Synthesis and lithium ion selectivity of 14-crown-4 derivatives having bulky subunits: cis and trans isomers of 2-phenylcyclohexano-14-crown-4, 2,3-diphenylcyclohexano-14-crown-4 and 2,3-di-(1-adamantyl)-14-crown-4

Article abstract of DOI:10.1039/a706606f

cis- and trans-2-Phenylcyclohexano-14-crown-4, cis- and trans-2,3-diphenylcyclohexano-14-crown-4, and cis- and trans-2,3-di-(1-adamantyl)-14-crown-4 have been prepared and their ion selectivities toward alkali metal cations examined by means of the extraction of alkali metal picrates, the measurement of stability constants of the complexes with lithium or sodium perchlorate, and the electrode response potential measurement for the ion-sensitive membranes. The remarkable dependence of the complexation ability to Li⁺ on the cisltrans stereochemistry of the ionophores is discussed on the basis of their geometries estimated by molecular mechanics calculations and the structures of cis- and trans-diphenylcyclohexano derivatives determined by X-ray structure analyses. In order to assist the discussion, the X-ray structure analyses of the lithium picrate complex of decalino-14-crown-4 and the 2:1 'sandwich-type' complex of benzo-14-crown-4 with sodium perchlorate have also been undertaken.

Full text of DOI:10.1039/a706606f

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and lithium ion selectivity of 14-crown-4 derivatives having
bulky subunits: cis and trans isomers of 2-phenylcyclohexano-
14-crown-4, 2,3-diphenylcyclohexano-14-crown-4 and
2,3-di-(1-adamantyl)-14-crown-4
Yoshito Tobe,*,a Yuko Tsuchiya,a Hidekazu Iketani,a Koichiro Naemura,a
Kazuya Kobiro,*,b,d Mayumi Kaji,b Sachiko Tsuzukib and Koji Suzukic
a Department of Chemistry, Faculty of Engineering Science, Osaka University,
1-3 Machikaneyama, Toyonaka, Osaka 560, Japan
b Niihama National College of Technology, 7-1 Yagumo-cho, Niihama, Ehime 792, Japan
c Department of Applied Chemistry, Faculty of Science and Technology, Keio University,
3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223, Japan
d Present address: Inoue Photochirogenesis Project, ERATO, 4-6-3 Kami-shinden, Toyonaka,
Osaka 565, Japan
cis- and trans-2-Phenylcyclohexano-14-crown-4, cis- and trans-2,3-diphenylcyclohexano-14-crown-4, and
cis- and trans-2,3-di-(1-adamantyl)-14-crown-4 have been prepared and their ion selectivities toward alkali
metal cations examined by means of the extraction of alkali metal picrates, the measurement of stability
constants of the complexes with lithium or sodium perchlorate, and the electrode response potential
measurement for the ion-sensitive membranes. The remarkable dependence of the complexation ability
to Li؉ on the cis/trans stereochemistry of the ionophores is discussed on the basis of their geometries
estimated by molecular mechanics calculations and the structures of cis- and trans-diphenylcyclohexano
derivatives determined by X-ray structure analyses. In order to assist the discussion, the X-ray structure
analyses of the lithium picrate complex of decalino-14-crown-4 and the 2:1 ‘sandwich-type’ complex of
benzo-14-crown-4 with sodium perchlorate have also been undertaken.
R
Introduction
Much interest has been focused on lithium ionophores with
regard to their ion-selective analysis and extraction and their
O
O
O
O
O
O
O
biological applications.1 For example, lithium ion-selective elec-
trodes have potential applications to medical and clinical use in
the therapy of manic-depressive psychosis.2 The most import-
ant issue in the use of lithium ion ionophores in biological
system is their selectivity against the interfering sodium ion
which is most abundant. A number of Liϩ-selective ionophores
such as dioxadiamides,3 a phenanthroline derivative,4 and
crown-4 ethers5 have been synthesized, many of which exhib-
ited Liϩ/Naϩ selectivity up to a value of 102. A hybrid system
O
2a R = H
2b R = Me
1
picrates.8 Recently, an improved synthesis of a didecalino
derivative 2b and related compounds was reported by Sachle-
ben et al., who confirmed the excellent properties of the dide-
calino system as a Liϩ extractant.9 However, we found the iono-
phores having two bulky subunits were far less selective than
those with one bulky subunit as ion carriers for Liϩ-selective
electrodes.7c
Recently, we reported that the phenylcyclohexano unit and
the adamantyl group incorporated in homochiral 18-crown-6
ethers and related phenolic crown ethers exerted effective steric
hindrance as chiral barriers in complexations with chiral
amines or amino acids.10 It occurred to us that these bulky sub-
units may well serve as alternatives for the decalino group in
Liϩ-selective ionophores. Moreover, it would also be interesting
to investigate the steric effects on the Liϩ/Naϩ selectivity for the
pairs of cis/trans diastereomers. In this connection, we prepared
cis and trans isomers of 2-phenylcyclohexano-14-crown-4 (3
and 4), 2,3-diphenylcyclohexano-14-crown-4 (5 and 6), and 2,3-
di-(1-adamantyl)-14-crown-4 (7 and 8), and investigated their
selectivity and binding ability towards alkali metal cations.11
The Liϩ/Naϩ selectivities were determined by means of the
extraction of alkali metal picrates, the measurement of stability
composed of a 14-crown-4 ring and amide groups exhibited
pot
Liϩ/Naϩ selectivity in the order of 103 (log K = Ϫ3.25).6 How-
Li,j
ever, still higher selectivity is required for their practical use. In
this connection, it is important to elucidate fundamental fac-
tors which govern the Liϩ/Naϩ selectivity.
We reported previously that decalino-14-crown-4 1 and its
derivatives having a bulky subunit on one of the ethano bridges
of the macro ring exhibited excellent Liϩ/Naϩ selectivities (in
the order of 103).7 The basis of this molecular design is two-
fold: (1) The decalin unit would preorganize the crown ring to
bind Liϩ favourably. We believed, at first intuitively, that the
O᎐C᎐C᎐O torsion angle involving the decalin unit must be near
60Њ or even smaller, which is suitable for binding the smaller Liϩ
ion rather than Naϩ owing to the favourable orientation of the
lone-pair electrons of the donor oxygen atoms. (2) The bulky
subunit would prevent the formation of 2:1 (crown ether:
cation) complexes with the larger Naϩ ion because of the steric
hindrance between the two ionophores. We also reported that
crown ethers such as didecalino-14-crown-4 2a having two bulky
subunits, one at each side of the ethano bridge, exhibited even
better Liϩ/Naϩ selectivity in the extraction of alkali metal
J. Chem. Soc., Perkin Trans. 1, 1998
485

View Article Online
constants with lithium and sodium perchlorate, and the elec-
trode response potential measurement for the ion-sensitive
membranes. The difference between the complexation abilities
of the ionophores 1 and 3–8 is discussed on the basis of their
the reported procedures. cis-Diol 11 was prepared by the intra-
molecular pinacol coupling of 1,6-diphenylhexane-1,6-dione14
using the protocol of Corey et al.15 (Scheme 1). During the
Ph
O
OH
OH
Mg (Hg), TiCl4
Ph
Ph
O
O
O
O
O
O
O
O
O
Ph
11
H
H
Ad
OH
OR
Ad
Ad
O
Ad
Ad
OH
OR
(i) NaBH4 or
3
4
+
LiAlH4
(ii) H+
OR
Ad
R = H (16a)ref. 17
R = H (13)
15 : 1
6 : 4
R = H (14)
R = COMe (16b)ref. 17
R = SiButMe2 (16c)
R = COMe
R = COMe
O
O
O
O
O
O
O
O
R = H (13)
R = H (14)
1 : 23
Ad = 1-adamantyl
Scheme 1 Stereoselective preparation of cis-diol 11 and stereospecific
preparation of meso-diol 13 and ( )-diol 14
6
5
course of our work, a stereoselective preparation of meso-di-(1-
adamantyl)ethanediol 13 by the NaBH4 reduction of 1,2-di-(1-
adamantyl)-2-hydroxyethanone 16a16 was reported [meso-
13:( )-14 = 5:1];17 the observed high stereoselectivity was
attributed to chelation control. In fact, the NaBH4 reduction
of non-chelating acetate 16b was reported to afford a mixture
of erythro- and threo-hydroxyacetates in the ratio 1.5:1.17
In order to improve the stereoselectivity in the reduction
under non-chelation control, we thought that the hydroxy
group of acyloin 16a should be converted to a highly sterically
demanding group. Indeed, the LiAlH4 reduction of tert-
butyldimethylsilyl ether 16c followed by hydrolysis gave
( )-diol 14 with high stereoselectivity [meso-13:( )-14 = 1:23;
Scheme 1].
O
O
O
O
O
O
O
O
8
7
geometries estimated by molecular mechanics calculations, and
the structures of cis- and trans-diphenylcyclohexano derivatives
5 and 6 were determined by X-ray crystallographic analyses.
The condensation of diols 9–14 with ditosyl (15a) or dimesyl
(15b) derivatives was carried out under high-dilution conditions
in tetrahydrofuran (THF) using NaH as a base and LiClO4 as a
template. The isolated yields of the crown ethers 3–8 were 50,
35, 4, 3, 2, and 50%, respectively. It should be noted that the
yields of crowns trans-4, trans-6, and cis-7 were considerably
lower than those for the corresponding isomers, cis-3, cis-5,
and trans-8. The low yields of the trans isomers 4 and 6 are
ascribed to the low population of the conformation suitable for
cyclization. Namely, since it is not possible to form a 14-
membered ring from diaxial alkoxides derived from diols 10
and 12, the products must be formed from energetically less
favourable diequatorial alkoxides in which the phenyl group(s)
occupies an axial position. Similarly, the low efficiency in the
cyclization of meso-diol 13 is due to the anti orientation of
the hydroxy groups in its most stable conformer, in which the
two adamantyl groups occupy anti positions. The O᎐C᎐C᎐O
torsion angle of the most stable conformer of 13 was estimated
to be 168Њ by MM3* calculations.18
Results and discussion
Syntheses of ionophores 3–8
The crown ethers 3–8 were prepared by condensation of the
corresponding diols 9–14 with ditosyl derivative 15b5b or dimesyl
Ph
Ph
Ph
Ph
OH
OH
OH
OH
OH
OH
OH
OH
Ph
Ph
H
H
9
10
11
12
OH
OH
Ad
Ad
OH
Ad
Ad
O
Ad
O
O
OR
OH
OR
Ad
Alkali metal ion selectivities of ionophores 3–8
OR
In order to achieve insight into the selectivities of the
ionophores 3–8 toward alkali metal ions, we first carried out
extraction experiments of alkali metal picrates in a water/
CH2Cl2 system. As shown in Table 1, cis-diphenylcyclohexano
derivative 5 exhibited the largest extractability for Liϩ, which
was comparable to that of decalino-14-crown-4 1.7a Although
the extractabilities of cis-phenylcyclohexano and trans-
diadamantyl derivatives 3 and 8 were moderate, they showed
high Liϩ/Naϩ selectivity values which were as large as those of
16a R = H
16b R = COMe
16c R = SiButMe2
13
14
15a R = H
15b R = Ts
15c R = Ms
Ad = 1-adamantyl
derivative 15c of 4,7-dioxadecane-1,10-diol 15a. cis- and trans-
1-Phenylcyclohexane-1,2-diol 912 and 1013 and trans-1,2-
diphenylcyclohexane-1,2-diol 1210a were prepared according to
486
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Table 1 The extraction of alkali metal picrates in water/CH2Cl2
system by ionophores 1 and 3–8
Table 2 The stability constants for the lithium and sodium ion com-
plexes of ionophores 3–6a
Extractabilitya (%)
K(LiClO4)
K(NaClO4)
Liϩ/Naϩ
selectivity
Liϩ/Naϩ
selectivity
Ionophore
(molϪ1 dm3)
(molϪ1 dm3)
Ionophore
Liϩ
Naϩ
Kϩ
Rbϩ
1
<1
<1
1.0
<1
d
<1
Csϩ
0
<1
<1
1.1
<1
d
<1
3
4
5
6
5100
400
7000
130
28
b
23
b
180
c
300
c
1b
3
81
42.8
3.1
74.1
1.6
<1
5
7.7
<1
4.8
<1
<1
1.2
1
1.4
16
7.7
c
15
c
c
32
4
<1
1.1
<1
5
7
6
a Determined by Li or 23Na NMR spectrometry in cyclohexanone at
80 ЊC. b Not determined because the association constants were too small
to be determined with reasonable accuracy. c Not determined.
7
d
8
38.9
<1
a [ionophore] = 7.00 × 10Ϫ4 mol dmϪ3, [picric acid] = 7.00 × 10Ϫ5 mol
dmϪ3, [alkali metal hydroxide] = 1.00 × 10Ϫ1 mol dmϪ3 at 25 ЊC. The
extractability (Ex) is defined as follows: Ex = [M(CE)Pic]org/[MPic]0,
where M, CE, Pic and [MPic]0 denote alkali metal, crown ether, picrate,
and the initial concentration of alkali metal picrate in the aqueous
phase, respectively. The initial concentration of alkali metal picrate in
the aqueous phase is equal to that of picric acid in the aqueous phase,
because a large excess of a metal hydroxide was used. The equation is
thus rewritten as follows: Ex = [M(CE)Pic]org/[HPic]0, where [HPic]0 is
the initial picric acid concentration in the aqueous phase. b Ref. 7a.
c The Liϩ/Naϩ selectivity was not determined because of the low
extractability of crowns 4 and 6 for Naϩ and of crown 7 for both Liϩ
and Naϩ. d Not determined.
examined by molecular mechanics calculations using the
MM3* force-field.18 The most stable conformers are shown in
Fig. 2 and the steric energies and O᎐C᎐C᎐O torsion angles
involving the bulky subunits are listed in Table 4. As shown in
Table 4, cis isomers 3 and 5 are considerably more stable than
the corresponding trans isomers 4 and 6, as is trans-8 compared
with cis-7. Fig. 2 shows that the geometries of the 14-membered
ring of crowns 1, 3, 4 and 6 look similar while that of cis-
diphenyl derivative 5 adopts a nearly symmetrical pseudo-chair
conformation. Closer scrutiny of the geometries (Table 4), how-
ever, reveals that the O᎐C᎐C᎐O torsion angles of trans isomers
4 and 6 are considerably larger (approximately 70Њ) than those
of cis isomers 1, 3 and 5 (59–66Њ).
The single-crystal X-ray analyses were undertaken for cis-
and trans-diphenylcyclohexano derivatives 5 and 6 (Fig. 3) and
the structures were compared with those estimated by MM3*
calculations. One of the oxygen atoms (O3) in the X-ray struc-
ture of cis-5 has a large thermal parameter, presumably due to a
disorder of this atom. In contrast to the calculated structure,
the 14-membered ring of cis-5 does not adopt a pseudo-chair
conformation. On the other hand, the structure of trans isomer
6 possesses a crystallographic C2 axis passing through the
midpoints of bonds C(7)᎐C(7*), C(1)᎐C(1*) and C(5)᎐(5*) and
the crown ring adopts a pseudo-chair conformation.20 The
observed O᎐C᎐C᎐O torsion angles (55.6Њ for 5, 76Њ for 6) are
qualitatively in agreement with those for the calculated struc-
tures.21 Thus, taking into account the conformational flexibility
of the 14-membered ring,22 it is reasonable to assume that the
calculated structures represent at least one of the lowest-energy
conformations.
compounds 1 and 5. On the other hand, trans isomers 4 and 6
exhibited only low extractability for Liϩ and nearly negligible
extractability for other metal cations. Moreover, cis-di-
adamantyl derivative 7 did not show any extractability even for
Liϩ under similar conditions.
Next, the binding constants for phenylcyclohexano deriv-
atives 3–6 with lithium and sodium perchlorate were deter-
mined by the titration method using 7Li and 23Na NMR spectra
in order to obtain more quantitative information on the ion
selectivities (Table 2). The NMR measurements were under-
taken in cyclohexanone at 80 ЊC in order to attain the rapid
equilibrium between free and complexed ions on the NMR
time-scale.19 The use of low boiling solvents such as THF,
acetone, and acetonitrile led to the broadening or splitting of
the signals owing to the slow exchange rate between the two
cationic species. The formation of 1:1 (ionophere:cation)
complexes under these conditions was checked by Job’s plot20
for all complexes for which the stability constants were deter-
mined. The binding constants for diadamantyl derivatives 7 and
8 were not measured, however, because of the limited solubility
of the ionophores in cyclohexanone. As can be seen from Table
2, cis isomers 3 and 5 showed high Liϩ/Naϩ selectivity as well as
high stability constants for Liϩ. On the other hand, the binding
constants of trans derivatives 4 and 6 to Liϩ were much smaller
than those of the corresponding cis isomers. The stability con-
stants for the Naϩ complexes of compounds 4 and 6 were too
small to be determined with reasonable accuracy.
With regard to the conformation of diadamantyl derivatives
7 and 8, the 14-membered ring of compound cis-7 is substan-
tially deformed into a conformation having a large O᎐C᎐C᎐O
torsion angle (92Њ). The corresponding trans-isomer 8, on the
other hand, adopts a nearly eclipsed conformation with a small
O᎐C᎐C᎐O dihedral angle (0Њ).
Relationship between structure and complexation ability and ion
selectivity of ionophores 1 and 3–8
Finally, the performance of the ionophores 3–6 and 8 as ion
carriers for ion-selective electrodes was investigated by the
measurement of the electrode response potential using poly-
(vinyl chloride) (PVC)-membrane electrodes containing these
ionophores. The selectivity coefficients are summarized in Fig.
In order to understand the remarkable difference between the
complexation abilities of the ionophores 1 and 3–8, we first
compared the O᎐C᎐C᎐O torsion angles of the subunit portion
of the ionophores with those of the metal-ion complexes.
Recently, Hay and Rustad devised the MM3Ј parameters for
aliphatic ether–alkali metal complexes based on ab initio mol-
ecular orbital and molecular mechanics (MM3) calculations.23
They reported that the best-fit O᎐C᎐C᎐O torsion angle for the
Liϩ-dimethoxyethane chelate was smaller (54Њ) than those
for the corresponding Naϩ (58Њ) and Kϩ (62Њ) chelates. The
observed torsion angles for the Liϩ complexes of syn-dicyclo-
hexano-14-crown-4 1722 and didecalino derivative 2b24a are
consistent with the calculated value (55Њ and 51Њ, respectively).
However, Sachleben reported that the torsion angles of the Liϩ
complexes of nonamethyl-14-crown-4 1824b and tetrakis(spiro-
cyclopentano) derivative 1924a are larger than the best-fit angle
(73Њ and 74Њ, respectively). Since the Liϩ-extractability of com-
pound 2b is twice as much as that of compound 18 and 103-
1 and Table 3. As shown in Fig. 1, cis-diphenyl derivative 5
pot
Li,Na
exhibited excellent selectivity for Liϩ (log K
= Ϫ3.1), in
good agreement with its extractability and binding constants.
This selectivity is comparable to that of decalino derivative 1
(Ϫ3.3).7b,c To our dismay, however, cis-phenylcyclohexano and
trans-diadamantyl derivatives 3 and 8, which showed moderate
selectivity in the metal-extraction and -binding constants,
exhibited substantially lower selectivities (Ϫ1.9 and Ϫ2.0,
respectively) than that of compound 5.
Structures of ionophores 1 and 3–8
In order to understand the difference between the complexation
abilities of the ionophores 1 and 3–8, their structures were
J. Chem. Soc., Perkin Trans. 1, 1998
487

View Article Online
pot
Li,j
Table 3 Ion selectivity factors (log K ) of the electrodes based on the ionophores 3–6 and 8
pot
Li,j
log K
ϩ
Ionophore
Liϩ
Naϩ
Ϫ1.9
Ϫ1.8
Ϫ3.1
Ϫ1.3
Ϫ2
Kϩ
Rbϩ
Ϫ2.7
Ϫ1.8
Ϫ3.8
Ϫ3.5
Ϫ1.8
Csϩ
NH4
Mg2ϩ
Ϫ5.1
Ϫ4.1
Ϫ5.5
Ϫ4.4
Ϫ3.8
Ca2ϩ
Ϫ4.7
Ϫ3.7
Ϫ5.4
Ϫ4.1
Ϫ3.7
Sr2ϩ
Ϫ4.7
Ϫ3.7
Ϫ5
Ba2ϩ
Ϫ4.7
Ϫ3.6
Ϫ5
3
4
5
6
8
0
0
0
0
0
Ϫ1.8
Ϫ1.8
Ϫ3.6
Ϫ3.4
Ϫ1.9
Ϫ3.3
Ϫ1.6
Ϫ3.7
Ϫ1.6
Ϫ1.7
Ϫ3.1
Ϫ1.8
Ϫ3.7
Ϫ1.9
Ϫ1.9
Ϫ4.2
Ϫ3.6
Ϫ4
Ϫ3.8
pot
Li,j
Fig. 1 Selectivity coefficients (log K j = interfering ion) of the electrodes based on ionophores 3–6 and 8. The membrane compositions were 3%
(by weight) ionophore, 67.3–67.4% membrane solvent BBPA [bis(1-butylphenyl) adipate], 20 mol% (relative to the ionophore) potassium tetrakis-(p-
chlorophenyl)borate (KTpClPB) and 28.8–28.9% poly(vinyl chloride) (PVC). The measurements were carried out at 25.0 0.5 ЊC.
times larger than that of spiro compound 19,9 we assume that
the small O᎐C᎐C᎐O torsion angle (54Њ) is one of the important
factors for strong binding to Liϩ. In order to confirm this
assumption, we undertook an X-ray structure analysis of the
lithium picrate complex of decalino-14-crown-4 1, which exhib-
ited high Liϩ-extractability.7a As shown in Fig. 4, the structure is
similar to that of the LiNCS complex of didecalino-14-crown-4
derivative 2b.24a The crown ring of the complex adopts a
pseudo-chair conformation and the Liϩ is located 0.70 Å above
the mean-plane of the four oxygen atoms [O(1)᎐O(2)᎐
O(3)᎐O(4)]. The O᎐C᎐C᎐O torsion angle of the decalin unit is
51Њ, supporting the above assumption. It is clearly seen from
Table 4 that the ionophores 1 and 5 with relatively small torsion
angles (62Њ and 60Њ, respectively) exhibited the best complex-
ation ability to Liϩ. Consequently, it is deduced that the pre-
organization of the O᎐C᎐C᎐O torsion angle to a best-fit angle
plays an important role in the complexation ability of iono-
phores to Liϩ.
pre-organized to bind Liϩ (the MM3* O᎐C᎐C᎐O torsion
angles: 66Њ and 0Њ, respectively; Table 4), possess moderate
binding ability to Liϩ. The flexibility of the crown rings was
estimated from the variable-temperature NMR (VT-NMR)
spectra. The VT-NMR spectra of the cis-phenylcyclohexano
and cis-diphenylcyclohexano derivatives 3 and 5 revealed that
a chair–chair interconversion of the cyclohexane subunits
took place rapidly on the NMR time-scale (conformers A
and B; Scheme 2a). For example, while 18 signals were
observed in the 13C NMR spectrum of compound 3 ([2H8]tolu-
ene) at 70 ЊC owing to the rapid equilibrium between the two
conformers, the signals split into two sets of peaks at Ϫ80 ЊC
with an approximate intensity ratio of 1:3. Similarly, the 13C
NMR signals of compound 5 split into two sets of peaks at
Ϫ80 ЊC with equal intensities. The barriers for the conform-
ational change were estimated by simulating the exchange rates
at various temperature based on a line-shape analysis using
DNMR2.25 The ∆G‡ (25 ЊC)-value for the rotation was calcu-
lated from an Eyring plot to be 53.5 5.0 and 45.6 4.2 kJ
molϪ1, for compounds 3 and 5, respectively. Consequently, cis
Next, we looked at the conformational flexibility of the
crown ring because the ionophores 3 and 8, which are not
488
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Table 4 The energies and O᎐C᎐C᎐O torsion angles estimated by the
molecular mechanics calculations (MM3*) for ionophores 1 and 3–8
Ionophore
Steric energya (kJ molϪ1
)
Torsion angleb (deg)
1
3
4
5
6
7
8
245.6
62
66
71 (72)
60 (59)c
69 (69)d
92 (90)
0 (1)
328.5 (0.0)
247.1 (8.6)
333.2 (0.0)
344.4 (11.2)
413.2 (21.9)
391.3 (0.0)
a The steric energies relative to the more stable conformer of the
diastereomeric pairs are given in parentheses. b The O᎐C᎐C᎐O torsion
angle involving the bulky subunits. In the case of compounds 4–8,
there are three to five conformers with geometries similar to those of
the minimum structures. The average O᎐C᎐C᎐O torsion angles for
these conformers, whose steric energies are within 4.2 kJ molϪ1 of the
minima, are given in parentheses (unweighted for their population).
c The observed (X-ray analysis) angle is 55.6Њ. d The observed (X-ray
analysis) angle is 76Њ.
Fig. 2 CHEM3D illustrations for the most stable conformers of
ionophores 1 and 3–8 derived by the conformation search by Macro-
Model using the MM3* parameters. Meshed circles are carbon atoms
and those with spots are oxygen atoms. Hydrogens are not shown, for
clarity.
isomers 3 and 5 can adopt a best-fit conformation without
substantial energy cost to bind Liϩ. On the other hand, such
a conformational change was not observed for trans isomers
4 and 6, because it is not possible to form a 14-membered
ring for the diaxial oxygen atoms (conformer D; Scheme 2b).
Moreover, contraction of the O᎐C᎐C᎐O torsion angle would
bring about considerable strain in the cyclohexane unit (con-
former C; Scheme 2b). Therefore, a substantial energy cost
would be paid by trans isomers 4 and 6 in adopting a best-fit
conformation to bind Liϩ.
Fig. 3 Molecular structures of cis-diphenylcyclohexano-14-crown-4
(5: top) and trans-diphenylcyclohexano-14-crown-4 (6: bottom) deter-
mined by X-ray crystallographic analyses. Hydrogens are not shown,
for clarity.
Finally, the steric hindrance of the bulky subunits, which
prevents formation of a 2:1 complex, leading to a high Liϩ/
Naϩ selectivity, should be considered. Regarding the formation
of 2:1 ‘sandwich-type’ complexes, such host:guest stoichio-
metry has frequently been observed in complexes with large
cations relative to the ring size of the crown ethers. For
example, it has been shown that dibenzo-18-crown-6 formed
2:1 and 3:2 complexes with large cations such as Rbϩ and
Csϩ.26 On the other hand, we have reported that didecalino-18-
crown-6 2a gave only 1:1 complexes with any cation tested,
owing to the steric hindrance of the decalin group.27 Similarly,
we found that only 1:1 crystalline complexes were isolated
from the reaction of decalino-14-crown-4 1 with LiClO4 or
NaClO4 regardless of the initial molar ratio of 1 and the salts.
Moreover, in solution, ionophores 3–6 formed 1:1 complexes
with LiClO4, and the formation of 1:1 complexes with
NaClO4 was also confirmed for hosts 3 and 5, as described
above. With regard to a planar oblate ionophore, benzo-14-
crown-4 20,28 we found that it formed only a 1:1 complex with
LiClO4 with a stability constant of 1100 molϪ1 dm3 in
cyclohexanone at 80 ЊC. On the other hand, compound 20
formed both 1:1 and 2:1 complexes with NaClO4 as judged
from Job’s plot. In accord with the stoichiometry in solution,
we isolated a 1:1 complex of compound 20 with LiNCS and a
cis-Diadamantyl derivative 7 exhibited, in the 1H NMR spec-
trum (400 MHz, [2H8]toluene, 30 ЊC), two singlets due to the
methine protons (Ad-CH-O) at δ 3.41 and 2.89. The signal did
not change even at 80 ЊC, and so the barrier for conformational
change should be higher than 70 kJ molϪ1 (interconversion
between conformers E and G; Scheme 2c). This is due to the
highly strained geometry of the transition state of the rotation,
in which the two adamantyl groups are eclipsed (conformation
F; Scheme 2c). Consequently, it must be highly unfavourable
for compound 7 to reduce the inherently large O᎐C᎐C᎐O
torsion angle in order to adopt a best-fit conformation for bind-
ing a metal cation. On the other hand, the most stable con-
former of trans-diadamantyl derivative 8 estimated from the
MM3* calculations adopts an eclipsed conformation (con-
former H; Scheme 2d). It is, therefore, not difficult for compound
8 to adopt a gauche conformation (conformation I; Scheme 2d)
which is suitable for cation binding. Indeed, a gauche conform-
ation (O᎐C᎐C᎐O torsion angle 64Њ) was found by the MM3*
calculations to be only 13.7 kJ molϪ1 higher than the most
stable conformer H. Consequently, we deduced that the steric
energy required to adopt a best-fit conformation for Liϩ bind-
ing is also an important factor in determining the binding
ability to Liϩ.
J. Chem. Soc., Perkin Trans. 1, 1998
489

View Article Online
R
Ph
Ph
O
O
R
O
O
(a)
B
A
Ph
R
Ph
R
O
Ph
O
O
O
(b)
O
R
O
C
D
Ad
O
Ad Ad
H H
Ad
Ad
O
Ad
H
(c)
O
O
H
O
O
H
H
E
F
G
Fig. 4 Molecular structure of lithium picrate complex of decalino-14-
crown-4 1 determined by X-ray crystallographic analysis. Hydrogens
are not shown, for clarity.
H
Ad H
Ad
Me
Ad
Ad
O
O
H
O
(d)
O
O
O
H
H
H
H
Me
Me
Me
Me
Ad H
H
Ad
O
O
O
O
O
O
O
O
H
I
Me
Me
Me
Me
Scheme 2 Schematic representation of the conformational behaviour
of ionophores 3–8. (a) 3 (R = H) and 5 (R = Ph), (b) 4 (R = H) and 6
(R = Ph), (c) 7 and (d) 8.
H
17
18
Me
O
O
O
O
O
O
O
O
20
19
2:1 complex with NaClO4.29 Fig. 5 shows the molecular struc-
ture of the 2:1 ‘sandwich-type’ complex determined by an
X-ray crystallographic analysis. The sodium atom is located
1.55 Å above the mean plane of the four oxygen atoms
[O(1)᎐O(2)᎐O(3)᎐O(4)]. The side view of the molecular struc-
ture indicates clearly that there is little steric interaction
between the benzene rings of the two ionophores. One can
easily imagine that substituting the benzene ring of com-
pound 20 with a bulky group, such as decalino or diphenyl-
cyclohexano, would bring about substantial steric repulsion
between the two ionophores.
In conclusion, we have found that cis-diphenylcyclohexano
derivative 5 exhibited the highest selectivity as well as complex-
ation ability toward Liϩ, which were as good as those of the
previously reported decalino-14-crown-4 1. While cis-phenyl-
cyclohexano and trans-diadamantyl derivatives 3 and 8 showed
moderate complexation ability toward Liϩ, trans-phenylcyclo-
hexano and trans-diphenylcyclohexano derivatives 4 and 6
showed small binding ability and cis-diadamantyl derivative 7
did not show any binding ability. The relatively high Liϩ-
selectivity of cis-phenylcyclohexano, cis-diphenylcyclohexano,
and trans-diadamantyl derivatives 3, 5 and 8 is ascribed to the
pre-organization of the O᎐C᎐C᎐O torsion angle, small steric
energy required to adopt a best-fit conformation for Liϩ bind-
Fig. 5 Molecular structure of sodium perchlorate (1:2) complex of
benzo-14-crown-4 20, determined by X-ray crystallographic analysis.
Hydrogens and perchlorate anion are not shown, for clarity.
ing, and the steric hindrance of the bulky subunits which pre-
vents the formation of a 2:1 complex.
Experimental
1H and 13C NMR spectra were recorded on a JEOL JNM-GX-
270 (1H, 270 MHz; 13C, 67.5 MHz) or an AL-400 spectrometer
(1H, 400 MHz) in CDCl3 at 30 ЊC. J-Values are given in Hz. IR
and mass spectra were taken with a Hitachi 260-10 and a JEOL
JMS-DX-303-HF spectrometer, respectively. GLC was con-
ducted with a Shimadzu GC-8A gas chromatograph equipped
with a 5% SE-52 column. Analytical high-performance liquid
chromatography (HPLC) was carried out with a Shimadzu
LC-10AS chromatograph equipped with a Inertsil ODS
column, and preparative HPLC separations were performed
with a JAI LC-908 chromatograph equipped with JAIGEL 1H
and 2H with CHCl3 eluent. Light petroleum refers to the
fraction with distillation range 40–65 ЊC.
490
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Materials
title crown 4 (338 mg, 3.8%) as an oil, νmax(neat)/cmϪ1 1120, 765
cis- and trans-1-Phenylcyclohexane-1,2-diol 912 and 10,13 trans-
1,2-diphenylcyclohexane-1,2-diol 12,10a 1,6-diphenylhexane-
1,6-dione,14 di-(1-adamantyl)-2-hydroxyethanone 16a,16 4,7-
dioxadecane-1,10-diol 15a,5b,c and its ditosyl derivative 15b,5b
were prepared according to the literature procedures.
and 705; δH 7.75 (2 H, d, J 7.2), 7.21–7.34 (3 H, m), 3.38–4.06
(11 H, m), 3.13 (1 H, ddd, J 4.2, 5.9 and 9.9), 2.48 (1 H, dd, J 9.7
and 13.2) and 1.3–1.9 (12 H, m); δC 139.4 (s), 130.2 (d), 127.3
(d), 127.0 (d), 86.7 (d), 81.4 (s), 71.2 (t), 70.0 (t), 67.6 (t), 67.1
(t), 66.4 (t), 59.4 (t), 37.4 (t), 30.9 (t), 30.6 (t), 29.7 (t), 25.4 (t)
and 23.9 (t); m/z (EI) 334 (Mϩ, 25%), 158 (72) and 91 (100)
[Found (EI): Mϩ, 334.2144. C20H30O4 requires M, 334.2126].
For elemental analysis, the LiClO4 complex of compound 4 was
prepared as described below for those of crowns 1 and 20, mp
208–210 ЊC (Found: C, 54.3; H, 6.8. C20H30ClLiO8 requires C,
54.5; H, 6.9%).
cis-2,3-Diphenylcyclohexano-14-crown-4 5. The reaction of
diol 11 (7.40 g, 27.6 mmol) with ditosyl compound 15b (13.8 g,
28.6 mmol) was carried out as described above using 60%
NaH (4.00 g, 100 mmol) and LiClO4 (5.60 g, 52.8 mmol) in
2000 cm3 of THF for 14 days. During the reaction, another
batch of 60% NaH (9.6 g, 240 mmol) and diester 15a (15.8 g,
32.5 mmol) was added. The reaction mixture was worked up as
above to give the title crown 5 (3.98 g, 35%) as a solid, mp 63–
65 ЊC; νmax(KBr)/cmϪ1 1120 and 700; δH 7.2–7.5 (10 H, m), 3.99
(2 H, dt, J 3.5 and 8.7), 3.7–3.9 (6 H, m), 3.44 (2 H, dt, J 3.2 and
8.4), 3.00 (2 H, ddd, J 4.2, 5.9 and 9.6) and 1.6–2.0 (12 H, m); δC
141.9 (s), 129.7 (d), 126.4 (d), 126.3 (d), 81.9 (s), 70.8 (t), 68.8
(t), 58.1 (t), 30.0 (t), 29.7 (t) and 22.2 (t); m/z (EI) 410 (Mϩ, 2%)
and 158 (100) (Found: C, 76.1; H, 8.4. C26H34O4 requires C,
76.05; H, 8.35%).
Preparation of ionophores
cis-1,2-Diphenylcyclohexane-1,2-diol 11. To an ice–salt-
cooled suspension of magnesium amalgam prepared from
magnesium powder (730 mg, 30.0 mg-atom) and mercury()
chloride (236 mg, 0.87 mmol) in THF (11 cm3) was added
titanium() chloride (2.33 g, 12.3 mmol) followed by the add-
ition of a solution of 1,6-diphenylhexane-1,6-dione (1.0 g, 3.8
mmol) in 25 cm3 of THF. The mixture was stirred at room
temp. for 28 h; then saturated aq. K2CO3 was added. The mix-
ture was filtered through a pad of Celite and the filtrate was
extracted with diethyl ether. The organic layer was washed with
saturated aq. NaCl and dried over MgSO4. Removal of the
solvent followed by chromatography on silica gel (hexane–
EtOAc, 9:1) gave title diol 11 (240 mg, 24%) as an oil which
solidified on storage, mp 80–82 ЊC; νmax(KBr)/cmϪ1 3350, 1055,
750 and 700; δH 7.0–7.1 (10 H, m), 2.94 (2 H, br s), 2.2–2.3 (2 H,
m) and 1.9–2.0 (6 H, m); δC 144.1 (s), 127.1 (d), 126.8 (d), 126.6
(d), 77.2 (s), 35.8 (t) and 21.9 (t); m/z (EI) 268 (Mϩ, 8%) and
105 (100) [Found (EI): Mϩ, 268.1463. C18H20O2 requires M,
268.1444].
4,7-Dioxadecane-1,10-diyl bismethanesulfonate 15c. To a
solution of diol 15a (10.0 g, 56.1 mmol) and triethylamine (55
cm3) in CH2Cl2 (200 cm3) was added a solution of methane-
sulfonyl chloride (37.2 g, 325 mmol) in the same solvent (150
cm3) at Ϫ20 ЊC. The mixture was stirred overnight at room
temp.; then water was added. The organic layer was separated
and washed successively with 10% aq. HCl, saturated aq.
NaHCO3, and saturated aq. NaCl and dried over MgSO4. After
removal of the solvent in vacuo, the residue was chromato-
graphed on silica gel (hexane–EtOAc, 1:1) to afford bis-
sulfonate 15c (18.4 g, 98%) as a pale yellow oil, νmax(neat)/cmϪ1
1350, 1170, 1110, 980 and 940; δH 4.35 (4 H, t, J 6.4), 3.59 (8 H,
s ϩ t, J 5.9), 3.01 (6 H, s) and 2.02 (4 H, quintet, J 6.2).
cis-2-Phenylcyclohexano-14-crown-4 3. To a gently refluxed
suspension of 60% NaH (4.20 g, 105 mmol) and LiClO4 (5.60 g,
52.8 mmol) in THF (1000 cm3) was added dropwise a solution
of diol 9 (5.00 g, 26.0 mmol) and dimesyl compound 15c (9.60
g, 28.6 mmol) in the same solvent (1000 cm3) under nitrogen
during 10.5 h. The mixture was heated for 22 h while another
batch of 60% NaH (1.00 g, 25 mmol) was added. After being
cooled, the mixture was treated with ice–water and most of the
THF was evaporated off. The residue was extracted with diethyl
ether and the extract was washed with saturated aq. NaCl and
then dried (MgSO4). The solvent was removed and the residue
was chromatographed on silica gel (hexane–EtOAc, 92:8) to
afford the title crown 3 (2.45 g, 28%) as a solid. The reaction of
diol 9 with ditosyl analogue 15b gave the crown 3 in 50%
yield under similar conditions, mp 88–90 ЊC (from hexane);
νmax(KBr)/cmϪ1 1120, 760 and 710; δH 7.42–7.46 (2 H, m), 7.30–
7.36 (2 H, m), 7.23 (1 H, tt, J 1.5 and 7.2), 3.94 (1 H, dd, J 2.5
and 8.1), 3.84–3.89 (1 H, m), 3.55–3.78 (10 H, m), 3.21 (1 H, dt,
J 15.2 and 5.2) and 1.2–2.1 (12 H, m); δC 142.7 (s), 127.7 (d),
127.5 (d), 126.6 (d), 79.8 (s), 79.7 (br d), 70.7 (br t), 70.5 (t), 67.4
(t), 66.8 (t), 63.8 (t), 59.0 (br t), 36.3 (br t), 30.6 (t), 3.05 (t), 26.3
(t), 22.8 (br t) and 22.1 (t); m/z (EI) 334 (Mϩ, 34%) and 158
(100) (Found: C, 71.75; H, 9.1. C20H30O4 requires C, 71.8; H,
9.0%).
trans-2,3-Diphenylcyclohexano-14-crown-4 6. The reaction of
diol 12 (7.00 g, 26.0 mmol) with ditosyl derivative 15b (13.8 g,
28.6 mmol) was carried out as described above using 60% NaH
(4.00 g, 100 mmol) and LiClO4 (5.60 g, 52.8 mmol) in 2000 cm3
of THF for 14 days. During the reaction, another batch of 60%
NaH (7.2 g, 180 mmol) and 15b (12.6 g, 25.9 mmol) was added.
The reaction mixture was worked up as above and subsequent
preparative HPLC purification gave the title crown 6 (328 mg,
3.0%) as a solid, mp 151–152 ЊC; νmax(KBr)/cmϪ1 1140, 1120,
1100 and 710; δH 7.93 (4 H, d, J 7.4), 7.35 (4 H, t, J 7.4), 7.26 (2
H, t, J 7.2), 3.52–3.76 (8 H, m), 3.35 (2 H, ddd, J 4.2, 7.9 and
9.6), 3.26 (2 H, ddd, J 4.8, 7.3 and 9.0), 2.28–2.34 (2 H, m), 1.9–
2.1 (4 H, m) and 1.4–1.7 (6 H, m); δC 141.6 (s), 130.8 (d), 127.1
(d), 126.1 (d), 85.6 (s), 70.0 (t), 67.6 (t), 60.8 (t), 35.0 (t), 29.8 (t)
and 23.3 (t); m/z (CI) 411 (Mϩ ϩ 1, 54%) and 161 (100) (Found:
C, 76.3; H, 8.4%).
meso-1,2-Di-(1-adamantyl)ethane-1,2-diol 13. To a solution
of acyloin 16a (5.00 g, 15.2 mmol) in 650 cm3 of ethanol was
added portionwise NaBH4 (1.73 g, 45.7 mmol). The mixture
was stirred at room temp. for 44 h; another portion of the
hydride (1.73 g) was added during this period. Then the mixture
was diluted with water and extracted with CHCl3. The extract
was washed with saturated aq. NaCl and dried over MgSO4.
The solvent was removed in vacuo to leave a solid (4.90 g, 98%).
1
The ratio of meso-13:( )-14 was determined by the H NMR
spectrum to be 1:19. In order to remove the trace ( )-isomer,
the product (5.49 g, 16.6 mmol) obtained as described above
was dissolved in 960 cm3 of CHCl3 and the solution was
treated with lead tetraacetate (736 mg, 1.66 mmol) at room
temp. for 1 h. The mixture was filtered and the solvent was
removed by evaporation in vacuo. Chromatography on silica
gel (CHCl3) gave meso-13 (3.83 g, 70% recovery) as a single
diastereomer (by 1H NMR spectroscopy), mp 260–262 ЊC
(from EtOH); νmax(KBr)/cmϪ1 3420 and 1000; δH 3.15 (2 H,
br s), 2.00 (6 H, br m) and 1.4–1.8 (26 H, m); δC 79.8 (d),
38.3 (t), 37.7 (t), 37.4 (s) and 28.6 (d); m/z (EI) 330 (Mϩ,
100%) and 165 (54) (Found: C, 79.8; H, 10.3. C22H34O
requires C, 79.95; H, 10.35%).
trans-2-Phenylcyclohexano-14-crown-4 4. The reaction of diol
10 (5.20 g, 27.0 mmol) with dimesyl compound 15c (9.60 g, 28.7
mmol) was carried out as described above using 60% NaH (4.20
g, 105 mmol) and LiClO4 (5.60 g, 52.8 mmol) in 2000 cm3 of
THF for 20 h. The reaction mixture was worked up as above
and the product was purified by preparative HPLC to give the
1,2-Di-(1-adamantyl)-2-(tert-butyldimethylsiloxy)ethanone
16c. To an ice-cooled suspension of oil-free potassium hydride
(140 mg, 3.3 mmol) and 18-crown-6 (8 mg, 0.03 mmol) in 4 cm3
of THF was added dropwise a solution of acyloin 16a (1.00 g,
3.05 mmol) in 20 cm3 of THF under nitrogen. The ice-bath was
J. Chem. Soc., Perkin Trans. 1, 1998
491

View Article Online
removed and a solution of tert-butyl(chloro)dimethylsilane
(0.50 g, 3.3 mmol) in 20 cm3 of THF was added. The mixture
was stirred at room temp. for 1.5 h; then water was added to
quench the reaction. The mixture was extracted with diethyl
ether and the extract was dried over MgSO4. The solvent was
removed in vacuo and the residue was chromatographed on sil-
ica gel (hexane–benzene, 2:1) to give siloxy product 16c (792 g,
60%) as a solid, mp 179–180 ЊC (from benzene–MeOH);
νmax(KBr)/cmϪ1 1710, 1260, 1250, 1090, 1085, 1010, 865, 845
and 780; δH 4.16 (1 H, s), 1.46–2.04 (30 H, m), 0.88 (9 H, s), 0.08
(3 H, s) and 0.01 (3 H, s); m/z (FAB) 443 (Mϩ ϩ 1, 17%) and
279 (100) (Found: C, 75.65; H, 10.4. C28H46O2Si requires
C, 75.95; H, 10.45%).
( )-1,2-Di-(1-adamantyl)ethane-1,2-diol 14. To an ice-cooled
suspension of LiAlH4 (1.42 g, 37.3 mmol) in 22 cm3 of THF
was added dropwise a solution of the above ketone 16c (1.42 g,
37.3 mmol) in 180 cm3 of THF under nitrogen. The ice-bath
was removed and the mixture was stirred at room temp. for
3 h; then water was added to quench the reaction followed by
addition of aq. 1  NaOH. The mixture was extracted with
diethyl ether and the extract was dried over MgSO4. The sol-
vent was removed in vacuo and the residue was washed with
hexane to leave title diol 14 as a solid (3.57 g, 87%), mp 283–
285 ЊC (from benzene, sealed tube); νmax(KBr)/cmϪ1 3400 and
1085; δH 3.20 (2 H, br s), 2.17 (2 H, br s), 1.99 (6 H, br m) and
1.5–1.8 (24 H, m); δC 74.0 (d), 38.1 (t), 37.2 (t), 36.8 (s) and
28.3 (d); m/z (EI) 330 (Mϩ, 100%) (Found: C, 80.2; H, 10.45.
C22H34O requires C, 79.95; H, 10.35%). The ratio of meso-
13:( )-14 in the crude product was determined by the 1H NMR
spectrum to be 23:1.
cis-2,3-Di-(1-adamantyl)-14-crown-4 7. The reaction of diol
13 (2.80 g, 8.48 mmol) was carried out as described above using
15b (a total of 57.1 g, 118 mmol), 60% NaH (11.0 g, 275 mmol),
and LiClO4 (1.61 g, 15.2 mmol) in 440 cm3 of THF. The product
7 (137 mg, 2%) was isolated by chromatography on silica gel
(EtOAc–hexane, 1:9) as a solid, mp 162–163 ЊC (from EtOAc);
νmax(KBr)/cmϪ1 1130, 1110, 1070 and 1030; δH 3.9–4.0 (2 H, m),
3.6–3.8 (4 H, m), 3.5–3.6 (4 H, m), 3.41 (1 H, dt, J 4.2 and 4.0),
3.31 (1 H, s), 3.01 (1 H, dt, J 4.2 and 4.0), 2.85 (1 H, s) and 1.6–
2.0 (34 H, m); δC 92.7 (d), 86.2 (d), 70.6 (t), 70.5 (t), 70.0 (t), 67.6
(t), 66.9 (t), 64.9 (t), 40.6 (t), 39.9 (t), 38.4 (s), 37.5 (t), 37.0 (t),
36.9 (s), 30.6 (t), 29.4 (t), 28.9 (d) and 28.7 (d); m/z (EI) 472
(Mϩ, 6%), 173 (93) and 135 (100) (Found: C, 75.9; H, 10.4.
C30H48O4 requires C, 76.25; H, 10.25%).
trans-2,3-Di-(1-adamantyl)-14-crown-4 8. The reaction of
diol 14 (2.50 g, 7.58 mmol) was carried out as described above
using ditosyl compound 15b (a total of 14.1 g, 28.9 mmol), 60%
NaH (1.76 g, 43.6 mmol), and LiClO4 (1.61 g, 15.2 mmol) in
1000 cm3 of THF for 7 days. The product 8 (1.80 g, 50%) was
isolated by chromatography on silica gel (EtOAc–hexane, 1:9)
as a solid, mp 218–219 ЊC (from EtOAc); νmax(KBr)/cmϪ1 1050,
1020 and 1005; δH 4.03 (2 H, ddd, J 2.9, 8.8 and 11.3), 3.87 (2 H,
ddd, J 2.9, 8.8 and 11.2), 3.7–3.8 (2 H, m), 3.5–3.6 (4 H, m),
3.10 (2 H, ddd, J 4.0, 4.0 and 8.0), 2.82 (2 H, s), 1.95 (6 H, br m)
and 1.5–1.8 (28 H, m); δC 84.1 (d), 71.2 (t), 68.6 (t), 68.3 (t), 39.1
(t), 37.3 (t), 37.2 (s), 31.4 (t) and 28.6 (d); m/z (EI) 472 (Mϩ, 6%)
and 173 (100) (Found: C, 76.0; H, 10.25%).
An analytical sample was obtained by preparative HPLC,
νmax(neat)/cmϪ1 1260, 1120, 1050 and 750; δH 6.94 (4 H, m), 4.15
(4 H, t, J 5.2), 3.79 (4 H, t, J 5.2), 3.68 (4 H, s) and 2.00 (4 H,
m); δC 149.8 (2 C, s), 122.1 (2 C, d), 117.2 (2 C, d), 70.7 (2 C, t),
67.3 (4 C, t) and 29.9 (2 C, t).
Formation of crystalline complexes of crown ethers 1 and 20 with
lithium perchlorate, lithium thiocyanate, and sodium perchlorate
The crystalline complexes were prepared by mixing equimolar
or two molar amounts of crown ether 1 or 20 (one or two mol
equiv. used) with lithium or sodium perchlorate in acetone or
methanol. Slow evaporation off of the solvent at room temp.
or vapour diffusion with diethyl ether afforded crystals of the
complexes in good yields (64–99%). Since attempts to prepare
crystalline products of compound 20 with LiClO4 were unsuc-
cessful, LiNCS was used instead to give a 1:1 complex in 72–
75% yield. The stoichiometry of the product was determined by
elemental analyses. Complex 1ؒLiClO4; mp 214–215 ЊC (Found:
C, 51.55; H, 7.85. C18H32ClLiO8 requires C, 51.6; H, 7.7%).
Complex 1ؒNaClO4; mp 198–200 ЊC (Found: C, 49.5; H, 7.55.
C18H32ClNaO8 requires C, 49.7; H, 7.4%). Complex 20ؒLiNCS;
mp 203–205 ЊC (Found: C, 56.6; H, 6.3; N, 4.25. C15H20LiNO4S
requires C, 56.8; H, 6.35; N, 4.4%). Complex (20)2ؒNaClO4; mp
184–185 ЊC (Found: C, 53.75; H, 6.45; Cl, 5.45. C28H40ClNaO12
requires C, 53.65; H, 6.45; Cl, 5.65%).
Electrode preparation and emf measurements
Ion-sensitive membranes of the PVC matrix type were prepared
according to the previous procedures.30 The polymeric mem-
brane composition was 3 wt% ionophore, 67.3–67.4 wt%
membrane solvent BBPA, 28.8–29.9 wt% PVC, and 20 mol%
(relative to the ionophore) potassium tetrakis-(p-chlorophenyl)-
borate (KTpClPB, Dojindo Laboratories, Kumamoto, Japan).
The membrane thickness was ~100 µm. A 6-mm diameter circle
was cut from a prepared membrane and placed on the tip of a
PVC ion-selective electrode body assembly (Liquid Electrode
Membrane Kit, DKK Co., Ltd., Tokyo, Japan). The prepared
electrodes were immersed in 0.1 mol dmϪ3 aq. LiCl for over
24 h for preconditioning before use. The external reference
electrode was a double-junction-type Ag–AgCl electrode (HS-
305DS, Toa Electronics, Ltd., Tokyo, Japan). The electrode
response potential (emf) measurements were performed
according to the reported procedure at 25 0.5 ЊC using the
electrochemical cell system,30 Ag;AgCl, 3 mol dmϪ3 KCl | 0.3
mol dmϪ3 NH4NO3 | test solution | membrane | 0.1 mol dmϪ3
LiCl, AgCl;Ag.
All test solutions were made from chloride salts without any
pot
pH-adjusting buffer reagent. The selectivity coefficient K
i,j
where i is the primary ion (Liϩ) and j is the interfering ion, were
calculated from the response potentials in an alkali metal or
alkaline earth metal chloride solution using the separate-
solution method (SSM; [j] = 0.1 mol dmϪ3) according to the
recommendations of IUPAC31 and JIS.32
X-Ray crystallographic structure analyses of crown ethers 5 and
6, lithium picrate complex of crown 1, and sodium perchlorate
complex of crown 20
Benzo-14-crown-4 20. A mixture of NaOH (4.19 g, 105
mmol), LiClO4 (12.1 g, 114 mmol), and 80 cm3 of dimethyl
sulfoxide (DMSO) was stirred at 60 ЊC for 30 min. To the mix-
ture was added a solution of o-dihydroxybenzene (10.1 g, 91.7
mmol) in 50 cm3 of DMSO and the mixture was stirred for 1 h
at 60 ЊC, then a solution of 1,10-dichloro-4,7-dioxadecane5c
(19.6 g, 91.2 mmol) in 10 cm3 of DMSO was added during 3 h.
The mixture was stirred for 20 h at 110 ЊC. After being cooled,
the mixture was diluted with water and extracted by CH2Cl2.
The organic phase was washed with brine and dried over
MgSO4. The solvent was evaporated off and the residue was
chromatographed on silica gel (light petroleum–diethyl ether,
98:2) to give the title crown 20 (9.11 g, 40%) as an oil.
The lithium picrate complex of crown 1 was prepared as
follows: A solution of crown ether 1 (30 mg, 0.096 mmol) in 1
cm3 of CH2Cl2 was shaken with a solution of picric acid (300
mg, 1.31 mmol) and LiOHؒH2O (500 mg, 11.9 mmol) in 0.5 cm3
of water. The organic phase was separated and the solvent was
evaporated off to leave a solid, which was dissolved in a mixture
of acetone and benzene. The solution was stored in an atmos-
phere of light petroleum (vapour-diffusion method) to give
yellow single crystals; mp 193–195 ЊC. The sodium perchlorate
complex of complex 20 was prepared as described above.
X-Ray diffraction data were collected at 23 ЊC on a Rigaku
AFC-7R or AFC-5R (for lithium picrate complex of crown
1) diffractometer with Mo-Kα radiation (λ = 0.710 69 Å). The
492
J. Chem. Soc., Perkin Trans. 1, 1998

View Article Online
Springer, New York, 1986, ch. 7; U. Olsher, R. M. Izatt, J. S.
structures were solved and refined by using the program pack-
age TEXSAN.33,
Crystal data for compound 5. C26H34O4, M = 410.55, mono-
Bradshaw and N. K. Dalley, Chem. Rev., 1991, 91, 137;
H. Sakamoto, Bunseki, 1992, 468 (Chem. Abstr., 117: 123502).
2 A. D. Amdisen, Handbook of Lithium Therapy, MTP Press,
Lancaster, 1986.
3 N. N. L. Kirsch, R. J. J. Funck, E. Pretsch and W. Simon, Helv.
Chim. Acta, 1977, 60, 2326; A. F. Zhukov, D. Erne, D. Ammann,
M. Güggi, E. Pretsch and W. Simon, Anal. Chim. Acta, 1981, 131,
117; E. Metzger, D. Ammann, D. Asper and W. Simon, Anal.
Chem., 1986, 58, 132.
4 H. Sugihara, T. Okada and K. Hiratani, Chem. Lett., 1987, 2391.
5 (a) U. Olsher, J. Am. Chem. Soc., 1982, 104, 4006; (b) S. Kitazawa,
K. Kimura, H. Yano and T. Shono, J. Am. Chem. Soc., 1984, 106,
6978; (c) B. P. Czech, D. A. Bebb, B. Son and R. A. Bartsh, J. Org.
Chem., 1984, 49, 4805; (d) R. A. Bartsh, B. P. Czech, S. I. Kang,
L. E. Stewart, W. Walkowiak, W. A. Charewicz, G. S. Heo and
B. Son, J. Am. Chem. Soc., 1985, 107, 4997; (e) K. Kimura, H. Yano,
S. Kitazawa and T. Shono, J. Chem. Soc., Perkin Trans. 2, 1986,
1945; ( f ) K. Kimura, O. Oishi, T. Murata and T. Shono, Anal.
Chem., 1987, 59, 2331; (g) A. S. Attiyat, G. D. Christian, R. Y. Xie,
X. Wen and R. A. Bartsch, Anal. Chem., 1988, 60, 2561.
6 R. Kataky, P. E. Nicholson, D. Parker and A. K. Covington,
Analyst (London), 1991, 116, 135.
7 (a) K. Kobiro, T. Hiro, T. Matsuoka, K. Kakiuchi, Y. Tobe and
Y. Odaira, Bull. Chem. Soc. Jpn., 1988, 61, 4164; (b) K. Kobiro,
Y. Tobe, K. Watanabe, H. Yamada and K. Suzuki, Anal. Lett.,
1993, 26, 49; (c) K. Suzuki, H. Yamada, K. Sato, K. Watanabe,
H. Hisamoto, Y. Tobe and K. Kobiro, Anal. Chem., 1993, 65, 3404.
For a review, see: (d) K. Kobiro, Coord. Chem. Rev., 1996, 148, 135.
8 K. Kobiro, T. Matsuoka, S. Takada, K. Kakiuchi, Y. Tobe and
Y. Odaira, Chem. Lett., 1986, 713.
†
clinic, space group P21, a = 7.999(1), b = 17.241(1), c = 8.724(1)
Å, β = 108.48(1)Њ, V = 1141.1(2) Å3, Z = 2, Dc = 1.195 g cmϪ3
,
crystal dimensions 0.40 × 0.40 × 0.33 mm. A total of 2914
reflections were collected using the ω–2θ scan technique to a
maximum 2θ-value of 55.0Њ. The structure was solved by direct
methods (SHELXS86)34 and refined by a full-matrix least-
squares method with 270 variables and 2142 observed reflec-
tions [I > 3σ(I)]. The non-hydrogen atoms were refined aniso-
tropically. Hydrogen atoms were included but not refined. The
final R- and Rw-values were 0.084 and 0.095, respectively. The
maximum peak and the minimum peak in the final difference
map were 0.74 and Ϫ1.01 eϪ ÅϪ3
.
Crystal data for compound 6. C26H34O4, M = 410.55, mono-
clinic, space group C2/c, a = 17.263(4), b = 11.720(2), c =
11.716(2) Å, β = 109.18(1)Њ, V = 2238.8(6) Å3, Z = 4, Dc = 1.218
g cmϪ3, crystal dimensions 0.26 × 0.26 × 0.50 mm. A total of
2794 reflections were collected using the ω–2θ scan technique to
a maximum 2θ-value of 55.0Њ. The structure was solved by
direct methods (MULTAN88)35 and refined by a full-matrix
least-squares method with 136 variables and 1421 observed
reflections [I > 3σ(I)]. The non-hydrogen atoms were refined
anisotropically. Hydrogen atoms were included but not refined.
The final R- and Rw-values were 0.062 and 0.039, respectively.
The maximum peak and the minimum peak in the final differ-
9 R. A. Sachleben, M. C. Davis, J. J. Bruce, E. S. Ripple, J. L. Driver
and B. A. Moyer, Tetrahedron Lett., 1993, 34, 5373.
ence map were 0.55 and Ϫ0.32 eϪ ÅϪ3
.
Crystal data for lithium picrate complex of crown 1.
C24H34LiN3O11, M = 547.49, monoclinic, space group P21/c,
a = 11.429(2), b = 14.270(2), c = 16.453(2) Å, β = 101.57(1)Њ,
V = 2628.6(7) Å3, Z = 4, Dc = 1.383 g cmϪ3, crystal dimensions
0.35 × 0.28 × 0.35 mm. A total of 6589 reflections were
collected using the ω–2θ scan technique to a maximum of 2θ-
value of 55.0Њ. The structure was solved by direct methods
(SHELXS86)34 and refined by a full-matrix least-squares
method with 352 variables and 3109 observed reflections
[I > 3σ(I)]. The non-hydrogen atoms were refined anisotropi-
cally. Hydrogen atoms were included but not refined. The final
R- and Rw-values were 0.056 and 0.037, respectively. The maxi-
mum peak and the minimum peak in the final difference map
10 (a) K. Naemura, H. Miyabe, Y. Shingai and Y. Tobe, J. Chem. Soc.,
Perkin Trans. 1, 1993, 1073; (b) K. Naemura, K. Ueno, S. Takeuchi,
Y. Tobe, T. Kaneda and Y. Sakata, J. Am. Chem. Soc., 1993, 115,
8475; (c) K. Naemura, T. Mizooku, K. Kamada, K. Hirose, Y. Tobe,
M. Sawada and Y. Takai, Tetrahedron: Asymmetry, 1994, 5, 1549.
11 For a preliminary account of this work, see: K. Kobiro, M. Kaji,
S. Tsuzuki, Y. Tobe, Y. Tsuchiya, K. Naemura and K. Suzuki, Chem.
Lett., 1995, 831.
12 G. Berti, F. Bottari, B. Macchia and F. Macchia, Tetrahedron, 1965,
21, 3277.
13 G. Berti and F. Macchia, Tetrahedron Lett., 1965, 3421.
14 R. C. Fuson and J. T. Walker, Org. Synth., 1945, Coll. Vol. II, p. 169.
15 E. J. Corey, R. L. Danheiser and S. Chandrasekaran, J. Org. Chem.,
1976, 41, 260; A. Furstner, R. Csuk, C. Rohrer and H. Weidmann,
J. Chem. Soc., Perkin Trans. 1, 1988, 1729.
were 0.39 and Ϫ0.38 eϪ ÅϪ3
.
16 H. Stetter and E. Rauscher, Chem. Ber., 1960, 93, 1161.
17 A. P. Marchand, D. Xing, S. G. Bott, K. Ogawa and J. Harada,
Tetrahedron Lett., 1994, 35, 8935.
Crystal data for sodium perchlorate complex (1:2) of crown
20. C28H40ClNaO12, M = 627.06, monoclinic, space group C2/c,
a = 19.306(4), b = 15.798(4), c = 11.581(7) Å, β = 120.95(3)Њ,
V = 3029(2) Å3, Z = 4, Dc = 1.375 g cmϪ3, crystal dimensions
0.20 × 0.20 × 0.30 mm. A total of 3706 reflections were col-
lected using the ω–2θ scan technique to a maximum 2θ-value of
55.0Њ. The structure was solved by direct methods (SAPI91)36
and refined by a full-matrix least-squares method with 210 vari-
ables and 2138 observed reflections [I > 3σ(I)]. The non-
hydrogen atoms were refined anisotropically. Hydrogen atoms
were included but not refined. Owing to the crystallographic C2
symmetry, the oxygen atoms O(5), O(6), O(7) and O(8) of the
perchlorate anion were refined as a disordered structure around
the Cl atom with equivalent populations. The final R- and Rw-
values were 0.053 and 0.039, respectively. The maximum peak
and the minimum peak in the final difference map were 0.22
18 The most stable conformations were elucidated by a conformation
search using MacroModel V5.5: F. Mohamadi, N. G. J. Richards,
W. C. Guida, R. Liskamp, M. Lipton, C. Caulfield, G. Chang,
T. Hendrickson and W. C. Still, J. Comput. Chem., 1990, 11, 440.
19 For exchange rates of Liϩ and Naϩ complexes of crown ethers and
cryptands, see: E. Shchori, J. Jagur-Grodzinski and M. Shporer,
J. Am. Chem. Soc., 1973, 95, 3842; Y. M. Cahen, J. L. Dye and A. I.
Popov, J. Phys. Chem., 1975, 79, 1289, 1292; C. Chen, S. J. Wang and
S. C. Wu, Inorg. Chem., 1984, 23, 3901; N. Okoroafor and A. I.
Popov, Inorg. Chim. Acta, 1988, 148, 91.
20 K. A. Konnors, in Binding Constants, Wiley, New York, 1987,
pp. 24–28.
21 Owing to the difference between the conformations of the crown
rings of the crystal and MM3*-calculated structures, the overall
root-mean-square (RMS) deviations for the non-hydrogen atoms of
the observed and calculated structures 5 and 6 are 0.721 and 0.855
Å, respectively. However, the RMS deviations for the O᎐C᎐C᎐O
moiety are small; 0.100 and 0.074 Å, for structures 5 and 6,
respectively.
and Ϫ0.26 eϪ ÅϪ3
.
22 For example, it has been demonstrated that anti-dicyclohexano-14-
crown-4 adopts different conformations in solution and in the solid
state: G. W. Buchanan, R. A. Kirby and J. P. Charland, J. Am. Chem.
Soc., 1988, 110, 2477.
23 B. P. Hay and J. R. Rustad, J. Am. Chem. Soc., 1994, 116, 6316.
24 (a) J. H. Burns, R. A. Sachleben and M. C. Davis, Inorg. Chim.
Acta, 1994, 223, 125; (b) R. A. Sachleben and J. H. Burns, J. Chem.
Soc., Perkin Trans. 2, 1992, 1971.
† 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. 1, available via the RSC Web
pages (http://chemistry.rsc.org/rsc/p1pifa.htm). Any request to the
CCDC for this material should quote the full literature citation and the
reference number 207/170.
25 QCPE, 1969, No. 140. We are indebted to Professors Y. Fukazawa
and S. Usui of Hiroshima University for the PC version of this
program.
References and notes
1 For reviews, see D. Ammann, Ion-Selective Microelectrodes,
26 C. J. Pedersen, J. Am. Chem. Soc., 1970, 92, 386.
J. Chem. Soc., Perkin Trans. 1, 1998
493

View Article Online
27 K. Kobiro, K. Kakiuchi, Y. Tobe and Y. Odaira, Chem. Lett., 1984,
487.
28 Although crown ether 20 is known (CA Registry No. 88718-32-5),
its preparation and characterization have not been reported. See the
Experimental section for detail.
33 TEXRAY Structure Analysis Package, Molecular Structure Cor-
poration, 1985.
34 G. M. Sheldrick, Crystallographic Computing 3, ed. G. M. Sheldrick,
C. Kruger and R. Goddard, Oxford University Press, 1985,
p. 175.
29 In contrast to compound 20, it has been well documented that 12-
crown-4 forms a 2:1 complex with both Naϩ and Liϩ; P. P. Power,
Acc. Chem. Res., 1988, 21, 147 and references cited therein; F. P. van
Remoortere and F. P. Boer, Inorg. Chem., 1974, 13, 2071; F. P. Boer,
M. A. Neuman, F. P. van Remoortere and E. C. Steiner, Inorg.
Chem., 1974, 13, 2826.
35 T. Debaerdemaeker, G. Germain, P. Main, L. S. Refaat, C. Tate and
M. M. Woolfson, Computer Programs for the Automatic Solution
of Crystal Structures from X-ray Diffraction Data, University of
York, 1988.
36 H.-F. Fan, Structure Analysis Programs with Intelligent Control,
Rigaku Corporation, Tokyo, 1991.
30 K. Suzuki, K. Tohda, H. Aruga, M. Matsuzoe, H. Inoue and
T. Shirai, Anal. Chem., 1988, 60, 1714.
31 IUPAC Recommendation for Nomenclature of Ion-Selective
Electrodes, Pure Appl. Chem., 1976, 48, 29.
32 JIS K-0122, Japanese Standards Association, Tokyo, 1981.
Paper 7/06606F
Received 10th September 1997
Accepted 27th October 1997
494
J. Chem. Soc., Perkin Trans. 1, 1998