Synthesis and complexation properties of carbonyl-containing thiacalix[4]arenes

Article abstract of DOI:10.1007/s11172-008-0191-8

Stereoisomers of thiacalix[4]arenes unsubstituted at the upper rim and containing four carbonyl fragments have been synthesized for the first time. Their structures were studied by 1D and 2D NMR spectroscopy, IR spectroscopy, and mass spectrometry. The co

Full text of DOI:10.1007/s11172-008-0191-8

Russian Chemical Bulletin, International Edition, Vol. 57, No. 7, pp. 1477—1485, July, 2008
1477
Synthesis and complexation properties
of carbonylꢀcontaining thiacalix[4]arenes
S. E. Solovieva, S. R. Kleshnina, M. N. Kozlova, L. F. Galiullina, A. T. Gubaidullin,
Sh. K. Latypov, I. S. Antipin, and A. I. Konovalov
A. E. Arbuzov Institute of Organic and Physical Chemistry,
Kazan Research Center, Russian Academy of Sciences,
8 ul. Akad. Arbuzova, 420088 Kazan, Russian Federation.
Eꢀmail: svsol@iopc.knc.ru
Stereoisomers of thiacalix[4]arenes unsubstituted at the upper rim and containing four carbonyl
fragments have been synthesized for the first time. Their structures were studied by 1D and 2D NMR
spectroscopy, IR spectroscopy, and mass spectrometry. The complexation properties of the macroꢀ
cycles toward alkaline metal cations were estimated by the picrate extraction method. The absence
of the preorganization effect in the case of the thiacalixarenes unsubstituted at the upper rim is the
main reason for the sharp decrease in their extraction ability.
Key words: thiacalix[4]arenes, template synthesis, NMR spectroscopy, extraction, preorganizaꢀ
tion, alkaline metals.
Calixꢀ and thiacalix[4]arenes 1—6 are threeꢀdimenꢀ
sional structures with distinctly pronounced molecular
cavities.1—4 They contain active reaction centers, which
makes it possible to modify the upper and lower rims for
the creation of preorganized receptor structures. Addiꢀ
tional possibilities for the design of molecular receptors
are due to the existence of several stereoisomeric forms
of the calix[4]arene platforms: cone, partial cone, and
1,3ꢀ and 1,2ꢀalternates. Therefore, they find growing use
in various areas of supramolecular chemistry.2—6 Tetꢀ
raesters 3,7—9 phenylcarbonyl derivatives 4 (see Refs 10
and 11), and tetraamides 5 (see Refs 10, 12, and 13) based
on pꢀtertꢀbutylthiacalix[4]arene 1 in different conformaꢀ
tions have been synthesized earlier. It was found that the
efficiency and selectivity of extraction were affected by
both the spacial structure (macrocycle conformation)
and the nature of substituents at the lower rim.
Data on the influence of the tertꢀbutyl group at the
upper rim on the complexation properties of the thiaꢀ
calix[4]arenes substituted at the lower rim are presently
lacking. At the first glance, this effect should be negligiꢀ
ble, because the tertꢀbutyl groups possess weak electronꢀ
donating properties and are arranged rather far from
the complexation centers on the lower rim. However, it
should be mentioned that the calixarenes are allosteric
molecular systems14—17 in which even minor structural
changes at one side of the macrocycle can result in
a substantial steric reorganization of the binding sites
localized at another side.
Comꢀ
pound
R1
R2
Comꢀ
pound
R1
R2
1
2
3
4
5
But
H
But
But
But
H
H
6
7
8
9
H
H
H
H
CH2COOEt
CH2C(O)Ph
CH2C(O)NEt2
CH2C(O)NBu2
CH2COOEt
CH2C(O)Ph
CH2C(O)NEt2
on unsubstituted at the upper rim thiacalix[4]arene 2
(R1 = H). In order to obtain different stereoisomeric
forms (cone, partial cone, and 1,3ꢀalternate), we used the
In this connection, in the present work we synthesized
tetrasubstituted at the lower rim derivatives 7—9 based
Published in Russian in Izvestiya Akademii Nauk. Seriya Khimicheskaya, No. 7, pp. 1448—1456, July, 2008.
1066ꢀ5285/08/5707ꢀ1477 © 2008 Springer Science+Business Media, Inc.

1478
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 7, July, 2008
Solovieva et al.
H(10), H(10´)
a
template effect of an alkaline metal cation, which has earliꢀ
er been applied successfully to the functionalization of
pꢀtertꢀbutylthiacalix[4]arene (1) derivatives.11—13
H(2), H(6)
H(1)
H(7)
H(9), H(9´)
Results and Discussion
8
8
7
7
6
6
5
4
4
3
3
2
2
δ
δ
It is found for the reactions of thiacalix[4]arene 2 with
N,Nꢀdiethylbromoacetamide and αꢀbromoacetophenone
in acetone that alkaline metal carbonates used as bases
exert a substantial effect on the stereochemical result of
the reactions, although the stereoselectivity of the proꢀ
cess is lower, as a rule, compared to that of tertꢀbutyl
analog 1. The high stereoselectivity of these reactions was
observed for sodium and cesium carbonates: the cone
stereoisomers of compounds 7 and 8 are formed in the
presence of sodium carbonates in 66 and 59% yield,
respectively, and the 1,3ꢀalternate stereoisomers of the
same compounds in the presence of cesium carbonates
are formed in 64 and 69% yield, respectively. When poꢀ
tassium carbonate was used, the 1,3ꢀalternate stereoisoꢀ
mers of compounds 7 and 8 were isolated from the reacꢀ
tion mixture in 45 and 50% yield, respectively.
1
2
5
5
5
5
8
8
8
7
7
7
6
6
6
4
4
4
3
3
3
2
2
2
δ
δ
δ
3
4
Due to the absence of methylene bridges in a thiaꢀ
calixarene molecule (the NMR signals of their proꢀ
tons are usually used for the identification of the conforꢀ
mation of classical calix[4]arene derivatives),1 it is nonꢀ
trivial problem to establish the spatial structure of
the thiacalix[4]arene derivatives: the 1H and 13C NMR
spectra of the macrocycles in the symmetric cone and
1,3ꢀalternate conformations resemble each other by
the number and multiplicity of the signals. The strucꢀ
tures of macrocycles 7—9 were completely determined
by 2D NMR spectroscopy (2D COSY, HSQC, and
HMBC).18,19
b
The spatial structures of macrocycles 7 and 8 were
ultimately determined by an analysis of the nuclear Overꢀ
hauser effects (NOE). All experiments on NOE meaꢀ
surement were carried out in a rotating coordinate sysꢀ
tem. For example, the NOE on the protons at the C(7)
atom, Nꢀethyl substituent, and aromatic ring is obꢀ
served in the 1D ROESY spectrum20 of compound 8 by
the excitation of the proton at the C(1) atom (Fig. 1, a,
curve 1). A similar pattern is observed by the excitation
of the protons of the С(7)H2 methylene group (see Fig. 1, a,
curve 2) and Nꢀethyl fragments (see Fig. 1, a, curves 3
and 4). The observed effects indicate that these groups of
protons spatially are not far from each other, suggesting
unambiguously that this compound exists in the 1,3ꢀalꢀ
ternate stereoisomeric form (see Fig. 1, b).
Fig. 1. 1D ROESY spectra (a) and the main NOE values (b)
for compound 8 (CDCl3, 20 °С). The spectra correspond to the exciꢀ
tation of protons at the C(1) (1), C(7) (2), C(9) and C(9´) (3), and
C(10) and C(10´) (4) atoms.
raising (to 323 K) were unsuccessful. This evidences that
the chemical shifts of the corresponding exchangeable
protons differ strongly by the equilibrium components
(conformers). The observed pattern could be a result
of the equilibrium pinched cone—cone—pinched cone
(Fig. 2), which has previously been found for similar
structures.21
The lowꢀtemperature NMR experiments completely
confirmed this hypothesis. With a temperature decrease
compound 7 exhibits the strong broadening and then
splitting of the signals (Fig. 3), and the spectrum under
1
In the H NMR spectra of the cone stereoisomers of
compounds 7 and 8 at room temperature, the majority of
signals are strongly broadened, indicating the conformaꢀ
tional exchange, whose rate is intermediate in the
1Н NMR time scale. The attempts to obtain the spectra
under the fast exchange conditions by the temperature

Carbonylꢀcontaining thiacalix[4]arenes
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 7, July, 2008
1479
are unexpectedly high compared to analogous parameters
for the classical calixarenes. For instance, for the tetraꢀOꢀ
propyl calix[4]arene derivative under the comparable conꢀ
ditions, ΔG≠303 is lower than 9.8 kcal mol–1 21
The higher
.
activation energies of the interconversion for the larger
macrocyclic ring of the thiacalix[4]arene derivatives comꢀ
pared to the activation energies of the calix[4]arenes are
possibly caused by the stabilization of the С2v forms in the
Fig. 2. Interconversion between the pinched cone conformations
in compounds 7 and 8.
the slow exchange conditions is observed at 243 K. The
1H NMR spectrum contains two sets of signals (see Fig. 3)
corresponding to a lower C2v symmetry of the pinched
cone conformation. All signals in each half of this strucꢀ
ture were unambiguously assigned on the basis of
the 1D/2D homoꢀ and heterocorrelation experiments.
The combined use of the NOESY and ROESY methꢀ
ods made it possible to differentiate the NOE due to
the chemical exchange and to identify the spectral comꢀ
ponents of the exchange.
Theoretical calculations of chemical shifts of protons
are a reliable tool in an analysis of spacial structures of
compounds bearing anisotropic groups. Therefore, the
shielding effects of the protons of the calixarene platform
were calculated for compound 7. Geometry optimizaꢀ
tion (MM2 method)22—24 in the cone conformation gives
the structure with the С2v symmetry (Fig. 4) in which the
protons of the adjacent aromatic rings of the calixarene
rim exist in different magnetic environments.
This agrees qualitatively with the 1H NMR spectral
data at low temperatures in which the chemical shifts of
the corresponding signals of all adjacent groups differ
strongly. Moreover, the calculated shielding effects for
the protons of the calixarene rim (in the framework of
the circular current model on the geometry optimized by
the MM2 method)25—27 agree with those observed for the
corresponding protons at the quantitative level (Fig. 5).
An analysis of the full lineshapes (WinDNMR v.7.1.6
program)28 made it possible to determine the rate conꢀ
stant for the exchange between the forms pinched cone 1
and pinched cone 2 in the 223—273 K temperature range.
The activation parameters of this process in CDCl3
323 K
8.0
8.0
8.0
8.0
8.0
8.0
7.6
7.6
7.6
7.6
7.6
7.6
7.2
7.2
7.2
7.2
7.2
7.2
6.8
6.8
6.8
6.8
6.8
6.8
6.4
6.4
6.4
6.4
6.4
6.4
6.0
6.0
6.0
6.0
6.0
6.0
δ
313 K
δ
303 K
δ
293 K
δ
273 K
δ
263 K
δ
253 K
8.0
8.0
7.6
7.6
7.2
6.8
6.8
6.8
6.4
6.4
6.0
6.0
6.0
δ
243 K
were determined by the Eyring equation29,30
:
7.2
δ
7*
ΔH≠ = 10.6 kcal mol–1, ΔS≠ = –8.4 cal mol–1 deg–1, and
2, 6
2*, 6*
12*
11*, 13*
7
10, 14
11, 13
223 K
1*
1
12
10*, 14*
ΔG≠ = 13.1 kcal mol–1. The latter value is well consisꢀ
303
tent with the data obtained earlier for the tetraꢀOꢀpropyl
8.0
7.6
7.2
6.4
δ
and tetraꢀOꢀethyl derivatives of thiacalix[4]arene 2
(see Ref. 21) in the same solvent: ΔG≠
= 13.3 and
Fig. 3. Temperature dependence of the 1H NMR spectra of comꢀ
pound 7 in the cone conformation in CDCl3 (the numbers of
protons with asterisks designate the equivalent groups in the pinched
cone conformation).
303
11.9 kcal mol–1, respectively.
It should be mentioned that the activation energies of
the interconversion of the thiacalix[4]arene derivatives

1480
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 7, July, 2008
Solovieva et al.
Fig. 4. Optimized by the MM2 method geometry of compound 7 in the cone conformation: side view from two sides.
the С4v symmetry.8,31,32 These regularities are the same in
the classical analogs of compounds 1 and 2.31,32
case of the thiacalixarenes due to the conjugation of the
bridging sulfur atoms with the aromatic rings. The higher
stability of the pinched cone conformation in the case of
the unsubstituted at the upper rim thiacalix[4]arenes is
indirectly specified by the fact that starting
thiacalix[4]arene 2 in the crystalline state also has the С2v
symmetry according to the IR spectroscopic31,32 and Xꢀ
ray diffraction8 data.
This behavior principally differs from that of the tertꢀbuꢀ
tyl derivatives. According to the 2D NMR spectroscopy
(weak crossꢀpeaks of structurally equivalent protons),
the fast chemical exchange, viz., interconversion of
the С2v—С2v´ forms (293 K), occurs in amide 5 bearing
the tertꢀbutyl group.12 Crystalline macrocycle 1 with
the tertꢀbutyl substituents in the paraꢀposition in crystalꢀ
line phase is characterized by the cone conformation with
The aforesaid indicates that the steric arrangement of
substituents at the lower rim of thiacalixarenes unsubstiꢀ
tuted at the upper rim or bearing the tertꢀbutyl group at
this rim should substantially be different; therefore, their
complexation ability will also differ. This is indicated by
the data on extraction of alkaline metal cations by tetꢀ
raesters of the classical calix[4]arenes with both the subꢀ
stituted and free upper rim,33 and the macrocycles bearꢀ
ing the tertꢀbutyl groups at the upper rim exhibit the
better extraction properties. The selectivity remains unꢀ
changed: the classical analogs of compounds 3 and 6 in
the cone conformation preferentially extract the sodium
cation, and the same analogs in the 1,3ꢀalternate conforꢀ
mation preferentially extract the potassium cation. There
are no similar data on tetraamides and tetraphenylcarboꢀ
nyl derivatives of the classical calix[4]arenes.
In order to estimate the ability of new thiacalix[4]arene
derivatives 7 and 8 to bind alkaline metal ions, the liquid
extraction of their picrates (in a mutually saturated waꢀ
ter—dichloromethane system) was studied. The data on
the extraction of the alkaline metal cations by compounds
6—9 and literature data7,11,13 on the extraction of these
cations by tertꢀbutyl analogs 3—5 under comparable conꢀ
ditions are presented in Table 1.
In all cases, the extraction ability of the macrocyclic
ligands decreases substantially upon the removal of the
tertꢀbutyl group from the upper rim (see Table 1). The
decreasing the extraction ability depends on the macroꢀ
cycle conformation. For the stereoisomers in the cone
conformation, the effect of tertꢀbutyl group removal
turned out to be much greater than that for the stereoisoꢀ
mers in the 1,3ꢀalternate conformation. For instance,
macrocycles 6 and 7 in the cone conformation almost
completely lost the ability to extract alkaline metal catꢀ
ions. The selectivity of binding changes simultaneously
with the efficiency. The ester (3) and phenylcarbonyl (4)
derivatives in the 1,3ꢀalternate conformation extract
the potassium cation most efficiently, whereas macrocyꢀ
Δδ
2
1.2
0.8
0.4
2
1
1
1
2
H(2), H(6)
H(1)
H(7)
Fig. 5. Comparison of the theoretical25—27 (1) and experimental (2)
differences in the chemical shifts (Δδ) for protons in the adjacent
fragments of thiacalix[4]arene 7 in the pinched cone conformation
(Δδ = δ – δ ).
i
i
i*

Carbonylꢀcontaining thiacalix[4]arenes
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 7, July, 2008
1481
Table 1. Degree of extraction (Е) of alkaline metal picrates
by derivatives 3—9a
derivative 5 extracts the alkaline metal cations so efficientꢀ
ly that the concentrations of both the ligand and metal
ions should be decreased to estimate the selectivity of
extraction (see Table 1 and Fig. 6, a). As a whole, the better
ability to bind amide derivatives 5 and 8 is due to the fact
that the amides are stronger electron donating groups than
ester and phenylcarbonyl substituents of derivatives 3, 4,
6, and 7.
It turned out that amide 8 in the cone conformation
with the unsubstituted upper rim is selective to the lithiꢀ
um cation, whereas its tertꢀbutyl analog (tetraamide 5)
preferentially binds the sodium and potassium cations,
although its selectivity with respect to other alkaline metꢀ
al cations is not so high. From this point of view,
the extraction behavior of tetraamide 8 in the cone conꢀ
formation is remarkable: the extraction ability of the macꢀ
rocycle decreases with an increase in the cation size.
Taking into account the substantially higher energy
of lithium cation transfer from the aqueous to organic
phase in the series of alkaline metal cations,34 one can
speak about the creation of the highly efficient complexꢀ
ation agent to the lithium cation.
A different pattern is observed for the 1,3ꢀalternate
stereoisomers. Macrocycle 5 is characterized by the abꢀ
sence of selectivity toward the alkaline metal cations
(see Fig. 6, b), while the removal of the tertꢀbutyl group
from the upper rim produces selectivity. However, unꢀ
like compound 8 in the cone conformation, an inverse
dependence is observed in this case: the extraction ability
increases with an increase in the cation size.
The observed differences in the extraction ability
of macrocycles 5 and 8 can be caused by several factors,
first of all, by a decrease in the lipophilicity of the ligand
due to the removal of four hydrophobic tertꢀbutyl substiꢀ
tuents from the upper rim. Therefore, we synthesized
N,Nꢀdibutylamide derivative 9 in the 1,3ꢀalternate conꢀ
formation, whose lipophilicity is approximately equal
Comꢀ
pound
Conforꢀ
mation
E (%)
Refeꢀ
rence
Li+
Na+
K+
Cs+
3
1,3ꢀAlternate
Cone
1,3ꢀAlternate
Cone
5
5
9
54
33
85
100
(99)
100
(78)
3
8
7
0
81
70
76
84
25
99
67
7
7
7
11
11
13
4
19
12
100
(89)
100
(62)
0
0
3
0
76
10
100
(99)
100
(45)
34
7
39
0
96
14
96
46
5b
1,3ꢀAlternate
100
(94)
100
(80)
11
0
15
14
98
Cone
13
c
6
7
8
9
1,3ꢀAlternate
Cone
1,3ꢀAlternate
Cone
1,3ꢀAlternate
Cone
1,3ꢀAlternate
—
c
—
c
11
95
10
—
23
98
c
—
a The initial concentrations of ligands 3—9 in the organic phase
([L]0), metal hydroxide ([MOH]0) in the aqueous phase, and picric
acid ([HPic]0) in the aqueous phase were 2.5•10–3, 0.1, and
2.5•0–4 mol L–1, respectively.
b The data for [5]0 = 3.5•10–4 mol L–1, [MOH]0 = 7.0•10–5 mol L–1
and [HPic]0 = 5.0•10–5 mol L–1 are given in parentheses.
c Data of this work.
,
cles 6 and 7 in the same conformation have a very weak
extraction ability toward small alkaline metal cations but
rather efficiently extract the cesium cation and manifest a
good selectivity with respect to this cation.
As in the case of the aboveꢀconsidered ester and pheꢀ
nylcarbonyl derivatives, the efficiency and selectivity
change substantially for amides 5 and 8 in the both conꢀ
formations (Fig. 6, a, b). In addition, under the condiꢀ
tions used for the extraction of compound 8, tertꢀbutyl
E (%)
E (%)
a
b
100
80
60
40
20
100
80
60
40
20
5
5
8
8
Li+
Na+
K+
Cs+
Li+
Na+
K+
Cs+
Fig. 6. Dependences of the degree of extraction (Е) of the alkaline metal cations by compounds 5 (see Ref. 13) and 8 in the cone (a) and
1,3ꢀalternate (b) conformations. Extraction conditions: [5]0 = 3.5•10–4 mol L–1, [MOH]0 = 7.0•10–5 mol L–1, [HPic]0 = 5.0•10–5 mol L–1
[8]0 = 2.5•10–3 mol L–1, [MOH]0 = 0.1 mol L–1, and [HPic]0 = 2.5•10–4 mol L–1
;
.

1482
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 7, July, 2008
Solovieva et al.
to that of macrocycle 5.35 The data on the extraction of
the alkaline metal cations by compounds 8 and 9 are
presented in Table 1. It is seen that more lipophilic amide
9 extracts the alkaline metal cations nearly in the same
way as compound 8. Thus, the lipophilicity of the macroꢀ
cycles exerts no decisive effect on the observed change in
the extraction ability upon the removal of the tertꢀbutyl
group from the upper rim.
Thus, new derivatives of the unsubstituted at the upꢀ
per rim thiacalix[4]arenes 7—9 in the cone and 1,3ꢀalterꢀ
nate conformations have been synthesized for the first
time and characterized. The study of the extraction propꢀ
erties of the synthesized compounds by the picrate methꢀ
od showed that the removal of the tertꢀbutyl groups deꢀ
creases the extraction ability of the macrocycles, and the
selectivity can either increase or decrease depending on
the stereoisomer nature. Amide 8 is a less efficient but
more selective extragent compared to analogous pꢀtertꢀ
butyl thiacalix[4]arene derivative 5. Compound 8 in
the cone conformation is a new efficient and selective
extragent for the Li+ cation, and compounds 6 and 7 in
the 1,3ꢀalternate conformation are selective extragents of
the cesium cation. An increase in the lipophilicity of
compound 9 in the absence of tertꢀbutyl groups at
the upper rim exerts no effect on the extraction of
the alkaline metal cations. It is most likely that the main
factor of the lower extraction ability of the thiacalixareꢀ
nes unsubstituted at the upper rim is the absence of
the preorganization effect of macrocyclic platform.
Another possible reason for serious differences in the
complexation ability can be the effect of ligand preorganiꢀ
zation. We succeeded to obtain a single crystal of tetꢀ
raamide 8 in the 1,3ꢀalternate conformation by crystallizaꢀ
tion from acetonitrile and studied it by Xꢀray diffraction
analysis. The comparison of these and earlier13 obtained
data on the structure of compound 5 showed that the
steric structures of the free ligands differed strongly (Fig. 7).
The tertꢀbutyl groups of ligand 5 preorganized
the cavity for cation incorporation: the distance between
the oxygen atoms of the carbonyl groups at one part of
the macrocycle is 6.65 Å, the distance between the pheꢀ
noxylic oxygen atoms is 5.66 Å, and the distance between
the benzene rings at the opposite part of the macrocycle is
5.65 Å, i.e., the aromatic rings are parallel. In the abꢀ
sence of tertꢀbutyl groups, the aromatic rings in comꢀ
pound 8 are brought together and the amide groups move
apart: the distance between the oxygen atoms of the carꢀ
bonyl groups at one part of the macrocycle is 8.14 Å,
the distance between the phenoxylic oxygen atoms is
6.48 Å, and the distance between the benzene rings at
the opposite part of the macrocycle is 3.9 Å. The data
presented convincingly prove that macrocycle 8, unlike
macrocycle 5, is not sterically preorganized for the interꢀ
action with the cation, and an additional energy should be
consumed during the complexation process to reorganize
the binding sites of the ligand. Therefore, the absence of
the preorganization effect in the case of the thiacalixareꢀ
nes unsubstituted at the upper rim is the main reason for
their lower extraction ability.
Experimental
Solvents were purified prior to use according to known proꢀ
cedures.36
1D and 2D NMR spectra (DEPT, NOESY, 2D COSY, 2D
HSQC, and 2D HMBC) in CDCl3 were recorded on a Bruker
Avanceꢀ600 spectrometer (600.13 (1Н) and 150.92 MHz (13С));
residual signals of CНCl3 (δН 7.26) and CDCl3 (δС 77.16)37 were
used as internal standards; the NOE values in the laboratory and
rotating coordinates were measured by the 1D DPFGSE method.20
A VTꢀ2200 temperature unit was used for lowꢀtemperature NMR
experiments. The temperature was calibrated for a methanol sample
using a standard procedure. Prior to experiment the samples were
kept in a sensor for 15 min to achieve the temperature equilibrium.
The MM2 calculations (force constants)22—24 were performed
using the CS Chem3D Ultra 6.0 program (CambridgeSoft Corp.).
The shielding effects were calculated (in the framework of the circular
a
b
6.65 Å
8.14 Å
6.48 Å
5.66 Å
5.65 Å
3.9 Å
Fig. 7. Crystal structures of ligands 5 (a) and 8 (b) in the 1,3ꢀalternate conformation according to the Xꢀray diffraction data.

Carbonylꢀcontaining thiacalix[4]arenes
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 7, July, 2008
1483
current model)25—27 by the Shield program. In the presence
of chemical exchange, the full lineshape was calculated by
the WinDNMR v.7.1.6 program.28 The activation parameters were
calculated by the Eyring equation.29,30
concentrated to a volume of 5 mL, the precipitate formed was filꢀ
tered off and washed with methanol, and target product 7 in the 1,3ꢀ
alternate conformation was isolated. The yield was 0.88 g (45%), Rf
0.18 (chloroform as eluent).
IR spectra were recorded on a Bruker Vector 2ꢀ2 FTꢀIR
spectrometer (resolution ability 1 cm–1, accumulation 64 scans) in
KBr pellets in the 400—4000 cm–1 interval. Mass spectra were
obtained on a MALDIꢀTOF Dynamo Finnigan mass spectrometer.
1,8,9ꢀTrihydroxyanthracene or pꢀnitroaniline was used as the matꢀ
rix. UV spectra were recorded on a UVꢀVIS Spectrometer Lambꢀ
da 35 instrument (Perkin—Elmer).
Elemental analysis was performed on a Perkin—Elmer PE 2400
series 2 СHNS/О analyzer. Melting points of the substances were
determined on a Boetius smallꢀscale heating stage with an RNMK
05 visual device. The purity of the substances was monitored by
TLC on Silufol UV 254 plates (developed by the UV light of
a VLꢀ6.LC lamp (6Wꢀ254 nm tube) and, in several cases, by iodine
vapors).
The compounds were studied at the NMR Division of
the Federal Collective Spectral Analytical Center for the Physicoꢀ
chemical Investigation of Structure, Properties, and Composition of
Substances and Materials and the Federal Center for Physicocheꢀ
mical Investigation of Substances and Materials (State Contracts of
the Ministry of Education and Science of the Russian Federation
02.451.11.7036 and 02.451.11.7019).
αꢀBromoacetophenone38 and thiacalix[4]arene 2 39 were synꢀ
thesized by known procedures. 25,26,27,28ꢀTetrakis[(ethoxyꢀ
carbonyl)methoxy]ꢀ2,8,14,20ꢀtetrathiacalix[4]arene (6) in the cone
and 1,3ꢀalternate conformations were synthesized using a described
procedure.9
Synthesis of 25,26,27,28ꢀtetrakis[(phenylcarbonyl)methoxy]ꢀ
2,8,14,20ꢀtetrathiacalix[4]arene stereoisomers (7) (general proceꢀ
dure). A mixture of thiacalix[4]arene 2 (1.0 g, 2.0 mmol),
αꢀbromoacetophenone (3.98 g, 20 mmol), and anhydrous alkaline
metal carbonate (20 mmol) in anhydrous acetone (50 mL) was
refluxed for 31 h under argon atmosphere. The reaction course was
monitored by TLC. Then the reaction mixture was treated using
different methods depending on the base used.
Base Na2CO3. Dilute hydrochloric acid (50 mL) and chloroꢀ
form (100 mL) were added to the precipitate that was filtered from
the reaction mixture, the layers were separated, the organic layer was
washed with water to pH 7 and dried over MgSO4, and chloroform
was removed to a minimum volume (~10 mL). The precipitate
formed was filtered off and washed with a minor amount of chloroꢀ
form, and target product 7 in the cone conformation was isolated.
The yield was 1.18 g (61%), Rf 0.66 (acetone—hexane (1 : 1) as
eluent). The product (0.09 g, 5%) was additionally isolated from
the acetone filtrate.
Base К2CO3. The MALDIꢀTOF mass spectrum of the reaction
mixture, m/z: 991 [M + Na]+ (tetrasubstituted derivative). Accordꢀ
ing to the TLC data, the reaction mixture contained two products:
with Rf 0.30 (presumably, the partial cone stereoisomer) and Rf 0.18
(1,3ꢀalternate) (chloroform as eluent); no starting thiacalix[4]arene
was found.
The reaction mixture was filtered, and the filtrate was concenꢀ
trated on an oil pump to a small volume. The precipitate formed was
filtered off and washed with diethyl ether. The solvent from the filtrate
was removed until a viscous resinꢀlike residue was formed to which
diethyl ether was added. The precipitate formed was filtered off,
dissolved in dichloromethane (20 mL), treated with dilute HCl,
washed with water to pH 7, and dried over MgSO4. The mixture was
Base Cs2CO3. The reaction mixture was filtered, and the preꢀ
cipitate was washed with acetone (75 mL). The solvent from
the joined acetone filtrates was removed to dryness, and diethyl ether
(75 mL) was added to the solid residue. Then the precipitate
was filtered off and washed several times by diethyl ether.
A small amount of chloroform was added to the precipitate. The
nonꢀdissolved portion was filtered off, and target product 7 in
the 1,3ꢀalternate conformation was isolated. The yield was
1.23 g (64%).
Compound 7, cone conformation. White powder, m.p. >350 °C.
Found (%): C, 69.34; H, 4.10; S, 13.12. C56H40O8S4. Calculatꢀ
ed (%): С, 69.42; H, 4.13; S, 13.22. IR, ν/cm–1: 1704 (С=О), 3056
(CH). 1H NMR (313 K), δ: 5.9 (br.s, 8 H, OCH2); 6.7 (br.s, 4 H,
Н(1)); 7.0 (br.s, 8 H, Н(2), Н(6)); 7.2 (br.s, 8 H, Н(11), Н(13)); 7.4
(t, 4 H, Н(12), J = 7.5 Hz); 7.9 (d, 8 H, Н(10), Н(14), J = 7.5 Hz).
13С NMR (223 K), δ: 74.1, 75.2 (С(7)), 123.0, 126.2 (С(1)), 126.8,
127.9 (С(10), С(14)), 128.3, 128.7 (С(11), С(13)), 130.7, 131.6
(С(5), С(3)), 133.4, 133.7 (С(12)), 131.8, 137.6 (С(6), С(2)),
155.2, 159.7 (С(4)), 193.6, 195.2 (С(8)). MALDIꢀTOF MS, m/z:
991 [M + Na]+.
Compound 7, 1,3ꢀalternate conformation. White powder,
m.p. >350 °C. Found (%): C, 69.34; H, 4.19; S, 13.24.
C56H40O8S4. Calculated (%): С, 69.42; H, 4.13; S, 13.22. IR,
ν/cm–1: 1705 (С=О), 3058 (CH). 1H NMR, δ: 5.36 (s, 8 H,
OCH2); 6.80 (t, 4 H, pꢀH arom., J = 7.0 Hz); 7.52 (t, 8 H,
mꢀH arom., J = 8.0 Hz + d, 8 H, mꢀH arom., J = 7.0 Hz); 7.63
(t, 4 H, pꢀH arom., J= 8.0 Hz); 8.00 (d, 8 H, oꢀH arom., J = 8.0 Hz).
MALDIꢀTOF MS, m/z: 1101 [M + Cs]+.
Synthesis of 25,26,27,28ꢀtetrakis[(N,Nꢀdiethylcarbamoyl)ꢀ
methoxy]ꢀ2,8,14,20ꢀtetrathiacalix[4]arene stereoisomers (8) (genꢀ
eral procedure). A mixture of thiacalix[4]arene 2 (1.00 g, 2.0 mmol),
N,Nꢀdiethylbromoacetamide (5.7 mL, 40 mmol), and anhydrous
metal carbonate (20 mmol) in anhydrous acetone (50 mL)
was refluxed for 48 h under argon atmosphere. The reaction course
was monitored by TLC. Then the reaction mixture was treated using
different methods depending on the base used.
Base Na2CO3. The reaction mixture was poured into water and
extracted with chloroform (~350 mL), and the extract was treated
with dilute hydrochloric acid, washed with water to pH 7, and dried
over MgSO4. Chloroform was removed in vacuo. An excess of the
starting BrCH2CONEt2 was removed in vacuum of an oil pump
(bath temperature 130 °С) nearly to dryness, and the residue (resinꢀ
like mass) was subjected to column chromatography on silica gel 60
(Lancaster, 220—440 mesh, for column chromatography, chloroꢀ
form—hexane (1 : 2) mixture as eluent). Target product 8 in the cone
conformation was isolated. The yield was 1.10 g (59%), Rf 0.87
(acetone—hexane (1 : 1) as eluent).
Base К2CO3. The MALDIꢀTOF mass spectrum of the reaction
mixture, m/z: 951 [M + H]+, 973 [M + Na]+, 989 [M + K]+
(tetrasubstituted products). According to the TLC data (aceꢀ
tone—hexane (1 : 1.5) as eluent), the reaction mixture contained two
products with Rf 0.24 and 0.51 (presumably, the partial cone
stereoisomer). The precipitate was filtered off from the reaction
mixture and washed with acetone. The filtrate was evaporated on
a rotary evaporator until a viscous resinꢀlike residue was formed
(bath temperature 32 °С). Colorless crystals that formed were filꢀ
tered off, washed with diethyl ether, and recrystallized from acetoniꢀ

1484
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 7, July, 2008
Solovieva et al.
trile, and product 8 in the 1,3ꢀalternate conformation was isolated.
The yield was 0.95 g (50%).
were prepared by the dissolution of weighed samples of picric acid in
aqueous solutions of metal hydroxides (MOH), which were pretiꢀ
trated with a 0.1 M solution of hydrochloric acid.
Base Cs2CO3. The reaction mixture was cooled to room temꢀ
perature, water was added, and the mixture was treated with dilute
hydrochloric acid. The mixture was extracted with chloroform
(250 mL), and the organic layer was washed with water to pH 7 and
dried over MgSO4. Chloroform was removed on a rotary evaporator
(bath temperature 30 °С). An excess of BrCH2CONEt2 was reꢀ
moved in vacuum of a rotary evaporator (bath temperature >200 °С)
almost to dryness. After cooling to room temperature, diethyl ether
was added to the solid residue. The precipitate was filtered off,
washed with diethyl ether (~30 mL), and recrystallized from acetoꢀ
nitrile, and target product 8 in the 1,3ꢀalternate conformation
was isolated. The yield was 1.3 g (69%), Rf 0.24 (acetone—hexane
(1 : 1.5) as eluent).
A solution of ligands 6—9 in dichloromethane was added
to 5 mL of an aqueous solution of MOH (С = 0.1 mol L–1) and
HPic (С = 2.5•10–4 mol L–1). The biphase system was stirred for
30 min in a closed flask on a magnetic stirrer (IKAꢀWERKE RO
15 power) and then left to stay for 90 min for phase separation.
Absorbances of the aqueous phase before and after extraction (А0 and
Аi, respectively) were determined at 355 nm.40,41 The extraction
percentage (Е) was calculated by the equation
E = [(A0 – Ai)/A0]•100.
Experiments on extraction were carried out for the initial conꢀ
centration of ligands 3—9 equal to 2.5•10–3 mol L–1 and for
Compound 8, cone conformation. Yellowish powder, m.p. 295 °C.
Found (%): C, 60.34; H, 6.31; N, 5.75; S, 13.42. C48H60N4O8S4.
[5]0 = 3.5•10–4 mol L–1
.
Xꢀray diffraction analysis of compound 8 in the 1,3ꢀalternate
conformation was performed on an EnrafꢀNonius CADꢀ4 fourꢀ
circle automated diffractometer. The crystals of compound 8
(C48H60N4O8S4) were tetragonal, at 20 °С а = b = 16.829(4) Å,
c = 17.889(6) Å, V = 5066(2) Å3, Z = 4, dcalc = 1.24 g cm–3, space
group I41/a. The unit cell parameters and intensities of 5370 reflecꢀ
tions, 1518 of which had I ≥ 2σ, were measured at 20 °С (CuꢀKα
radiation, λ(CuꢀKα) = 1.54184 Å, graphite monochromator, ω/2θ
scan mode, θ = 74.14°). No decrease in the intensities of three
control reflections was observed during the experiment, and an
absorption correction was applied (μ(Cu) = 2.16 mm–1). The crystal
structure was solved by a direct method using the SIR program,42
and nonꢀhydrogen atoms were refined in the fullꢀmatrix anisotropic
approximation for F 2 using the SHELXL program.43 The coordiꢀ
nates of hydrogen atoms were calculated on the basis of stereochemꢀ
ical criteria and refined by the corresponding riding models. The final
values of the R factors were R = 0.0561, Rw = 0.1471 for 1518
reflections with I2 ≥ 2σ; R = 0.1060, Rw = 0.1752 for all reflections.
The fitting parameter for F 2 was 1.022. All calculations were perꢀ
formed by the MolEN44 and WINGX45 programs, and the figures
of molecules were drawn using the PLATON program.46 The Xꢀray
diffraction data for compound 8 were deposited with the Cambridge
Structure Data Collection (CCDC no. 665 719).
Calculated (%): С, 60.76; H, 6.33; N, 5.91; S, 13.50. IR, ν/cm–1
:
1672 (C=O), 2957 (CH). 1H NMR, δ: 0.84, 0.88 (both t, 12 H each,
CH3, J = 4.7 Hz); 3.38, 3.55 (both br.s, 8 H each, NCH2); 5.22
(br.s, 8 H, OCH2); 6.63 (t, 4 H, pꢀH arom., J = 5.2 Hz); 7.51
(d, 8 H, mꢀH arom., J = 5.1 Hz). MALDIꢀTOF MS, m/z: 973
[М + Na]+.
Compound 8, 1,3ꢀalternate conformation. White crystals,
m.p. 243 °C. Found (%): C, 60.71; H, 6.23; N, 5.82; S, 13.30.
C48H60N4O8S4. Calculated (%): С, 60.76; H, 6.33; N, 5.91;
S, 13.50. IR, ν/cm–1: 1660 (C=O), 2960 (CH). 1H NMR, δ: 1.17,
1.26 (both t, 12 H each, CH3, J = 6.8 Hz); 3.35, 3.52 (both q,
8 H each, NCH2, J = 6.8 Hz); 4.68 (s, 8 H, OCH2); 6.87 (t, 4 H,
pꢀH arom., J = 7.6 Hz); 7.70 (d, 8 H, mꢀH arom., J = 7.6 Hz).
13С NMR, δ: 13.09 (С(10)), 14.50 (С(10´)), 40.19 (С(9)), 41.56
(С(9´)), 71.05 (С(7)), 123.85 (С(1)), 128.8 (С(3), С(5)), 138.62
(С(2), С(6)), 160.9 (С(4)), 166.72 (С(8)). MALDIꢀTOF MS, m/z:
952 [М + Н]+, 973 [М + Na]+, 989 [M + K]+, 1083 [M + Cs]+.
25,26,27,28ꢀTetrakis[(N,Nꢀdibutylcarbamoyl)methoxy]ꢀ
2,8,14,20ꢀtetrathiacalix[4]arene (9), 1,3ꢀalternate conformation.
A mixture of calix[4]arene 2 (0.363 g, 0.72 mmol), Cs2CO3 (2.36 g,
7.24 mmol), and BrCH2CONBu2 (3.62 g, 14.48 mmol) in anhyꢀ
drous acetone (20 mL) was stirred for 60 h under argon atmosphere
at the boiling point of acetone. The completeness of the reaction was
monitored by TLC and MALDIꢀTOF mass spectrometry. After the
end of the reaction, water was added to the reaction mixture, which
was extracted with chloroform (250 mL), the organic layer was
dried over MgSO4, and the solvent was removed until a viscous
resinꢀlike residue was formed and subjected to column chromatogꢀ
raphy on silica gel 60 (Lancaster, 220—440 mesh for column
chromatography, chloroform as eluent). The crystals, formed after
solvent excess was removed from the main fraction, were washed
with acetonitrile, and target product 9 in the 1,3ꢀalternate conforꢀ
mation was obtained. The yield was 0.42 g (58%), white needleꢀlike
crystals, m.p. >350 °C, Rf 0.23 (chloroform as eluent). Found (%):
C, 65.55; H, 7.97; N, 4.70; S, 10.93. C64H92N4O8S4. Calculated
(%): С, 65.53; H, 7.85; N, 4.78; S, 10.92. IR, ν/cm–1: 1660 (С=О),
Sh. K. Latypov thanks the Russian Science Support
Foundation (grant for young doctors of science).
This work was financially supported by the Russian
Foundation for Basic Research (Project Nos 07ꢀ03ꢀ00834,
05ꢀ03ꢀ33008, and 05ꢀ03ꢀ32558).
References
1. C. D. Gutsche, Calixarenes Revisited, Monographs in Supramoꢀ
lecular Chemistry, Ed. J. F. Stoddart, RSC, London, 1998, 233 pp.
2. Сalixarenes in Action, Eds L. Mandolini, R. Ungaro, Imperial
College Press, London, 2000, 271 pp.
1
3055 (CH). H NMR, δ: 0.92—1.00 (m, 24 H, CH3); 1.33, 1.43
(both m, 16 H each, CH2CH2); 3.33, 3.46 (both t, 8 H each,
NCH2, J = 7.5 Hz); 4.69 (s, 8 H, OCH2); 6.86 (t, 4 H, pꢀH arom.,
J = 7.6 Hz); 7.76 (d, 8 H, mꢀH arom., J = 7.6 Hz). MALDIꢀTOF
MS, m/z: 1173 [M + Н]+.
Extraction studies. Picric acid (HPic), LiOH, NaOH, KOH,
and CsOH (all reagent grade) were used. Dichloromethane (reagent
grade) was purified by a standard method.36 Alkaline metal picrates
3. Calixarenes 2001, Eds Z. Asfari, V. Bohmer, J. Harrowfield,
J. Vicens, Kluver Academic Publishers, Netherlands, 2001,
677 pp.
4. Сalixarenes in Nanoworld, Eds J. Vicens, J. Harrofield, Springꢀ
er, Netherlands, 2007, 396 pp.
5. N. Morohashi, F. Narumi, N. Iki, T. Hattori, S. Miyano, Chem.
Rev., 2006, 106, 5291.

Carbonylꢀcontaining thiacalix[4]arenes
Russ.Chem.Bull., Int.Ed., Vol. 57, No. 7, July, 2008
1485
6. P. Lhotak, Eur. J. Org. Chem., 2004, 1675.
7. N. Iki, F. Narumi, T. Fujimoto, N. Morohashi, S. Miyano,
J. Chem. Soc., Perkin Trans. 2, 1998, 2745.
28. http://www.chem.wisc.edu/areas/reich/plt/windnmr.htm.
29. O. Michinori, Application of Dynamic NMR Spectroscopy to
Organic Chemistry, VCH, Weinheim, 1985.
8. H. Akdas, L. Bringel, E. Graf, M. W. Hosseini, G. Mislin,
J. Pansanel, A. D. Cian, J. Fisher, Tetrahedron Lett., 1998,
39, 2311.
9. P. Lhotak, V. Stastny, P. Zlatuskova, I. Stibor, V. Michlova,
M. Tkadlecova, J. Havlicek, J. Sykora, Collect. Czech. Chem.
Commun., 2000, 65, 757.
10. R. Lamartine, C. Bavoux, F. Vocanson, A. Martin, G. Senlis,
M. Perrin, Tetrahedron Lett., 2001, 42, 1021.
11. I. I. Stoikov, O. A. Omran, S. E. Solovieva, Sh. K. Latypov,
K. M. Enikeev, A. T. Gubaidullin, I. S. Antipin, A. I. Konoꢀ
valov, Tetrahedron, 2003, 59, 1469.
12. S. E. Solovieva, A. O. Omran, M. Gruner, W. D. Habicher, I. S.
Antipin, A. I. Konovalov, Zh. Strukt. Khim., 2005, 46, S19
[Russ. J. Struct. Chem., 2005, 46 (Engl. Transl.)].
13. S. E. Solovieva, M. Gruner, A. O. Omran, A. T. Gubaidullin,
I. A. Litvinov, W. D. Habicher, I. S. Antipin, A. I. Konovalov,
Izv. Akad. Nauk, Ser. Khim., 2005, 2041 [Russ. Chem. Bull., Int.
Ed., 2005, 54, 2104].
14. A. Arduini, E. Ghidini, A. Pochini, R. Ungaro, G. D. Andreetti,
L. Calestani, F. Ugozzoli, J. Incl. Phenom., 1988, 6, 119.
15. A. Ikeda, Y. Suzuki, M. Yoshimura, S. Shinkai, Tetrahedron,
1998, 54, 2497.
30. H. Friebolin, Basic Oneꢀ and Twoꢀdimentional NMR Spectrosꢀ
copy, VCH, Weinheim, 1991.
31. V. I. Kovalenko, A. V. Chernova, E. I. Borisoglebskaya, S. A.
Katsyuba, V. V. Zverev, R. R. Shagidullin, I. S. Antipin, S. E.
Solovieva, I. I. Stoikov, A. I. Konovalov, Izv. Akad. Nauk, Ser.
Khim., 2002, 762 [Russ. Chem. Bull., Int. Ed., 2002, 51, 825].
32. S. Katsyuba, V. Kovalenko, A. Chernova, E. Vandyukova,
V. Zverev, R. Shagidullin, I. Antipin, S. Solovieva, I. Stoikov,
A. Konovalov, Org. Biomol. Chem., 2005, 3, 2558.
33. K. Ivamoto, S. Shinkai, J. Org. Chem., 1992, 57, 7066.
34. A. F. Danil de Namour, A. Pugliese, A. R. Casal, M. B. Llerena,
P. J. Aymoninoc, F. J. Sueros Velardea, Phys. Chem. Chem.
Phys., 2000, 2, 4355.
35. F. Daniels, R. A. Alberty, Physical Chemistry, J. Wiley and
Sons, New York—London—Sydney—Toronto, 1974.
36. A. Weisberger, E. S. Proskauer, J. A. Riddick, E. E. Toops,
Organic Solvents, Interscience, New York, 1955.
37. H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem., 1997,
62, 7512.
38. Organic Syntheses, Coll. Vol. 2, Ed. A. Blatt, New York, 1946.
39. H. Kumagai, M. Hasegawa, S. Miyanari, Y. Sugawa, Y. Sato,
T. Hori, S. Ueda, H. Kamiyama, S. Miyano, Tetrahedron Lett.,
1997, 38, 3971.
40. K. E. Koenig, G. M. Lein, P. Stuckler, T. Kaneda, D. J. Cram,
J. Am. Chem. Soc., 1979, 101, 3553.
41. R. C. Helgeson, G. R. Weisman, J. L. Toner, T. L. Tarnowski,
Y. Chao, J. M. Mayer, D. J. Cram, J. Am. Chem. Soc., 1979,
101, 4928.
16. Z. Asfari, V. Lamare, J. F. Dozol, J. Vicens, Tetrahedron Lett.,
1999, 40, 691.
17. J. Budka, P. Lhotak, V. Michlova, I. Stibor, Tetrahedron Lett.,
2001, 42, 1583.
18. A. E. Derome, Modern NMR Techniques for Chemistry Reꢀ
search, Pergamon, Cambridge, 1988, 280 pp.
19. AttaꢀurꢀRahman, One and Two Dimensional NMR Spectroscoꢀ
py, Elsevier, Amsterdam, 1989.
42. A. Аltomare, G. Cascarano, C. Giacovazzo, D. Viterbo, Acta
Crystallogr., Sect. A, 1991, 47, 744.
20. K. Stott, J. Stonehouse, J. Keeler, T. L. Hwang, A. J. Shaka,
J. Am. Chem. Soc., 1995, 117, 4199.
21. M. Cajan, P. Lhotak, J. Lang, H. Dvorakova, I. Stibor, J. Koca,
J. Chem. Soc., Perkin Trans. 2, 2002, 1922.
22. N. L. Allinger, J. Am. Chem. Soc., 1977, 99, 8127.
23. N. L. Allinger, R. A. Kok, M. R. Imam, J. Comput. Chem.,
1988, 9, 591.
24. J.ꢀH. Lii, S. Gallion, C. Bender, H. Wikstrom, N. L. Allinger,
K. M. Flurchick, M. Teeter, J. Comput. Chem., 1989, 10, 503.
25. J. S. Waugh, R. W. Fessenden, J. Am. Chem. Soc., 1957,
79, 846.
43. G. M. Sheldrick, SHELXꢀ97. Programs for Crystal Structure
Analysis (Release 97ꢀ2), University of Göttingen, Germany,
1997, 154 pp.
44. L. H. Straver, A. J. Schierbeek, MolEN. Structure Determinaꢀ
tion System. Vol. 1. Program Description, Nonius B.V., Delft,
1994, 180 pp.
45. L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
46. A. L. Spek, Acta Crystallogr., Sect. A, 1990, 46, 34.
26. C. E. Johnsonand, F. A. Bovey, J. Chem. Phys., 1958, 29, 1012.
27. C. W. Haigh, R. B. Mallion, Progr. Nucl. Magn. Reson. Specꢀ
trosc., 1980, 13, 303.
Received December 18, 2007;
in revised form February 22, 2008