Synthesis of tripodal aza crown ether calix[4]arenes and their supramolecular chemistry with transition-, alkali metal ions and anions

Article abstract of DOI:10.1016/S0040-4020(02)01447-3

Tripodal aza crown ether calix[4]arenes, 5a, 5b, 6a and 6b, have been synthesized. The structure of protonated 5a was elucidated by X-ray crystallography to be a self-threaded rotaxane. Complexation studies of 5a and 5b towards anions using Na⁺

Full text of DOI:10.1016/S0040-4020(02)01447-3

Tetrahedron 58 (2002) 10277–10285
Synthesis of tripodal aza crown ether calix[4]arenes
and their supramolecular chemistry with transition-,
alkali metal ions and anions
a
a
a
Thawatchai Tuntulani,a Praput Thavornyutikarn, Sirilux Poompradub, Nongnuj Jaiboon,
,
*
Vithaya Ruangpornvisuti,a Narongsak Chaichit,b Zouhair Asfaric and Jacques Vicensc
aDeparment of Chemistry, Faculty of Science, Chulalongkorn University, Bangkok 10330, Thailand
bDepartment of Physics, Faculty of Science, Thammasat University at Rangsit, Pathumthani 12121, Thailand
´ ´ ´
ECPM, Group de Chimie des Interactions Moleculaires Specifiques, associe au CNRS, 25, rue Becquerel,
c
F-67087 Strasbourg, Cedex 2, France
Received 27 August 2002; revised 14 October 2002; accepted 7 November 2002
Abstract—Tripodal aza crown ether calix[4]arenes, 5a, 5b, 6a and 6b, have been synthesized. The structure of protonated 5a was elucidated
by X-ray crystallography to be a self-threaded rotaxane. Complexation studies of 5a and 5b towards anions using Naþ as countercation were
carried out by 1H NMR titration in dimethylsulfoxide-d6 and the mixture of chloroform-d and methanol-d4, respectively. Ligands 5a and 5b
were able to form 1:1 complexes with Br2, I2 and NO32 and the complexation stability varied as follows: NO23 .I2.Br2. The effect of
countercation on anion complexation was also investigated. The results showed that the association constants of 5a towards Br2 in the
presence of various cations varied as Kþ.Bu4Nþ.Naþ. The enhancement in anion complexation ability of 5a may result from the
rearrangement of the tripodal ammonium unit in the presence of Kþ. The neutral forms, 6a and 6b, were able to form complexes with
transition metal ions such as Co2þ, Ni2þ, Cu2þ and Zn2þ. The stability of the complexes followed the sequence: Ni2þ,Co2þ!Cu2þ@Zn2þ
.
Compounds 6a and 6b may, therefore, potentially be used as either transition metal ion or anion receptors that can be controlled by pH of the
solution. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
receptors for cations, anions and neutral molecules.10 It is
our goal to design and synthesize a versatile receptor for
alkali metals, transition metals and anions which can be
switched in its function by external stimuli. We chose p-tert-
butylcalix[4]arene as the building block and combined it
with a crown ether and the tren unit to synthesize a
compound that has great potential to bind either metal ions
or anions. We report herein the synthesis of the tripodal-
amine capped benzo crown p-tert-butylcalix[4]arenes, 6a
and 6b. The complexation of 6a and 6b with transition metal
ions has been studied by potentiometric titration. In
addition, the anion complexation studies of the ammonium
During the past decades, supramolecular chemistry has
become one of the most popular branches of chemistry due
to its versatile applications in environmental and material
science. Chemists have synthesized many types of receptors
for cations and anions.1 Lehn and co-workers have
synthesized macrobicyclic and macrotricyclic polyamines
and found that the protonated forms can bind anions such as
F
2 and Cl2.2 Bowman–James and co-workers employed a
bicyclic polyammonium receptor derivatized from tris-
(2-amino)ethylamine (tren), L, possessing C3 symmetry to
bind nitrate3 and fluoride4 ions. They found by X-ray
crystallography that such a polyammonium macrocycle was
able to include two nitrates in it cavity.5 In 2000, Nelson
and colleague reported the binding properties of the same
ligand with the transition metal series Co(II)–Zn(II).6
Our group has synthesized hosts for cations,7 anions8 and
neutral9 molecules using calix[4]arene as a building block.
Recently, derivatives of calixarenes have been employed as
Keywords: p-tert-butylcalix[4]arene; aza crown ether; ammonium groups;
anion receptor; metal ion receptor.
*
Corresponding author. Tel.: þ66-2-2187640; fax: þ66-2-2187598;
e-mail: tthawatc@chula.ac.th
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)01447-3

10278
T. Tuntulani et al. / Tetrahedron 58 (2002) 10277–10285
Scheme 1.
derivative of 6a and 6b, 5a and 5b, have been performed in
the presence of various countercations in order to under-
stand the role of cations towards anion binding ability.
reactions of p-tert-butylcalix[4]arene with 3.0 equiv. of
2-(20-bromoethoxy)benzaldehyde, 1a, and 4-(20-bromo-
ethoxy)benzaldehyde, 1b, respectively, were carried out in
the presence of a base to produce trialdehyde precursors, 3a
and 3b, for preparing the tripodal-amine capped calix[4]-
arene. The synthesis of 3a was reported previously in
acetonitrile using K2CO3 as base.12 This reaction gave only
6% yield of the desired trialdehyde derivative. Furthermore,
substitution reactions using K2CO3 always gave the
dialdehyde derivatives, 2a and 2b, in high yields.7,13
Since then, a number of bases and solvents have been
employed to optimize the yields of the desired products.
However, it was found that reactions in the presence of
strong bases such as KOH underwent Cannizzaro reactions
and gave both alcohol and carboxylic acid derivatives
instead.14Finally, we found that reactions using BaO in
DMF gave higher yields of trialdehyde calix[4]arenes, 3a
(21%) and 3b (46%), than those of dialdehyde calix[4]-
arenes, 2a (20%) and 2b (2%). It should be noted that the
yield of 3b was twice as much as that of 3a probably due to
the lower steric hindrance of the para isomer facilitating the
substitution reaction. The 1H NMR spectrum of 3a showed
(CvO)H at 10.41 and 9.74 ppm and at 9.76 and 9.68 ppm
for 3b in 1:2 integral ratios. FAB MS and elemental analysis
results of 3a and 3b were in agreement with the proposed
structure.
2. Results and discussion
2.1. Design, synthesis and characterization
Our desired receptors must contain both cation and anion
binding sites. Ammonium groups and crown ethers are well-
known to bind anions via electrostatic interactions and
cations via coordination bonds, respectively. In this paper,
we employ p-tert-butylcalix[4]arene as a supramolecular
building block. Three ethoxy benzaldehyde groups have
been attached to the calix[4]arene unit to form a crown
ether-like cation ionophore. Condensation reactions of
trialdehyde calix[4]arene with tris(2-amino)ethylamine
and the subsequent reduction and protonation generate
tripodal ammonium receptor unit, 5a and 5b. Compounds
5a and 5b thus possess both metal ion and anion ionophores
next to each other separated by a spacer such as a benzene
unit. The neutral forms of 5a and 5b, compounds 6a and 6b,
contain four nitrogen lone pairs and should thus form
complexes with transition metal ions.
The synthesis of desired compounds 5a, 5b,11 6a and 6b
was carried out as shown in Scheme 1. Substitution
Condensation reactions of 3a and 3b with 1.1 equiv. of

T. Tuntulani et al. / Tetrahedron 58 (2002) 10277–10285
10279
2.3. Anion binding studies and effects of alkali metal ions
towards anion binding ability
The charge and geometry of anions were considered in our
investigation. Therefore, we chose to investigate spherical
anions (F2, Br2 and I2), a trigonal planar anion (CO232), an
angular anion (AsO22) and tetrahedral anions (H2PO24 ,
1
HPO422, SO422 and PO432). H NMR (200 MHz) titrations
were employed in complexation studies of 5a and 5b
towards anions. NMR titrations for 5a were carried out in
DMSO-d6 due to the solubility of 5a and anion salts in that
solvent. The compound 5b, however, was not soluble in
DMSO-d6 but soluble in CDCl3. Anion salts dissolved quite
well in CD3OD. The NMR titrations for 5b were thus
carried out in the mixture of CDCl3 and CD3OD.
Figure 1. Crystal structure of 5a. Hydrogen atoms were omitted for clarity.
Although excess NaF and Na2SO4 was added to solutions of
5a and 5b, no chemical shift displacement of any signals in
the NMR spectra was observed. Therefore F2 and SO422 did
not form complexes with 5a and 5b. This is probably due to
the size of F2 and SO422 which is not appropriate for the
ligand cavity.
tris(2-amino)ethylamine in a mixture of CH3CN and
CH2Cl2 (high dilution) precipitated imines or Schiff base
products, 4a (95%) and 4b (97%). The signals due to
(CvO)H proton disappeared, and the signals due to
1
RNvCH protons showed at 8.83 and 8.93 ppm in the H
In the case of anions such as AsO22 , CO232, PO432, HPO242
and H2PO24 (using sodium salts), we observed interesting
phenomena when complexation studies of ligand 5a were
performed. Upon increasing the mole ratio of anions, white
solids precipitated from the mixture of 5a and AsO22 , CO322
and PO342. Therefore, NMR titrations could not be
completed with these anions and the association constants
cannot be calculated. However, upon addition of HPO422
and H2PO24 into the solution of 5a, moderate upfield shifts
of the signal ArCH2NH2þCH2– at d 9.39 and 9.78 ppm were
observed. When the anion ratio increased the spectrum
changed to the same as that of the neutral compound (6a)
implying that the deprotonation of 5a took place upon
complexing HPO242 and H2PO24 . Association constants of
the complexes of 5a with HPO242 and H2PO42 thus cannot be
calculated.
NMR spectrum of 4a and at 8.07 and 7.68 ppm for 4b. FAB
MS and elemental analysis results were correct for the
proposed structures. Reduction of 4a and 4b by 20 equiv. of
NaBH4 in CH2Cl2 and subsequent protonation with
HCl/CH3OH (0.74% v/v) yielded ammonium derivatives,
5a (81%) and 5b (84%), which showed very broad signals in
the 1H NMR spectrum due to the effect of positive charges.
Signals due to ArCH2NHþ2 CH2– appeared at 9.78 and
9.39 ppm with an integral ratio of 2:1 for 5a and at 8.71 and
8.23 ppm for 5b. Mass spectra of 5a and 5b showed a strong
signal at m/z 1192.1 corresponding to the molecular weight
of the neutralized species. Nevertheless, elemental analysis
result agreed with the proposed structures. It is interesting
that the position of (CvO)H, RNvCH and ArCH2NH2þ-
CH2– protons in 3a–5a appear more downfield than those
of 3b–5b probably due to the effect of magnetic anisotropy
of the adjacent phenyl ring. Neutralization of 5a and 5b with
NaOH in methanol provided the neutral tripodal-amine
capped benzo crown calix[4]arene, 6a (72%) and 6b (74%).
Spectroscopic and elemental analysis results of 6a and 6b
agreed well with the proposed structures.
The complexation studies of 5b towards AsO22 , CO322 and
PO342 were also carried out. It was found that complexation
occurred along with the deprotonation to give the neutral
compound 6b. Therefore, the association constants for these
complexes cannot be calculated. Although addition of
Na2HPO4 into ligand 5b caused the displacement of the
aromatic signal (at 6.5–8.0 ppm) of 5b, Na2HPO4 did not
completely dissolve into solution. The association constant
of the complex of 5b and HPO242 cannot be determined
correctly. Furthermore, upon addition of excess NaH2PO4·
H2O to a solution of 5b, no chemical shift displacement in
the NMR spectra was observed. Therefore, compound 5b
did not form a complex with H2PO24 .
2.2. X-Ray crystallography
The solid state structure of compound 5a has been
determined by X-ray crystallography (Fig. 1). The structure
is solvated by one molecule of CH3OH and two molecules
of H2O. The phenyl rings of the calix[4]arene unit are in a
pinched cone conformation. Interestingly, one of the ethoxy
benzyl chains connecting to the tren unit threads through the
cavity of the other two ethoxy benzyl chains. This structure
resembles a self-threaded rotaxane derivatised from two
homooxacalix[3]arenes.15 Recently, Vicens and colleagues
have also reported a similar structure of tripodal calix[4]-
(azo)crowns.16 Although suitable crystals of 5b for X-ray
Upon addition of NaBr, NaI and NaNO3 to the solutions of
5a, moderate upfield shifts of the signal ArCH2NHþ2 CH2–
at d 9.39 and 9.78 ppm and slightly shifts of aromatic
regions at d 7.00–8.00 ppm were observed in the 1H NMR
spectra. This indicates that anions form complexes with
5a in the cavity of the tripodally capped unit using the
electrostatic interactions. Job plots indicated that 5a formed
complexes with these anions in a 1:1 fashion. In the case of
5b, the signal of the ArCH2NHþ2 CH2– protons in ligand 5b
1
analysis cannot be obtained, the H NMR spectrum of 5b
suggests a more symmetrical orientation of the glycolic
chains. The difference in structure and size of the cavity of
5a and 5b leads to the different in transition metal ions and
anion binding ability, vide infra.

10280
T. Tuntulani et al. / Tetrahedron 58 (2002) 10277–10285
Table 1. Summary of association constants of ligands 5a and 5b towards
various anions using Naþ as countercation
Table 2. Association constants of ligands 5a and 5b towards Br2 and I2 in
the presence of various countercations
Anion
K
assoc (M21
)
Metal
Anion
Kassoc (M21
)
5a
5b
5a
5b
Br2
I2
58.6
77.2
190.2
53.0
57.3
106.3
Nþonea
K
Br2
Br2
I2
84.2
120.1
108.9
103.3
76.5
34.9
137.9
66.3
NO23
Nþonea
K
I2
All experiments were carried out at 298 K; errors estimated to be less than
15%.
All experiments were carried out at 298 K; errors estimated to be less than
15%.
a
Using Bu4Nþ as countercation.
disappeared because the protons on the ammonium group
exchanged with CD3OD. Nevertheless, the moderate down-
field shift of the protons in the para position of –CH2ArHa
and ROArHb was monitored upon addition of various ratios
of NaBr, NaI and NaNO3. Job plots indicated that 5b
also formed complexes with anions in a 1:1 ratio. The
association constants of 5a and 5b towards various anions
calculated by the program EQNMR17 are shown in Table 1.
It is found that using Naþ as countercation 5a and 5b form
complexes with Br2, I2 and NO23 and the stability of the
complexes varies as NO32.I2.Br2. It should be noted that
the interaction of 5a with NO23 is stronger than that of 5b.
The results indicate that 5a possesses a more appropriate
cavity for NO23 than 5b does.
We are also interested in the effects of alkali metal ions
towards anion binding, especially spherical anions like Br2
and I2. Upon addition of tetrabutylammonium and
potassium salts of Br2 and I2 to 5a moderate upfield shifts
of the signal ArCH2NH2þCH2– at d 9.39 and 9.78 ppm and
to 5b strongly downfield shifts of the signal –CH2ArHa and
1
ROArHb at d 7.38 and 6.93 ppm were observed in the H
NMR spectra. The plot showing the relationship between
chemical shifts of the signal due to –OArHCH2– and
concentrations of iodide anion in the presence of various
countercations are depicted in Fig. 2. Association constants
of 5a and 5b towards Br2 and I2 in the presence of various
countercations such as Kþ and Bu4Nþ calculated by the
program EQNMR are collected in Table 2.
Recently, receptors containing two individual recognition
units, one for a cation and one for an anion have attracted
attention. Reinhoudt and coworkers have synthesized an
elegant calix[4]arene derivative with cation binding ester
groups on the lower rim and anion binding urea groups
on the upper rim. The compound was able to bind Cl2
efficiently in the presence of Naþ.18 Beer and co-workers
have synthesized a number of ditopic receptors that
can undergo selective ion pair recognition. Rhenium(I)
bipyridyl amide crown ether receptors were found to
complex KCl ion pairs.19 A tripodal tris(amido benzo-15-
crown-5) ligand was found to cooperatively bind chloride,
iodide and perrhenate anions via co-bound crown ether
complexed sodium cations.20 Later, a heteroditopic bis-
(calix[4]arene) ferrocene receptor was found to give a
significant electrochemical response to bromide anions in
the presence of Liþ ion.21
With Bu4Nþ as countercation, 5a and 5b can form more
stable complexes with I2. However, the stability of 5b
towards I2 is higher than 5a. This signifies that the cavity of
5b is more suitable for binding a big anion such as I2. In the
presence of Kþ, 5a shows an increase in binding affinity
towards Br2 by nearly 1.5-fold. On the other hand, Naþ
does not cause any enhancement in the anion binding ability
of 5a. This result suggests that the crown ether unit of 5a
prefers binding Kþ over Naþ. A similar crown ether cavity
found in bis(calix[4]arene) in which two molecules of
calix[4]arene linked by four glycolic units has been reported
to bind Kþ selectively.22 From the crystal structure of 5a,
it is also possible that an alkali metal ion can coordinate to
the crown ether unit and induces the structural reorgani-
zation of 5a to be more appropriate for binding anions
(Scheme 2). Interestingly, the binding ability of 5b towards
Br2 and I2 decreases in the presence of Naþ and Kþ. This
observation in which the presence of alkali metal ions
Figure 2. Titration curves of 5b with I2 in the presence of Bu4Nþ, Naþ and
Kþ.
Scheme 2.

T. Tuntulani et al. / Tetrahedron 58 (2002) 10277–10285
10281
Table 3. Stepwise protonation constants of ligands 6a and 6b in
1.00£1022 M Bu4NCF3SO3 in methanol at 258C
Table 4. Stability constants of complexes of 6a and 6b towards Co2þ, Ni2þ
Cu2þ and Zn2þ ions in 1.00£1022 M Bu4NCF3SO3 in methanol at 258C
,
log Ki
6a
6b
Ligand Species
log b
Co2þ
Ni2þ
Cu2þ
Zn2þ
log K1
log K2
log K3
log K4
12.03^0.01
10.17^0.01
6.95^0.02
4.44^0.03
10.79^0.04
9.35^0.04
7.46^0.04
5.33^0.04
6a
6b
ML2þ
7.52^0.02 6.97^0.03 17.75^0.06 9.97^0.06
M2L4þ
–
–
–
–
–
–
15.31^0.06
1.17^0.07
ML(OH)þ
ML2þ
M2L4þ
8.75^0.04 6.08^0.03 17.50^0.07 8.52^0.07
decreases the anion binding ability of 5a and 5b can be
rationalized in term of binding competition. Alkali metal
ions (Naþ or Kþ) that cannot fit into the cavity size of the
crown ether unit in 5a or 5b retain cation–anion pairs in
selection which can compete in binding with the tripodal
ammonium unit of the ligands.
–
–
–
–
–
14.28^0.05
–
ML(OH)þ 0.42^0.06
and 6b towards Co2þ, Ni2þ, Cu2þ and Zn2þ were carried out
by means of potentiometric titration using a competitive
method with the proton. The stability constants of the
complexes between ligand 6a and 6b with Co2þ, Ni2þ, Cu2þ
and Zn2þ ions in 1.00£1022 M Bu4NCF3SO3 in methanol
are summarized in Table 4.
2.4. Transition metal ion binding studies
In order to investigate the complexation ability of
compounds 6a and 6b towards transition metal ions,
compounds 5a and 5b were used in potentiometric
titrations. Protonation constants and complex stability
constants were calculated using the program SUPER-
QUAD.23 Protonation constants of 6a and 6b in
1.00£1022 M Bu4NCF3SO3 in methanol at 258C are
shown in Table 3.
The copper complexes of both 6a and 6b are the most stable
complexes. Ligand 6a shows slightly higher stability
constants than 6b with the difference of 0.25 log units.
The very high stability constants of compounds 6a and 6b
with Cu2þ are attributed to the suitable geometry of the
ligand to accommodate the stable geometry of Cu2þ ion.
When Cu2þ is coordinated to the nitrogen atoms in the tren
cavity, the tren unit can arrange itself for the trigonal
bipyramidal geometry, which is the most common geometry
found in many copper(II) complexes. Another ligand that
may occupy the axial position of the metal center is a
solvent molecule such as water or methanol in the presence
of non-coordinating counteranions (such as ClO24 or
CF3SO23 ). X-Ray crystal structures of various copper(II)
complexes of symmetric and unsymmetric tripodal tetra-
mine ligands which their structures are similar to the tren
unit also possess the trigonal bipyramidal geometry.25
Both compounds 6a and 6b can bind up to four protons in
the pH range 2–12 and show high basicity in the first two
protonation steps, with the first constants log K1¼12.02 and
10.79 for 6a and 6b, respectively. Therefore, 6a and 6b are
more basic than the constitutive subunit tren for which
log K1¼10.12.24 However, it is obvious that the first
protonation constant of 6a is higher than that of 6b. This
outcome may stem from the fact that compound 6a has
nitrogen substituents at the ortho position in which the
neighboring nitrogen atoms can form intramolecular
hydrogen bonding better than in the para-position of
compound 6b. This result leads to the more partial negative
charge on nitrogen atoms and more positive charge at
hydrogen atoms in 6a. Therefore, the nitrogen donors in 6a
have higher electron density and higher proton affinity than
those of 6b.
Interestingly, the results show that Zn2þ forms complexes
with ligands 6a and 6b in both mononuclear and binuclear
fashions. It should be noted that the binding affinity of the
second zinc ion for the 1:1 complex of 6b is slightly higher
than that of 6a. This result can also be rationalized in term of
the electronic repulsion. The distance between the tren unit
and glycolic ether chains in 6b is longer than that in 6a.
Therefore, when another zinc ion is included in the position
near the lower rim of calix[4]arene, the total electronic
repulsion is decreased and leads to greater stability to form
a binuclear complex in the case of 6b. Ligand 6a also
undergoes a hydrolysis reaction and affords ZnL(OH)þ,
which predominates in an alkaline solution.
Compound 6b exhibits slightly lower protonation ability
for the second proton than its ortho analogue. However, for
the third and fourth protonation, ligand 6b shows higher
protonation constants than 6a. These results can be
rationalized in term of the electrostatic repulsion. Although
both calix[4]arenes have a similar protonation trend in the
same pH range, 6b is less sterically hindered and the binding
sites are further apart from each other. Thus, the third and
fourth proton bindings of calix[4]arene 6b are not so
difficult and the positive charge repulsion is minimized
inducing a higher proton affinity.
Thus far, 6a possesses higher stability constants towards
Ni2þ, Cu2þ and Zn2þ than 6b does. However, the Co2þ
complex of 6b is obviously more stable than that of 6a (the
difference is about 1.25 logarithm units). This may stem
from the difference in size of the binding sites of the ligands.
Among transition metal ions, both 6a and 6b from the least
stable complexes with Ni2þ. This is probably due to the
unsuitable geometry of the ligands. Compounds 6a and 6b
are too constrained to adjust themselves to accommodate
Ni2þ in a square planar geometry (the most stable one
for Ni2þ). Therefore, the stability of the mononuclear
Compounds 6a and 6b are expected to bind first-row
divalent transition metal ions such as Co2þ, Ni2þ, Cu2þ and
Zn2þ to give stable complexes. Nevertheless, ligand 6a has
a different size of binding cavity in comparison with 6b.
Therefore, we aimed to investigate the effects of the size of
the binding cavity and flexibility of the ligand on the
stability of the complex. The binding ability studies of 6a

10282
T. Tuntulani et al. / Tetrahedron 58 (2002) 10277–10285
determined using VG-Analytical ZAB HF Mass Spectro-
meter. The ESI-TOF mass spectra were obtained from a
Micromass LCT Mass Spectrometer and the electrospray
ion trap mass spectra were recorded on a Bruker Mass
Spectrometer.
4.2. Synthesis
All materials and reagents were standard analytical grade,
purchased from BDH, Fluka, J.T. Baker or Merck, and used
without further purification. Commercial grade solvents
such as acetone, dichloromethane, ethyl acetate, hexane and
˚
methanol were distilled and stored over 4 A molecular
Scheme 3.
sieves. DMF was dried according to the published procedure
and distilled before use.27 Chromatographic separations
were performed on silica gel column (kieselgel 60, 0.063–
0.200 mm, Merck). Thin layer chromatography (TLC) was
carried out using silica gel plates (kieselgel 60 F254, 1 mm,
Merck). 2-(20-Bromoethoxy)benzaldehyde, 1a,13 4-(20-bro-
moethoxy)benzaldehyde, 1b,7 and p-tert-butylcalix[4]arene28
were prepared according to the literature.
complexes of the two ligands follows the sequence: Ni2þ
,
Co2þ!Cu2þ@Zn2þ. This result is in the same trend with the
Irving–Williams series of the metal complex stability.26
3. Conclusion
In summary, we have synthesized two tripodal aza crown
ether calix[4]arenes, 6a and 6b, and studied anion
complexation with various anions using the acid forms, 5a
and 5b. Both 5a and 5b cannot form complexes with F2
4.2.1. Preparation of 25,26,27-tri((2-ethoxy)benzalde-
hyde-p-tert-butylcalix[4]arene, 3a and 25,26,27-tri((4-
ethoxy)benzaldehyde-p-tert-butylcalix[4]arene, 3b. In a
250 mL two-necked round bottom flask equipped with a
magnetic bar and a reflux condenser, a mixture of p-tert-
butylcalix[4]arene (6.05 g, 9.34 mmol), barium oxide (5.20 g,
33.90 mmol) and dry DMF (150 mL) was stirred for 1 h.
Into this mixture, 2-(20-bromoethoxy)benzaldehyde, 1a,
(6.56 g, 28.64 mmol) in DMF (50 mL) was then added
dropwise through an addition funnel. The mixture was
stirred and heated at 708C under nitrogen atmosphere for 7
days. The reaction was allowed to cool to room temperature,
and the solvent was evaporated under reduced pressure to
give an orange-brown residue. The residue was dissolved in
dichloromethane and it was then added 3 M hydrochloric
acid until the pH of the solution reached 1. The organic
phase was separated, and the aqueous layer was extracted
again with dichloromethane. The combined organic layer
was dried over sodium sulfate anhydrous. After filtration of
sodium sulfate, the solvent was removed to give an oily
orange-brown residue. The residue was redissolved in a
minimum amount of dichloromethane. The orange-brown
solution was eluted through a silica gel column with
dichloromethane as eluent. 25,26,27-Tri((2-ethoxy)benzal-
dehyde-p-tert-butylcalix[4]arene, 3a, eluted out of the
column after 25,27-di((2-ethoxy)benzaldehyde-p-tert-butyl-
calix[4]arene, 2a. White needle crystals of 3a can be
obtained by adding CH3OH into its (CH2Cl2) solution
(2.17 g, 21%).
and SO242. In the case of basic anionic guests such as CO232
,
PO342, AsO22 , HPO422 and H2PO24 , two phenomena,
complexation and deprotonation, have occurred con-
currently. This is the most crucial defect of anion hosts
using ammonium receptors. However, the main advantage
of using the electrostatic interaction for anions is that polar
and protic solvents can be used, unlike the hydrogen
bonding interaction in which the type of solvents employed
is limited. For spherical anions like Br2 and I2, it is found
that the presence of Kþ enhances the complexation of 5a
towards anions. Our results thus show the effect of metal
ions on anion binding and also demonstrate an example of
metal ion-controlled anion receptors. In the mean time, we
have shown that compounds 6a and 6b can bind transition
metal ions such as Co2þ, Ni2þ and Cu2þ in a 1:1 fashion.
However, the ligands form the most stable complexes with
Cu2þ ion. Strikingly, compounds 6a and 6b form complexes
with Zn2þ in both 1:1 and 1:2 ligand to metal ratios.
Compounds 6a and 6b may thus be used potentially as
switchable receptors for metal ions and anions. The binding
ability of the receptor can be switched by varying the pH of
the solution as illustrated in Scheme 3.
4. Experimental
4.1. Analytical instruments
In a similar manner to 3a, the reaction between p-tert-
butylcalix[4]arene (7.03 g, 10.83 mmol) and 4-2(20-bromo-
ethoxy)benzaldehyde (7.42 g, 32.63 mmol) in dry DMF
(50 mL) resulted in compounds 2b (0.20 g, 2%) and 3b
(5.53 g, 46%) as white solids.
The 1H NMR spectra were recorded at 200 MHz on a
Bruker ACF 200 nuclear magnetic resonance spectrometer,
400 MHz on a Bruker DRX 400 spectrometer and 500 MHz
on a GEOL spectrometer. In all cases, samples were
dissolved in DMSO-d6 or in mixed CDCl3 and CD3OD, and
chemical shifts were recorded using a residual proton signal
as internal reference.
3a: Mp: 239–2408C. 1H NMR spectrum (200 MHz, CDCl3)
d (ppm) 10.41 and 9.74 (s each, 2H and 1H, –Ar(CvO)H);
7.63–6.32 (m, 20H, aromatic protons); 5.22 (s, 1H, ArOH);
4.90, 4.42 and 4.16 (m, 12H, –OCH2CH2O–); 4.24
and 3.29 (m, 4H each, ArCHAHBAr); 1.36–0.82 (m, 36H,
Elemental analyses were analyzed on a Perkin–Elmer
CHON/S analyzer (PE2400 series II). Mass spectra were

T. Tuntulani et al. / Tetrahedron 58 (2002) 10277–10285
10283
–Ar-t-C4H9). FAB MS (m/z): 1092.5. Anal. calcd for 4a
(C71H80O10): C, 77.99; H, 7.37. Found: C, 78.11; H, 7.17.
4.2.3. Preparation of 25,26,27-N,N0,N00-tri((2-ethoxy)ben-
zyl)ethylenetetraamine-p-tert-butylcalix[4]arene·4HCl,
5a and 25,26,27-N,N0,N00-tri((4-ethoxy)benzyl) ethylene-
3b: Mp: 199–2008C. 1H NMR spectrum (500 MHz, CDCl3)
d (ppm) 9.76 and 9.68 (s each, 2H and 1H, –Ar(CvO)H),
7.57 and 7.43 (d each, JH–H¼8.7 Hz, 4H and 2H, –OArHa),
7.19 (s, 2H, HOArH), 7.14 (s, 2H, ROArH), 6.70 and 6.63 (d
each, JH–H¼8.7 Hz, 4H and 2H, –OArHb), 6.54 (s, 4H,
ROArH), 5.40 (s, 1H, HOAr), 4.86 (m, 2H, OCH2CH2O),
4.45 and 3.32 (d each, JH – H¼12.4 Hz, 4H each,
ArCHAHBAr), 4.28 (m, 2H, OCH2CH2O), 4.13 (s, 8H,
OCH2CH2O), 1.36 (s, 18H, HOAr-t-C4H9 and ROAr-t-
C4H9), 0.82 (s, 18H, ROAr-t-C4H9). MALDI-TOF MS
(m/z): 1093.6. Anal. calcd for 3b (C71H80O10): C, 77.99; H,
7.37. Found: C, 77.91; H, 7.52.
tetraamine-p-tert-butylcalix[4]arene·4HCl, 5b. In
a
500 mL one-necked round bottom flask equipped with a
magnetic bar and a reflux condenser, 4a (1.00 g, 0.84 mmol)
was dissolved in dry dichloromethane (50 mL). To the
solution was added excess sodium borohydride (0.63 g,
0.02 mmol) and the mixture was stirred overnight under
nitrogen atmosphere. A copious amount of deionized water
was then added to destroy excess sodium borohydride.
The organic phase was separated and washed again with
deionized water until the pH of the aqueous layer became
neutral. The combined organic layer was dried over
anhydrous sodium sulfate. After filtration of sodium sulfate,
the solvent was evaporated to dryness. The solid residue was
dissolved in a minimum amount of methanol and acidified
with 0.74% v/v hydrochloric acid in methanol until the pH
of the solution reached 1. Upon slow evaporation of the
solvent, white crystals of 5a were precipitated (0.92 g,
81%).
4.2.2. Preparation of 25,26,27-N,N0,N00-tri-((2-ethoxy)-
benzyl)ethylenetriimine-p-tert-butylcalix[4]arene, 4a,
and 25,26,27-N,N0,N00-tri-((4-ethoxy)benzyl)ethylenetri-
imine-p-tert-butylcalix[4]arene, 4b. In a 500 mL two-
necked round bottom flask equipped with a magnetic bar
and a reflux condenser, a mixture of 3a, (1.00 g, 0.92 mmol)
and acetonitrile (250 mL) was stirred. Tris(2-amino)ethyl-
amine (0.16 g, 1.10 mmol) in dichloromethane (10 mL) and
acetronitrile (50 mL) was then added dropwise through an
addition funnel in 30 min. The mixture was refluxed under
nitrogen atmosphere for 8 h. White solid precipitated from
the solution. The mixture was allowed to cool to room
temperature and filtered. The white solid 4a was washed
with acetronitrile and dried in vacuo (1.03 g, 95%).
In a similar fashion, the reaction between 4b (1.52 g,
1.283 mmol) and NaBH4 (0.92 g, 24.35 mmol) in dry
CH2Cl2 (300 mL) yielded 5b as a white solid (1.44 g, 84%).
5a: Mp: 290–2928C (dec.). 1H NMR spectrum (DMSO-d6)
d (ppm) 9.78 and 9.38 (s each, broad, 4H and 2H,
ArCH2NHþ2 Cl2); 7.86, 7.66, 7.57, 7.34 and 7.03 (m, 12H,
Ha, Hb, Hc and Hd); 7.17 and 7.11 (s each, 2H each, ROArH
and HOArH); 6.54 and 6.46 (s each, 2H each, ROArH);
5.80 (s, 1H, ArOH); 5.13 (m, broad, 2H, OCH2CH2O);
4.62–4.39 (m, 6H, H2NþCH2–Ar and 4H, ArCH2Ar); 4.18
(m, broad, 10H, OCH2CH2O and 4H, ArCH2Ar); 2.82–2.75
(m, 12H, þNHCH2CH2NþH2); 1.30, 1.20 and 0.73 (s each,
9H, 9H and 18H, HOAr-t-C4H9 and ROAr-t-C4H9).
ESI-TOF MS (m/z): 1192.1. Anal. calcd for 5a·4H2O
(C77H110N4O11Cl4): C, 65.69; H, 8.44; N, 3.78. Found: C,
65.61; H, 7.87; N, 3.97.
The reaction between 3b (3.05 g, 2.79 mmol) and tris-
(2-aminoethyl)amine (0.50 g, 3.43 mol) in acetonitrile
(250 mL) gave compound 4b as a white solid (3.23 g, 97%).
1
4a: Mp: 310–3128C (dec.). H NMR spectrum (200 MHz,
CDCl3) d (ppm) 8.93 and 8.83 (s each, 1H and 2H,
–CHvN–); 7.91–6.45 (m, 20H, aromatic protons); 5.30 (s,
1H, –ArOH); 5.16, 4.53 and 4.04 (m, 12H, –OCH2CH2O);
2.89 (m, 12H, –NCH2CH2N–); 4.39 and 4.33, 3.39 and
3.32 (d each, 2H each, JH–H¼13 Hz, ArCHAHBAr); 1.36,
1.27 and 0.79 (s each, 9H, 9H and 18H, ROAr-t-C4H9 and
HOAr-t-C4H9). FAB MS (m/z): 1185.7. Anal. calcd for
4a·H2O (C77H94N4O8): C, 76.84; H, 7.87; N, 4.65. Found:
C, 76.70; H, 7.61; N, 4.24.
1
5b: Mp: 283–2858C (dec.). H NMR spectrum (500 MHz,
CDCl3) d 8.71 and 8.23 (s each, broad, 4H and 2H,
ArCH2NHþ2 Cl2), 7.79 (d, JH–H¼8.6 Hz, 4H, –OArHa),
7.36 (d, JH–H¼8.5 Hz, 2H, –OArHa), 7.14 (s, 2H, HOArH),
7.10 (s, 2H, ROArH), 6.92 (d, JH–H¼8.7 Hz, 4H, –OArHb),
6.54 (m, 6H, ROArH and –OArHb), 6.12 (s, 1H, HOAr),
4.56 and 3.30 (d each, JH–H¼13.3 Hz, 4H, ArCHAHBAr),
4.55–4.40 (m, 14H, OCH2CH2O, ArCH2N and ArCHAHB-
Ar), 4.20–3.98 (m, 6H, OCH2CH2O, ArCH2N), 3.70 (s, br,
2H, NCH2CH2N), 3.41–3.10 (m, 10H, NCH2CH2N), 3.25
(d, JH–H¼13.0 Hz, 2H, ArCHAHBAr), 1.34 (s, 9H, HOAr-t-
C4H9), 1.32 (s, 9H, ROAr-t-C4H9), 0.82 (s, 18H, ROAr-t-
C4H9). MALDI-TOF MS (m/z): 1191.8. Anal. calcd for 5b
(C77H102N4O7Cl4): C, 69.15; H, 7.69; N, 4.19. Found: C,
69.19; H, 7.76; N, 4.16.
1
4b: Mp: 305–3078C (dec.). H NMR spectrum (500 MHz,
CDCl3) d (ppm) 8.07 and 7.86 (s each, 1H and 2H,
–CHvN), 7.38 (d, JH–H¼8.7 Hz, 4H, –OArHa), 7.20 (s,
2H, HOArH), 7.18 (s, 2H, ROArH), 6.73 (d, JH–H¼8.7 Hz,
4H, –OArHb), 6.62 (d, JH–H¼2.4 Hz, 2H, ROArHa), 6.52
(d, JH–H¼2.4 Hz, 2H, ROArHb), 6.32 (s, 1H, HOAr), 6.13
(d, JH–H¼8.8 Hz, 2H, ROArH), 6.02 (d, JH–H¼8.8 Hz,
2H, ROArH), 4.92 and 3.32 (d each, JH–H¼13.0 Hz, 4H,
ArCHAHBAr), 4.56 (m, 2H, OCH2CH2O), 4.33 and 3.23
(d each, JH–H¼13.0 Hz, 4H, ArCHAHBAr), 4.28–4.02
(m, 10H, OCH2CH2O), 3.74 (m, 4H, CHvNCH2CH2N),
3.64 (m, 2H, CHvN CH2CH2N), 2.83 (m, 4H,
CHvNCH2CH2N), 2.59 (m, 2H, CHvNCH2 CH2N), 1.39
(s, 9H, HOAr-t-C4H9), 1.36 (s, 9H, ROAr-t-C4H9), 0.83 (s,
18H, ROAr-t-C4H9). MALDI-TOF MS (m/z): 1184.6. Anal.
calcd for 4b (C77H92N4O7): C, 78.01; H, 7.82; N, 4.73.
Found: C, 77.95; H, 7.66; N, 4.77.
4.2.4. Preparation of 25,26,27-N,N0,N00-tri((2-ethoxy)ben-
zyl)ethylenetetraamine-p-tert-butylcalix[4]arene, 6a and
25,26,27-N,N0,N00-tri((4-ethoxy)benzyl)ethylenetetra-
amine-p-tert-butylcalix[4]arene, 6b. In a 50 mL round
bottom00 flask equipped with a magnetic bar, 25,26,27-
N,N0,N -tri-((2-ethoxy)benzylethylenetetraamine-p-tert-
butylcalix[4]arene·4HCl, 5a, (0.10 g, 0.07 mmol) was

10284
T. Tuntulani et al. / Tetrahedron 58 (2002) 10277–10285
dissolved in dry methanol (30 mL). NaOH solution
(CH3OH) was then slowly added until the pH of the
solution reached 10. The reaction was stirred under nitrogen
atmosphere for 1 h. The solvent was subsequently removed
under reduced pressure. The residue was redissolved in
dichloromethane and extracted with deionized water until
the aqueous phase contained no Cl2. The organic layer was
then dried over anhydrous sodium sulfate and concentrated
on a rotary evaporator. Upon slow evaporation of the
solvent, the white solid of 6a precipitated (0.064 g, 72 %).
dihydrogen phosphate, iodide, nitrate, sulfate and phosphate
was studied employing 1H NMR titrations. For 5a, typically,
a 0.0250 M solution of ligand 5a (0.0836 g, 0.0625 mmol)
in DMSO-d6 (2.50 mL) was prepared. To 0.20 mL of this
solution in NMR tubes were added 0.0–4.0 equiv. of
0.1000 M sodium salts (0.1500 mmol) in DMSO-d6
(1.50 mL). In each NMR tube, the amount of DMSO-d6
was then adjusted to the same quantity. For 5b, typically, a
0.1000 M solution of a sodium salt (0.1500 mmol) in
CD3OD (1.50 mL) was prepared. Ligand 5b was brought
into the NMR tubes and 0.0–4.0 equiv. of 0.1000 M sodium
salt were added. In each NMR tube, the amount of the
solvents was adjusted to the same quantity. The spectra were
recorded every 24 h until the complexation reached the
equilibrium. Plots between complex concentration (12X)d
and the mole fraction of ligands and anions (X) indicated
1:1 ligand:anion complexes for Br2, I2 and NO32. The result
of the experiment was a plot of displacement in chemical
shift as a function of the amount of added anion, which was
subjected to analysis by a non-linear curve-fitting method
using the program EQNMR.17
A similar reaction between 5b (1.69 g, 1.262 mmol) and a
solution of NaOH (0.20 g, 5.048 mmol) in methanol
(40 mL) yielded 6b as a white solid (1.12 g, 74%).
1
6a: Mp: 300–3028C (dec.). H NMR spectrum (400 MHz,
CDCl3) d (ppm) 7.20 and 6.56 (s each, 4H, t-C(CH3)-3-
ArHCH2 –); 7.47, 7.15, 6.93 and 6.25 (m, 12H,
–OArHOCH2–); 5.19 (s, 1H, –ArOH); 4.98, 4.32, 4.03
and 3.86 (m, 12H, –OCH2CH2O–); 4.18 and 3.75 (d, 2H
and 4H, JH–H¼14 Hz, ArCH2NR); 7.34 (t, 3H, Hc aro-
matic); 4.85, 4.45, 3.37 and 3.23 (d, 2H each, JH–H¼13 Hz,
ArCHACHBAr); 2.66–1.99 (m, broad, 12H, RNCH2CH2-
NR); 1.41, 1.39 and 0.86 (s each, 9H, 9H and 18H, HOAr-t-
C4H9 and ROAr-t-C4H9). FAB MS (m/z): 1191.7. Anal.
calcd for 6a (C77H98N4O7): C, 77.61; H, 8.29; N, 4.70.
Found: C, 77.57; H, 7.85; N, 4.32.
4.5. Potentiometric titration
All materials and solvents were standard analytical grade
purchased from Aldrich, Fluka and Merck, and used without
further purification unless otherwise noted. Transition metal
salts such as Co(ClO4)2·6H2O, Ni(CH3CO2)2·4H2O,
Cu(CF3SO3)2 and Zn(CF3SO3)2 were dried under reduced
pressure and stored in a desiccator. Compounds 5a and 5b
were dried under vacuum and kept in a dessicator before
use.
1
6b: Mp: 295–2988C (dec.). H NMR spectrum (200 MHz,
CDCl3) d (ppm) 7.18, 7.07, 7.05 and 7.01 (s each, 8H,
t-C(CH3)3ArHCH2–); 6.70–6.30 (m, 12H, –OArHOCH2–
); 5.39 (s, 1H, –ArOH); 4.85–4.60 (m, 6H, –OCH2CH2O–,
ArCHAHBAr); 3.27 (m, 2H, ArCHAHBAr); 3.74–3.45 (m,
6H, –OArCH2N–); 2.86, 2.67, 1.92 (m, broad, 12H,
RNCH2CH2NR); 1.32 and 0.83 (s each, 18H and 18H,
HOAr-t-C4H9 and ROAr-t-C4H9). FAB MS (m/z): 1191.7.
Anal. calcd for 6b (C77H98N4O7): C, 77.61; H, 8.29; N,
4.70. Found: C, 77.58; H, 8.22; N, 4.66.
Tetrabutylammonium trifluoromethanesulfonate (Bu4-
NCF3SO3) was used as an inert background electrolyte for
all experiments. This electrolyte was prepared by dissolving
a weighed quantity of Bu4NCF3SO3 in methanol. The
concentration of the inert background electrolyte was kept
constant at 0.01 M. Tetrabutylammonium hydroxide
(Bu4NOH), of concentration 0.05 M, was utilized as a
titrant base. Standard solutions of HCl (0.01 M) and HClO4
(0.05 M) in the inert background solution were used to
adjust the pH during electrode calibrations33 and titration
experiments. Solutions of metal ions Co2þ, Ni2þ, Cu2þ and
Zn2þ were prepared by dissolution of weighed quantities of
metal salts in the inert background solution. The concen-
tration of each metal solution was approximately
0.01 M. Stock solutions of 5a (0.001 M) and 5b (0.001 M)
were prepared by dissolving weighted quantities of
compounds 5a and 5b in 0.01 M Bu4NCF3SO3.
4.3. X-Ray crystallography
The crystal of 5a (0.20£0.20£0.10 mm3) was mounted on
the end of a hollow glass fiber approximately parallel to the
long dimension of the crystal using cyanoacrylate glue.
Preliminary examination and data collection were per-
˚
formed using MoKa X-radiation (l¼0.71073 A) on Bruker
AXS SMART area detector diffractrometer. The collected
data were reduced using the program SAINT.29 Empirical
absorption correction was done by the program SADABS.30
A total of 58283 reflections were measured within the u
range of 0.99–30.468. The structure was solved by direct
methods and refined with anisotropic thermal parameters for
all non-hydrogen atoms by full matrix least square using
SHELX-97 package.31 All hydrogen atoms were found in
different Fourier maps and were included in the refinement.
Due to vibrational disorder of the solvent of crystallization,
refinement converged with rather high R and wR values
(0.1355 and 0.3402, respectively).32
The stability constants of the complexes were obtained
potentiometrically using a competitive method with proton.
The concentrations of [Hþ] were measured by a combined
glass electrode (Mettler DG113-SC) connected to an
automatic titrator (Mettler DL25). All titration experiments
were performed at 258C. For complexation constants
determination, typically a 10 mL portion of a stock solution
of a ligand was added a metal solution under argon
atmosphere and subsequently titrated with Bu4NOH. For
each determination at least four experiments were per-
formed with different mole ratios of metal:ligand. The
stability constants were calculated using the program
4.4. 1H NMR titrations
Complexation of 5a and 5b towards various anions such as
arsenite, bromide, carbonate, fluoride, hydrogen phosphate,

T. Tuntulani et al. / Tetrahedron 58 (2002) 10277–10285
10285
SUPERQUAD.23 The protonation constants of the free
ligands, determined from titrations of ligands in the absence
of metal, and metal hydroxide species, MOHþ, were
included in the refinement procedures.
N.; Ruangpornvisuti, V.; Chaichit, N.; Asfari, Z.; Vicens, J.
Tetrahedron Lett. 2001, 42, 5541–5544.
12. Tuntulani, T.; Ruangpornvisuti, V.; Tantikunwattana, N.;
Ngampaiboonsombut, O.; Seangprasertkij-Magee, R.; Asfari,
Z.; Vicens, J. Tetrahedron Lett. 1997, 38, 3985–3988.
13. Seangprasertkij, R.; Asfari, Z.; Arnaud, F.; Vicens, J. J. Org.
Chem. 1994, 59, 1741–1744.
Supplementary material
14. Sukwattanasinitt, M.; Rojanathanes, R.; Tuntulani, T.;
Ruangpornvisuti, V. Sci. Asia 2002, 28, 25–28.
15. Zhong, Z.; Ikeda, A.; Shinkai, S. J. Am. Chem. Soc. 1999, 121,
11906–11907.
All potentiometric titration plots of 6a and 6b towards
Co2þ, Cu2þ and Zn2þ are deposited in the editorial office.
Crystallographic data for 5a are available upon request from
the Cambridge Crystallographic Data Base (CCDC
165241).
´
16. Oueslati, I.; Abidi, R.; Amri, H.; Thuery, P.; Nierlich, M.;
Asfari, Z.; Vicens, J. Tetrahedron Lett. 2001, 42, 1685–1689.
17. Hynes, M. J. J. Chem. Soc., Dalton Trans. 1993, 311–312.
18. Scheerder, J.; van Duynhoven, J. P. M.; Engbersen, J. F. J.;
Reinhoudt, D. N. Angew. Chem., Int. Ed. Engl. 1996, 35,
1090–1093.
Acknowledgements
This work was financially supported by the Thailand
Research Fund (Grant no. PDF4080055). The authors
thank Professor Michael J. Hynes for providing the program
EQNMR. We also appreciate the National Biotechnology
Center for allowing us to use the 400 MHz NMR
spectrometer.
19. Redman, J. E.; Beer, P. D.; Dent, S. W.; Drew, M. G. B. Chem.
Commun. 1998, 231–232.
20. Beer, P. D.; Hopkins, P. K.; McKinney, J. D. Chem. Commun.
1999, 1253–1254.
21. Cooper, J. B.; Drew, M. G. B.; Beer, P. D. J. Chem. Soc.,
Dalton Trans. 2000, 2721–2728.
22. Schmitt, P.; Beer, P. D.; Drew, M. G. B.; Sheen, P. D. Angew.
Chem., Int. Ed. Engl. 1997, 36, 1840–1842.
23. Gans, P.; Sabatini, A.; Vacca, A. J. Chem. Soc., Dalton Trans.
1985, 1195–1200.
References
24. Motekaitis, R. J.; Martell, A. E.; Lehn, J. M.; Watanabe, E. I.
Inorg. Chem. 1982, 21, 4253–4257.
1. Steed, J. W.; Atwood, J. L. Supramolecular Chemistry. Wiley:
New York, 2000.
25. Dittler-Klingemann, A. M.; Hahn, F. E. Inorg. Chem. 1996,
35, 1996–1999.
2. Dietrich, B.; Dilworth, B.; Lehn, J.-M.; Souchez, J. P.;
Cesario, M.; Guilhem, J.; Pascard, C. Helv. Chim. Acta 1996,
79, 569–587.
26. Shriver, D. F.; Atkins, P. W.; Langford, C. H. Inorganic
Chemistry. 2nd ed; Oxford University: Oxford, 1994.
27. Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals; Pergamon: Oxford, 1988; p 68.
3. Mason, S.; Clifford, T.; Seib, L.; Kuczera, K.; Bowman-James,
K. J. Am. Chem. Soc. 1998, 120, 8899–8900.
4. Mason, S.; Llinares, J. M.; Morton, M.; Clifford, T.; Bowman-
James, K. J. Am. Chem. Soc. 2000, 122, 1814–1815.
5. Clifford, T.; Danby, A.; Llinares, J. M.; Mason, S.; Alcock,
28. Gutsche, C. D.; Iqbal, M. Org. Synth. 1990, 68, 234–237.
29. SAINT: Data Integration Software, Version 4.0; Bruker AXS,
Inc.: Madison, USA, 1997.
´
N. W.; Powell, D.; Aguilar, J. A.; Garcıa-Espan˜a, E.;
30. Sheldrick, G. M. SADABS: Program for Empirical Absorption
Bowman-James, K. Inorg. Chem. 2001, 40, 4710–4720.
6. Arnaud-Neu, F.; Fuangswasdi, S.; Maubert, B.; Nelson, J.;
Mckee, V. Inorg. Chem. 2000, 39, 573–579.
¨
Correction of Area Detector Data; University of Gotingen:
¨
Gotingen, Germany, 1996.
31. Sheldrick, G. M. SHELXL97; University of Gotingen:
¨
7. Navakun, K.; Tuntulani, T.; Ruangpornvisuti, V. J. Incl.
Phenom. 2000, 38, 113–122.
¨
Gotingen, Germany, 1997.
32. Crystal data for 5a, C77H102N4O7Cl2(OH)2·(CH3OH)(H2O)2,
M¼1368.7, monoclinic, space group C2/c, Z¼8, a¼
8. Rojsajjakul, T.; Veravong, S.; Tumcharern, G.;
Seangprasertkij-Magee, R.; Tuntulani, T. Tetrahedron 1997,
53, 4669–4680.
˚
43.6552(14), b¼15.9085(5), c¼25.1856(7) A, b¼
3
109.4630(10)8, V¼16491.6(9) A , Dc¼1.119 g cm23, 23606
˚
9. Tuntulani, T.; Tumcharern, G.; Ruangpornvisuti, V. J. Incl.
Phenom. 2001, 39, 47–53.
¨
10. Asfari, Z.; Bohmer, V.; Harrowfield, J.; Vicens, J. Calixarenes
unique data, R1¼0.1355, wR2¼0.3402.
33. Arnaud-Neu, F.; Barrett, G.; Harris, S. J.; Owens, M.;
McKervey, M. A.; Schwing-Weill, M.-J.; Schwinte, P.
Inorg. Chem. 1993, 32, 2644–2650.
2001. Kluwer Academic: Dordrecht, 2001.
11. Tuntulani, T.; Poompradub, S.; Thavornyutikarn, P.; Jaiboon,