori, S. Ueda, H. Kamiyama and S. Miyano, Tetrahedron Lett.,
1997, 38, 3971.
11 N. Iki, N. Morohashi, F. Narumi and S. Miyano, Bull. Chem. Soc.
Jpn., 1998, 71, 1597.
∆ = δL Ϫ δ
(11)
where δL is the chemical shift value when the Naϩ ion is absent.
Finally ∆LT was plotted against [L]T/([L]T ϩ [M]T). The max-
imum was observed at [L]T/([L]T ϩ [M]T) = 0.5 which implied
the stoichiometry is 1:1.¶
Determination of stability constants of cone-Na1ϩ and cone-
Na2؉ complexes
The solutions of appropriate concentration were prepared as
above. For the cone-Na1ϩ complex, [L]T was maintained as 1.0
mM for all conditions, whereas [M]T was varied to be 0, 0.5, 1.0,
2.0, 3.0, 5.0, 8.0, and 12.0 mM. The chemical shift change,
∆ [eqn. (11)], for the -OCH2CO- protons was recorded. Finally
a set of [M]T versus ∆ data was applied to Lang’s routine25
to obtain K and ∆c, which is the ∆ value at [M]T→∞.||
For the cone-Na2ϩ complex, the sample of [L]T = 1.0 mM
and [M]T = 0.5 mM was prepared. Since the cone-Na2ϩ
1
complex and the free cone-2 gave discrete H NMR signals,
[MLϩ]/[L] was directly calculated (to be 0.76) from the peak
area of But protons of each species. Then [L] and [MLϩ] were
calculated from eqns. (12) and (9). Finally [Mϩ] was obtained
12 K. Araki, K. Iwamoto, S. Shinkai and T. Matsuda, Chem. Lett.,
1989, 1747.
13 C. D. Gutsche, B. Dhawan, J. A. Levine, K. H. No and L. J. Bauer,
Tetrahedron, 1983, 39, 409.
14 A. Arduini, W. M. McGregor, D. Paganuzzi, A. Pochini, A. Secchi,
F. Ugozzoli and R. Ungaro, J. Chem. Soc., Perkin Trans 2, 1996,
839.
[L]T
[L] =
[MLϩ]
(12)
1 ϩ
[L]
15 F. Arnaud-Neu, E. M. Collins, M. Deasy, G. Ferguson, S. J. Harris,
B. Kaitner, A. J. Lough, M. A. McKervey, E. Marques, B. L. Ruhl,
M. J. Schwing-Weill and E. M. Seward, J. Am. Chem. Soc., 1989,
111, 8681.
from eqn. (10) and was introduced into eqn. (8) to obtain K
(=1.08 × 104 MϪ1). All measurements were done at 25 ЊC.
16 See e.g., Y. Takeda, H. Sato and S. Sato, J. Solution Chem., 1992, 21,
1069; M. Billah, T. Honjo and K. Terada, Fresenius’ J. Anal. Chem.,
1993, 347, 107; Y. Yamamoto, Bunseki Kagaku, 1972, 21, 418.
17 P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman,
S. Garcia-Granada, R. O. Gould, J. M. M. Smits and C. Smykalla,
The DIRDIF program system, Technical Report of the
Crystallographic Laboratory, University of Nijmegen, The
Netherlands, 1992.
18 P. T. Beurskens, G. Admiraal, G. Beurskens, W. P. Bosman, R. de
Gelder, R. Israel and J. M. M. Smits, The DIRDIF-94 program
system, Technical Report of the Crystallography Laboratory,
University of Nijmegen, The Netherlands, 1994.
Acknowledgements
The authors wish to thank Dr Hirotsugu Takizawa and Dr
Kyota Ueda for their kind assistance in analyzing X-ray struc-
tures. We are grateful to the Ministry of Education, Science,
Sports and Culture, Japan (Grant-in-Aids No. 07555587, No.
08405058 and No. 09750940) and to Cosmo Research Institute
for financial support. We also thank Cosmo Research Institute
for a large sample of TCA.
¶ For raw data and plotting, refer to supplementary data.
|| For detailed procedure, see reference 25. For the raw data and the
Lang’s plot, refer to the supplementary material.
19 D. T. Cromer and J. T. Waber, International Tables for X-ray
Crystallography, vol. 4, The Kynoch Press, Birmingham, England,
Table 2.2 A, 1974.
20 J. A. Ibers and W. C. Hamilton, Acta Crystallogr., 1964, 17, 781.
21 D. C. Creagh and W. J. McAuley, International Tables for
Crystallography, vol. C, ed. A. J. C. Wilson, Kluwer Academic
Publishers, Boston, Table 4.2.6.8, pp. 219–222, 1992.
22 D. C. Creagh and J. H. Hubbell, International Tables for
Crystallography, vol. C, ed. A. J. C. Wilson, Kluwer Academic
Publishers, Boston, Table 4.2.4.3, pp. 200–206, 1992.
23 Crystal Structure Analysis Package, Molecular Structure
Corporation, 1985 and 1992.
24 K. A. Connors, Binding Constants. The Measurement of Molecular
Complex Stability, John Wiley & Sons, New York, 1987.
25 R. P. Lang, J. Am. Chem. Soc., 1962, 84, 1185.
References
1 V. Böhmer, Angew. Chem., Int. Ed. Engl., 1995, 34, 713; M. A.
McKervey and V. Böhmer, Chem. Br., 1992, 724; C. D. Gutsche,
Calixarenes, monographs in supramolecular chemistry, ed. J. F.
Stoddard, The Royal Society of Chemistry, Cambridge, 1989.
2 D. Diamond and M. A. McKervey, Chem. Soc. Rev., 1996, 15.
3 K. Ohto, M. Yano, K. Inoue, T. Yamamoto, M. Goto, F. Nakashio,
S. Shinkai and T. Nagasaki, Anal. Sci., 1995, 11, 893; J. M.
Harrowfield, M. Mocerino, B. J. Peachey, B. W. Skelton and A. H.
White, J. Chem. Soc., Dalton Trans., 1996, 1687.
4 See e.g., C. D. Gutsche and K. C. Nam, J. Am. Chem. Soc., 1988,
110, 6153; Y. K. Agrawal and M. Sanyal, Analyst, 1995, 120, 2759;
Paper 8/03734E
2750
J. Chem. Soc., Perkin Trans. 2, 1998, 2745–2750