An integrated approach (thermodynamic, structural, and computational) to the study of complexation of alkali-metal cations by a lower-rim calix[4]arene amide derivative in acetonitrile

Article abstract of DOI:10.1021/ic300474s

The calix[4]arene secondary-amide derivative L was synthesized, and its complexation with alkali-metal cations in acetonitrile (MeCN) was studied by means of spectrophotometric, NMR, conductometric, and microcalorimetric titrations at 25 °C. The stability constants of the 1:1 (metal/ligand) complexes determined by different methods were in excellent agreement. For the complexation of M⁺ (M = Li, Na, K) with L, both enthalpic and entropic contributions were favorable, with their values and mutual relations being quite strongly dependent on the cation. The enthalpic and overall stability was the largest in the case of the sodium complex. Molecular and crystal structures of free L, its methanol and MeCN solvates, the sodium complex, and its MeCN solvate were determined by single-crystal X-ray diffraction. The inclusion of a MeCN molecule in the calixarene hydrophobic cavity was observed both in solution and in the solid state. This specific interaction was found to be stronger in the case of metal complexes compared to the free ligand because of the better preorganization of the hydrophobic cone to accept the solvent molecule. Density functional theory calculations showed that the flattened cone conformation (C₂ point group) of L was generally more favorable than the square cone conformation (C₄ point group). In the complex with Na⁺, L was in square cone conformation, whereas in its adduct with MeCN, the conformation was slightly distorted from the full symmetry. These conformations were in agreement with those observed in the solid state. The classical molecular dynamics simulations indicated that the MeCN molecule enters the L hydrophobic cavity of both the free ligand and its alkali-metal complexes. The inclusion of MeCN in the cone of free L was accompanied by the conformational change from C₂ to C₄ symmetry. As in solution studies, in the case of ML⁺ complexes, an allosteric effect was observed: the ligand was already in the appropriate square cone conformation to bind the solvent molecule, allowing it to more easily and faster enter the calixarene cavity.

Full text of DOI:10.1021/ic300474s

Article
pubs.acs.org/IC
An Integrated Approach (Thermodynamic, Structural, and
Computational) to the Study of Complexation of Alkali-Metal Cations
by a Lower-Rim Calix[4]arene Amide Derivative in Acetonitrile
Gordan Horvat,† Vladimir Stilinovic,† Tomica Hrenar,† Branko Kaitner,† Leo Frkanec,‡
́
and Vladislav Tomisic*,†
̌
́
†Department of Chemistry, Faculty of Science, University of Zagreb, Horvatovac 102a, 10000 Zagreb, Croatia
‡Laboratory of Supramolecular and Nucleoside Chemistry, Department of Organic Chemistry and Biochemistry, Ruđer Bosk
Institute, Bijenicka c. 54, 10000 Zagreb, Croatia
̌
ovic
́
̌
S
* Supporting Information
ABSTRACT: The calix[4]arene secondary-amide derivative L was synthesized, and
its complexation with alkali-metal cations in acetonitrile (MeCN) was studied by
means of spectrophotometric, NMR, conductometric, and microcalorimetric titrations
at 25 °C. The stability constants of the 1:1 (metal/ligand) complexes determined by
different methods were in excellent agreement. For the complexation of M+ (M = Li,
Na, K) with L, both enthalpic and entropic contributions were favorable, with their
values and mutual relations being quite strongly dependent on the cation. The
enthalpic and overall stability was the largest in the case of the sodium complex.
Molecular and crystal structures of free L, its methanol and MeCN solvates, the
sodium complex, and its MeCN solvate were determined by single-crystal X-ray
diffraction. The inclusion of a MeCN molecule in the calixarene hydrophobic cavity
was observed both in solution and in the solid state. This specific interaction was
found to be stronger in the case of metal complexes compared to the free ligand
because of the better preorganization of the hydrophobic cone to accept the solvent molecule. Density functional theory
calculations showed that the flattened cone conformation (C2 point group) of L was generally more favorable than the square cone
conformation (C4 point group). In the complex with Na+, L was in square cone conformation, whereas in its adduct with MeCN,
the conformation was slightly distorted from the full symmetry. These conformations were in agreement with those observed in
the solid state. The classical molecular dynamics simulations indicated that the MeCN molecule enters the L hydrophobic cavity
of both the free ligand and its alkali-metal complexes. The inclusion of MeCN in the cone of free L was accompanied by the
conformational change from C2 to C4 symmetry. As in solution studies, in the case of ML+ complexes, an allosteric effect was
observed: the ligand was already in the appropriate square cone conformation to bind the solvent molecule, allowing it to more
easily and faster enter the calixarene cavity.
nature of the binding groups, the affinity of calixarenes toward
metal cations strongly depends on the reaction medium, i.e., on
the solvent used.1d,5i,k,6 In this respect, the specific interactions
of the ligand and the complex with the solvent molecules can
be particularly important. Inclusion of an acetonitrile (MeCN)
molecule in a calixarene hydrophobic cavity4a−f,7 can serve as a
striking example of such an interaction, which was shown to be
synergetic with the binding of the cation in the ligand
INTRODUCTION
■
Calixarene derivatives have been extensively studied because of
their ability to form inclusion complexes with a wide variety of
guest species.1 These macrocyclic receptors have been prepared
by lower- and/or upper-rim functionalization of the parent
calixarenes. By the appropriate choice of the substituents and
the number of repeating phenolic units comprising the
macrocycle, the efficient and, in some instances, selective
binders of cations,2 anions,2b,3 and neutral molecules2b,4 can be
obtained.
The p-tert-butyl lower-rim-functionalized calix[4]arenes with
carbonyl-containing substituents, which include calixarene
ketones, esters, and amides, have been proven to be effective
receptors for alkali, alkaline-earth, and transition-metal
cations.2c,d,f An interesting feature of tetrasubstituted secon-
dary-amide derivatives is the possibility of intramolecular
NH···OC hydrogen-bond formation. This was shown to
have a strong influence on their binding abilities.5 Besides the
hydrophilic cavity.5k,7a,8
Even though the complexation reaction between the
calixarene ligand (L) and the metal cation (Mz(M); z(M)
denotes the charge number) in solution (sln)
Mz(M)(sln) + L(sln) ⇄ MLz(M)(sln)
(1)
Received: March 2, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
g, 2.24 mmol) and triethylamine (0.36 mL, 2.61 mmol) were added to
the cooled (0 °C) dry CH2Cl2 (20 mL) under argon. Then, the acid
chloride of calix[4]arene tetracetic acid (the cone conformer)10 (0.45 g,
0.47 mmol) in dry CH2Cl2 (10 mL) was added at once with a syringe.
The reaction mixture was allowed to warm to room temperature and
stirred overnight. After filtration, the filtrate was washed with NaOH
(1.0 mol dm−3), hydrochloric acid (1.0 mol dm−3), brine, and water.
After drying (MgSO4) and evaporation of the solvent, the crude
product was crystallized from CH2Cl2/petroleum ether. Recrystalliza-
tion from methanol (MeOH) gave L (0.49 g, 71%) as white crystals.
Mp: 233−234 °C. Rf = 0.44 (5% MeOH/CH2Cl2).
seems rather simple, there are many processes and interactions
that are involved in such a reaction and that determine its
equilibrium. For that reason, in order to gain detailed insight
into the reactions studied, it is advisible to employ a number of
complementary experimental and computational techniques, so
that one may support the other. In the present work, such an
approach has been used in the investigation of the alkali-metal-
cation complexation by a calix[4]arene secondary-amide
derivative, namely, 5,11,17,23-tetra-tert-butyl-25,26,27,28-
tetrakis(N-hexylcarbamoylmethoxy)calix[4]arene (L, Scheme
1) in MeCN. The goal was to address as many processes/
1H NMR (600 MHz, CDCl3, 25 °C, TMS): δ 7.46 (4 H, t, J = 5.82,
NH), 6.77 (8 H, s, ArH), 4.47 (4 H, d, J = 12.9, ArCH2Ar), 4.47 (8 H,
s, OCH2CO), 3.37−3.33 (8 H, m, CH2NH), 3.23 (4 H, d, J = 13.0,
ArCH2Ar), 1.59−1.55 (8 H, m, CH2CH2NH), 1.35−1.28 (24 H, m,
CH2CH2CH2), 1.08 (36 H, s, C(CH3)3), 0.89 (12 H, t, J = 6.8, CH3).
13C NMR (75 MHz, CDCl3, 25 °C): δ 169.50 (CO), 152.95
(ArC−O), 145.75 (p-ArC), 132.69 (m-ArC), 125.80 (o-ArC), 74.63
(OCH2CO), 39.59 (NHCH2), 33.88 (C-tert-butyl), 31.60 (ArCH2Ar),
31. 49 (NCH2 CH2 ), 31. 32 (CH3 -tert-butyl), 29. 74
(CH2CH2CH2CH3), 26.81 (CH2CH2CH2CH3), 22.65 (CH2CH3),
14.05 (CH3).
a
Scheme 1. Synthesis of Compound L
FTIR (KBr pellet): νmax/cm−1 3289 (s, NH), 3088 (w, Ar−H),
1654 (CONH, amide I), 1545 (CONH, amide II), 1362 (w, tert-
butyl), 1195 (s, COC).
Elem anal. Calcd for C76H116N4O8 (1213.76): C, 75.21; H, 9.63; N,
a
4.62. Found: C, 74.90; H, 9.37; N, 4.32.
Reagents and conditions: (i) NaOH, EtOH/H2O; (ii) SOCl2; (iii)
+
MS: m/z 1213.9 [(M + H)+]. Calcd for C76H117N4O8 : m/z 1213.9,
Et3N, dry CH2Cl2, Ar.
1235.8 [(M + Na)+]. Calcd for C76H116N4O8Na+: m/z 1235.9.
(b) Preparation of compound L by aminolysis: To 100 mL of
absolute MeOH were added 1 mmol of the calix[4]arene ethyl ester
and 32 mmol of n-hexylamine. The mixture was kept for several weeks
at room temperature. Colorless transparent needles of a MeOH
solvate of L crystallized from the solution. The crystals were separated
from the mother liquor by suction filtration, washed with MeOH, and
dried to obtain a white powder, which was pure compound L (yield
54%). The spectroscopic data of the prepared compound were
identical with those described above.
Materials for Physicochemical Measurements. The salts used
for the investigation of L complexation were LiClO4 (Fluka, p.a., and
Sigma Aldrich, 99.99%), NaClO4·H2O (Fluka p.a.), NaClO4 (Sigma
Aldrich, 98+%), KClO4 (Merck, p.a.), CsNO3 (Sigma, 99.5%), and
RbNO3 (Sigma, 99.7%). The solvent, acetonitrile (MeCN; Merck,
Uvasol), was used without further purification. In most spectrophoto-
metric and in all potentiometric titrations, the ionic strength was kept
constant by the addition of Et4NClO4 (Fluka, p.a.).
Spectrophotometry. UV titrations were performed at 25.0 0.1
°C by means of a Varian Cary 5 double-beam spectrophotometer
equipped with a thermostatting device. The spectral changes of the L
solution (V0 = 2.0 cm3, c0 = 2.40 × 10−4−2.46 × 10−4 mol dm−3, and Ic
= 0.01 mol dm−3) were recorded upon the stepwise addition of an
alkali-metal salt solution (c = 9.59 × 10−4−3.88 × 10−3 mol dm−3 and
Ic = 0.01 mol dm−3) directly into the measuring cell of 1 cm path
length (Hellma, Suprasil QX). Absorbances were sampled at 1 nm
intervals, with an integration time of 0.2 s. Titrations for each M+/L
system were done in triplicate. The obtained spectrophotometric data
were processed using the SPECFIT program.11
interactions involved in the complexation reaction as possible
from both thermodynamic and structural points of view. In line
with this, detailed spectrophotometric, NMR, electrochemical,
and microcalorimetric investigations were carried out, along
with the X-ray solid-state structure determinations of the free
ligand and its complexes. Particular attention was focused on
the formation of the LMeCN adduct and the ternary
MLMeCN+ complexes (M denotes alkali metal). To the best
of our knowledge, for the first time, the equilibrium constants
for binding of the MeCN molecule with free calixarene and its
metal-ion complex in a MeCN solution were estimated. In
addition, the microscopic equilibrium constants for the
formation of LiL+ and LiLMeCN+ complex species were
assessed. To shed more light on the complexation properties
and structures of the species involved, classical molecular
dynamics (MD) and quantum-chemical computations were
also performed. Among other things, these studies provided
corroborations of the conclusions made on the basis of the
experimental results.
EXPERIMENTAL SECTION
■
Synthesis. All reagents were of the best grade commercially
available and were used without further purification. Solvents were
purified by standard procedures.9 The acid chloride of calix[4]arene
tetracetic acid (the cone conformer) was prepared as described in the
literature.10 Analytical thin-layer chromatography was performed on
silica gel plates (SiO2, Merck 60 F254). Melting points were determined
Potentiometry. The stability constant of the NaL+ complex in
MeCN was determined by the potentiometric titration of a 30.3 cm3
NaClO4 solution (c0 = 1.0 × 10−4 mol dm−3) with a solution of L (c =
7.34 × 10−4 mol dm−3) in a thermostatted titration vessel (t = 25.0
0.1 °C). The ionic strength of both solutions was set to 0.01 mol dm−3
by Et4NClO4. The indicator electrode was a sodium-selective glass
electrode (Metrohm, 6.0501.100) with an Ag/AgCl reference
electrode (Metrohm, 6.0733.100) filled with a MeCN solution of
Et4NCl (c = 0.01 mol dm−3). The working and reference half-cells
were connected with a salt bridge containing 0.01 mol dm−3
Et4NClO4. A Metrohm 713 pH meter was used for electromotivity
measurements. The cell was calibrated by the incremental addition of a
NaClO4 solution (c = 9.94 × 10−3 mol dm−3) to a 30.0 cm3 solution of
on a Kofler hot-bench apparatus and were not corrected. H and 13C
1
NMR spectra were recorded with a Bruker Avance 300 or 600 MHz
spectrometer (δ in ppm relative to (CH3)4Si as an internal standard, J
values are in hertz). IR spectra were recorded by means of an ABB
Bomem MB102 FTIR spectrometer. Mass spectrometry (MS)
measurements were conducted on an Agilent 6410 Triple-Quadrupole
mass spectrometer. Elemental analysis (C, H, and N) was provided by
the Analytical Services Laboratory of the Rudjer Bosk
̌ ́
ovic Institute,
Zagreb, Croatia.
5,11,17,23-Tetra-tert-butyl-25,26,27,28-tetrakis(N-
hexylcarbamoylmethoxy)calix[4]arene (L). (a) n-Hexylamine (0.48
B
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 1. Crystallographic Data and Details of Data Collection for the Crystal Structures of L, L·2MeOH, LMeCN, [NaL]ClO4,
and [NaLMeCN]barb
L
L·2MeOH
C78H124N4O10
LMeCN
C78H121N5O9
[NaL]ClO4
[NaLMeCN]barb
molecular formula
Mr
C76H116N4O8
1213.73
monoclinic
C2/c
C76H116N4O12NaCl
1336.17
tetragonal
P4/n
C88H142N7O15Na
1561.09
1277.81
triclinic
1272.8
cryst syst
space group
a/Å
triclinic
triclinic
P1
̅
P1
̅
P1
̅
31.981(2)
24.792(2)
18.5200(17)
90
14.9313(6)
16.3464(8)
18.3108(9)
108.310(4)
104.079(5)
106.294(4)
3793.6(3)
2
11.6750(16)
18.183(3)
20.382(3)
65.402(16)
74.233(13)
79.578(13)
3774.8(10)
2
15.1410(19)
15.1410(19)
17.627(3)
90
13.1464(4)
16.8767(6)
22.1602(8)
98.404(3)
90.787(3)
106.426(3)
4657.5(3)
2
b/Å
c/Å
α/deg
β/deg
90.575(7)
90
90
γ/deg
V/Å3
90
14683(2)
8
4041.0(9)
2
Z
ρcalc/g cm−3
λ(Mo Kα)/Å
T/K
cryst dimens/mm3
μ/mm−1
F(000)
1.098
1.119
1.12
1.089
1.109
0.71073
100.0(2)
0.31 × 0.24 × 0.09
0.07
0.71073
0.71073
0.71073
100.0(2)
0.54 × 0.50 × 0.11
0.109
0.71073
120.0(1)
0.56 × 0.47 × 0.33
0.073
120.0(1)
0.29 × 0.17 × 0.12
0.072
120.0(1)
0.36 × 0.12 × 0.11
0.079
5312
1400
1392
1472
1688
θ range/deg
hkl ranges
4−25
4−25.5
4−25
3.72−25
3.82−25
−38, 33; −29, 29; −22, −18, 16; −19, 19; −22, −13, 13; −21, 13; −24, −18, 18; −18, 10; −20, −15, 15; −18, 20; −24,
22
21
24
20
26
no. of measd reflns
no. of indep reflns
45804
12065
3617
30764
13514
6804
18772
11994
2591
23626
3548
1156
35524
15632
8850
no. of reflns with I >
4σ(I)
no. of param
Δρmax, Δρmin/e Å−3
R [F2 > 4σ(F2)]
wR (F2)
888
868
854
320
1076
0.923, −0.565
0.1292
0.367
0.886, −0.6887
0.0511
0.384, −0.272
0.061
0.343, −0.454
0.1551
0.4346
1.138
1.833, −0.865
0.1054
0.1304
0.1411
0.623
0.3246
GOF, S
0.983
1.982
1.103
Et4NClO4 (c = 0.01 mol dm−3) in MeCN. In each calibration
experiment, a Nernstian-like behavior was observed, with the slope of
the E versus p[Na] plot being about −58 mV. Titration was repeated
three times, and the obtained potentiometric data were analyzed with
the HYPERQUAD program.12
ΔrH = −31.9 kJ mol−1) were in good agreement with the literature
values (log K = 3.77 and ΔrH = −31.4 kJ mol−1).13 Thermograms
were processed using the Microcal OriginPro7.0 program.
In the calorimetric titrations, the enthalpy changes were recorded
upon the stepwise additions of a MeCN solution of alkali-metal salt to
a solution of L (V0 = 1.4182 cm3). Heats of dilution of salt solutions
were obtained by blank experiments and subtracted from the heats
measured in the titration experiment. The dependence of successive
enthalpy changes on the titration volumes was processed by a
nonlinear least-squares fitting procedure using the OriginPro7.5
program. All measurements were conducted three or more times.
X-ray Structure Determination. The single crystals of
L·2MeOH, LMeCN, and [NaL MeCN]barb (barb = 5,5-diethylbarbi-
turate) were obtained by the evaporation of a saturated solution at
room temperature, while [NaL]ClO4 was crystallized by the vapor
diffusion of tetrahydrofuran (THF). L crystallized from an ethanol
solution, L·2MeOH from a MeOH solution, and LMeCN from a
mixture of MeCN, MeOH, and water (5:5:1 volume ratio). Single
crystals of [NaLMeCN]barb were obtained by the evaporation of a
solution prepared by mixing of a solution of sodium 5,5-
diethylbarbiturate in a mixture of water and MeOH (1:3) with a
MeCN solution of L, so that the L/NaC7H11O3N2 mole ratio was
about 1:3.
The crystal and molecular structures were determined by single-
crystal X-ray diffraction. Diffraction measurements were made on an
Oxford Diffraction Xcalibur Kappa CCD X-ray diffractometer with
graphite-monochromated Mo Kα (λ = 0.71073 Å) radiation.14 The
data sets were collected using the ω-scan mode over a 2θ range up to
54°. The structures were solved by direct methods and refined using
the SHELXS and SHELXL programs.15 The structural refinement was
performed on F2 using all data. The hydrogen atoms were placed in
calculated positions and treated as riding on their parent atoms [C−H
Conductometry. Conductometric titrations were carried out at
25.0 0.1 °C by means of a Metrohm 712 conductometer. The cell
constant (0.892
0.001 cm−1) was determined before each
experiment using a 0.01 mol dm−3 aqueous KCl solution. The alkali-
metal salt solution (V0 = 21.0 or 22.0 cm3 and c0 = 9.0 × 10−5−1.16 ×
10−4 mol dm−3) was titrated with a ligand solution (c = 6.58 × 10−4−
8.21 × 10−4 mol dm−3) in a thermostatted vessel. The measured
conductivities were corrected for the conductivity of the solvent. The
obtained data were processed by the OriginPro7.5 program.
1H NMR Studies. 1H NMR titrations were carried out at 25 °C by
means of a Bruker Avance 600 MHz with a solvent signal used as a
standard for titrations in CD3CN or with a tetramethylsilane (TMS)
signal as a standard in CDCl3 solutions. In the titrations of L with
metal cations in MeCN, solutions made of alkali-metal salt (c = 2.87 ×
10−4−3.92 × 10−4 mol dm−3) and L (c = 5.32 × 10−4 mol dm−3) were
added in a CD3CN solution of L (c = 5.32 × 10−4 mol dm−3). In the
case of the studies of MeCN binding by L and NaL+ in CDCl3,
solutions of L (c = 1.70 × 10−3 mol dm−3) and [NaL]ClO4 (c = 1.24 ×
10−4 mol dm−3) were titrated with a MeCN solution (c = 8.43 × 10−3
mol dm−3). The temperature dependence of the 1H NMR spectrum of
L was measured in CD3CN (from −35 to +60 °C) and in CDCl3
(from −50 to +50 °C). Spectra were recorded at 32 pulses.
Calorimetry. Microcalorimetric measurements were performed by
a Microcal VP-ITC isothermal titration calorimeter at 25.0 °C. The
calorimeter was calibrated electrically, and its reliability was addition-
ally checked by carrying out the complexation of Ba2+ by 18-crown-6
in a aqueous medium at 25 °C. The results obtained (log K = 3.76 and
C
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
= 0.93 Å and Uiso(H) = 1.2Ueq(C); C−H = 0.97 Å and Uiso(H) = 1.2
Ueq(C)], except those involved in hydrogen bonding, which were
located from the electron difference map, if it was possible (in
structures of L·2MeOH and LMeCN). All calculations were
performed, and the drawings were prepared using the WinGX
crystallographic suite of programs.16 The crystal data and measure-
ment details are listed in Table 1. Further details are available in the
Supporting Information and from the Cambridge Crystallographic
Centre17 with quotation numbers 815093−815097.
Isotropic shieldings of L′ and tetramethysilane (TMS) in vacuo and
in solvent were calculated by applying the gauge-independent atomic
orbital (GIAO) and PCM GIAO22 methods on the previously
optimized geometries using the same basis set as that above. Chemical
shifts (1H and 13C) were calculated with respect to TMS.
MD Simulations. The MD simulations were carried out by means
of the GROMACS23 package (version 4.5.3). Intramolecular and
nonbonded intermolecular interactions were modeled by the OPLS-
AA (Optimized Parameters for Liquid Simulations-All Atoms) force
field,24 which has already been successfully used to simulate
calix[4]arene derivatives and their complexes with metal cations
and/or MeCN.25 Partial charges assigned to ring carbon atoms bound
to CH2 groups that link the monomers were assumed to be zero, as
described in ref 25. In all simulations, the initial molecular structure of
L was the one obtained from the NaL+ crystal. The L and ML+ species
were solvated in a cubical box (edge length 58.5 Å) of MeCN with
2156 molecules and with periodic boundary conditions. The solute
concentration in such a box was about 0.01 mol dm−3. The solvent box
was equilibrated prior to inclusion of L and its complexes, with the box
density after equilibration being close to the experimental one within
5%. During simulations of the systems comprising calixarene and metal
cations, the Cl− ion was included to neutralize the box. The chloride
counterion was kept fixed at the box periphery, whereas the complex
was initially positioned at the box center. In all simulations, an energy
minimization procedure was performed followed by a MD simulation
in NpT conditions for 50.5 ns, where first 0.2 ns were not used in the
data analysis. The Verlet algorithm26 with a time step of 1 fs was
employed.
A number of difficulties were encountered in the refinement of the
structural models, mostly because of weak diffraction of the samples
and because of the structural disorder of tert-butyl and hexyl groups. It
was therefore necessary to introduce multiple restraints of bond
distances and angles as well as thermal parameters in these parts of the
structures. In the structure of L, two hexyl chains were modeled as
disordered over two positions, although the actual disorder is probably
more complex and irresolvable, with hydrophobic parts of the crystal
structure resembling a hydrocarbon glass. In the structure of L·2MeOH,
one of the MeOH molecules is disordered and had to be modeled,
with the position of the hydroxyl hydrogen atom concluded based on
the proximity of an appropriate hydrogen-bond acceptor. The crystal
of LMeCN was very poorly diffracting, and therefore a large portion of
reflections were treated as unobserved, especially at 2θ values above
30°, which has naturally led to problems with resolution and the need
for numerous geometrical restraints in the model. In the structure of
[NaLMeCN]barb, together with a rotational disorder of tert-butyl
groups, there is an only partly resolvable disorder of the hexyl chains
present, similar to, although not as severe as in, L. The structure also
contains an extensive hydrogen-bonding network with four water
molecules independent of symmetry on which the hydrogen atoms
could not be located from the electron density map. The most
problematic is the structure of [NaL]ClO4, where the hexyl chains do
not obey the tetragonal crystallographic symmetry of the calixarene
cone and, therefore, could not be located from the electron density
map and had to be modeled. They were modeled as disordered over
two sets of positions in order to better account for the residual
electron density, which is almost uniform in the entire region occupied
by the hexyl groups. In addition, the tert-butyl groups and the
perchlorate anions are also disordered over two positions. A positional
disorder of the perchlorate anions occurred because of placement of
both the cations and anions on special positions because the positions
are of different multiplicities, with the cations repeated twice and the
anions apparently four times within the unit cell. This required the
The cutoff radius for nonbonded van der Waals and short-range
Coulomb interactions was 16 Å. Long-range Coulomb interactions
were treated by the Ewald method, as implemented in the Particle
Mesh Ewald procedure.27 The simulation temperature was kept at
298.15 K with the Nose−
́
Hoover28 algorithm using a time constant of
1 ps. The pressure was kept at 1 bar by the Martyna−Tuckerman−
Tobias−Klein29 algorithm and a time constant of 1 ps.
RESULTS AND DISCUSSION
■
Synthesis. Calix[4]arene tetrahexylamide L was synthesized
according to the previously described method,5b starting from
the corresponding tetraester (cone conformation) by (i)
hydrolysis to the tetraacid, (ii) activation to the acid chloride,10
and (iii) amide bond formation (Scheme 1). Alternatively, the
same compound can be prepared via aminolysis of the
tetraester with n-hexylamine, as described in the Experimental
Section.
1
occupancy of the anion to be /2 at each site, in order to retain the
electric neutrality of the model. It is possible that the latter problem
might be resolved by solving and refining the structure in an I-centered
supercell of quadruple volume. Unfortunately, an attempt of
refinement of such a model yielded most unsatisfactory results, with
unreasonable bond lengths and R of over 35%. However, as has been
shown recently,18 this is not necessarily an indication that the supercell
model is incorrect but rather that the diffraction data are not of
sufficient quality to support a model with a larger number of refinable
parameters. Therefore, it was concluded that the model with the
smaller unit cell should be used, even though it may actually represent
an average structure.
DFT Calculations. Quantum-chemical calculations were performed
using the Gaussian 03 program package19 with default convergence
criteria at the B3LYP/6-311++G(d,p) level of theory. To reduce the
computational time and numerical problems, methyl groups were
introduced in the substituents at the lower rim (L′, Figure 12) instead
of the hexyl groups. Geometry optimizations were performed
separately for the square cone and flattened cone conformations of
calix[4]arene, and its complex with the Na+ cation placed in the lower-
rim hydrophilic cavity. Solvent effects (chloroform and MeCN) were
introduced in the calculation using the reformulation of the polarizable
continuum method (PCM)20 known as the integral equation
formalism of the PCM of Tomasi and co-workers.21 In addition, the
adduct of L′ with one MeCN molecule in the hydrophobic cone was
formed and optimized.
Crystal Structures and Stereochemistries of L,
L·2MeOH, LMeCN, [NaL]ClO4, and [NaLMeCN]barb.
Solvent-free crystals of L were obtained from its ethanol
(EtOH) solution, while crystallization from MeOH yielded a
MeOH solvate, L·2MeOH, with two MeOH molecules per
molecule of L.
In the solvent-free crystal and the MeOH solvate, the
molecular structure of L (Figure 1) revealed almost identical
distorted calixarene cones of an approximate C2 symmetry
(Figure 2). The cone is characterized by alternating signs of the
torsion angles φ and χ (Table S1, Supporting Information)
about the methylene bonds of the macrocyclic ring.30
Deformation of the cone can be described as tilting of the
phenyl rings toward the macrocycle plane defined by the
methylene carbon atoms. In the molecule of L, two rings are
almost parallel (the planes of the rings are at an angle of 1.6° in
L and 2.7° in L·2MeOH) and almost perpendicular to the
plane of the macrocycle (81.6° and 80.7° in L; 82.0° and 84.6°
in L·2MeOH). The other two rings are tilted from the
macrocycle plane with angles of the phenyl rings to the plane of
42.0° and 48.3° in L (44.5° and 40.3° in L·2MeOH). The
D
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Intramolecular hydrogen bonding in crystal structures of (a)
a solvent-free crystal of L, (b) the MeOH solvate L·2MeOH, and (c)
the LMeCN adduct. Hydrogen atoms bonded to carbon atoms of L
and solvent molecules have been omitted for clarity.
nitrogen atoms are hydrogen-bonded to one carboxyl oxygen
atom, forming two intramolecular hydrogen bonds (N1−
H1···O7 and N4−H4···O7). The remaining amide groups
participate in intermolecular hydrogen bonding, either with
neighboring molecules of L (two bonds N2−H2···O5) or with
MeOH molecules of which one acts only as hydrogen donor to
an amide oxygen atom (O10−H···O6), while the other is
bridging between two molecules of L (O9−H···O8 and N3−
H···O9). The hydrogen-bonding network interconnects the
molecules in chains along the [101] direction (Table S2,
Supporting Information).
Figure 1. ORTEP plot of the L molecule in the structure of L·2MeOH
with the labeling scheme for non-carbon atoms. Thermal ellipsoids are
shown with 50% probability. Hydrogen atoms as well as minor
components of all disordered groups have been omitted for clarity.
The labeling of the atoms is analogous in all of the structures.
Crystallization of L from a mixture of MeCN, MeOH, and
water yielded LMeCN monohydrate. The MeCN molecule is
situated in the calixarene cone, with the methyl group of the
MeCN forming a weak C−H···π hydrogen bond with the
aromatic rings of the cone (C···centroid distances between
3.413 and 3.783 Å; Figure 4a). The cone is much more regular
Figure 2. Overlap of the molecules of L from the crystal structures of
the solvent-free crystal (green) and the MeOH solvate (red), showing
almost identical cone geometry in the two crystals. Hydrogen atoms as
well as minor components of all disordered groups have been omitted
for clarity.
Figure 4. Position of the MeCN molecule nested in the calixarene cone
in crystal structures of (a) LMeCN and (b) [NaLMeCN]barb. The
non-hydrogen atoms of the MeCN molecules are shown as thermal
motion ellipsoids plotted with 50% probability, and the MeCN
hydrogen atoms are shown as small spheres of arbitrary radius. The
hydrogen atoms of the calixarene unit have been omitted for clarity.
planes of the latter two rings are almost perpendicular to each
other, being at an angle of 89.7° (84.6° in L·2MeOH).
In the crystal structure of the solvent-free L, there is an
extensive network of hydrogen bonds present. Two amide
nitrogen atoms participate in an intramolecular hydrogen bond
with the same carbonyl oxygen atom (N2−H2n···O8 and N3−
H3n···O8), while the other nitrogen atoms are hydrogen-
bonded to carbonyl groups of neighboring molecules (N1−
H1n···O6 and N4−H4n···O7), forming a 2D hydrogen-
bonding network parallel to the bc plane (Table S2, Supporting
Information). Out of four carbonyl oxygen atoms, only O5
does not participate in strong hydrogen bonding but is in close
contact with hydrogen atoms from two methylene groups,
namely, C68−H68a from the same molecule and C60−H60a
from the neighboring one.
than when MeCN is absent, although it is still somewhat
flattened (Figure 5). The angles of phenyl rings to the
macrocycle plane are 74.74°, 68.27°, 58.00°, and 57.58°. Of the
four N−H groups, one is involved in an intramolecular
hydrogen bond (N3−H3···O6; Figure 3c) and one (N4−H4)
is positioned so that the hydrogen atom is positioned toward
the calixarene macrocycle and approximately equidistant (ca
2.73 Å) to ether oxygen atoms (O1, O2, and O3). N2−H2
forms a hydrogen bond with a water molecule, which further
bonds via O9−H6···O5 to a carbonyl oxygen atom of the same
calixarene molecule and via O9−H5···O7 to a carbonyl oxygen
atom of a neighboring calixarene molecule. The remaining N1−
H1 bonds via N1−H1···O8 to a neighboring calixarene
molecule (Table S2, Supporting Information). The intermo-
The intramolecular hydrogen bonding in the crystal structure
of the MeOH solvate (L·2MeOH) is very similar to that in
solvent-free L (Figure 3a,b) because here also two amide
E
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Overlap of the molecules of L from the crystal structures of
L·2MeOH, (black), the MeCN adduct LMeCN (red), and the sodium
complex [NaL]ClO4 (green). Hydrogen atoms have been omitted for
clarity.
Figure 6. Overlap of the [NaL]+ ions from the crystal structures of
[NaL]ClO4 (green) and [NaLMeCN]barb (red). tert-Butyl groups
have been omitted for clarity.
lecular hydrogen bonding forms a double chain of molecules
along the [100] direction.
between 3.563 and 3.610 Å; Figure 4b). In addition to the
intermolecular hydrogen bond mentioned above, the [NaL]+
ion participates in an extensive hydrogen-bonding network,
which also includes the 5,5-diethylbarbiturate anion, ethanol,
and four symmetrically independent water molecules. This
intermolecular hydrogen bonding forms a double chain of
molecules along the [100] direction.
In [NaL]ClO4, the sodium ion is located on the crystallo-
graphic 4-fold axis. The coordination polyhedron around the
sodium atom is an intermediate between a cube and a square
antiprism and is formed by four ether and four carbonyl oxygen
atoms with Na−O bond lengths of 2.556(5) and 2.368(17) Å,
respectively. The cone is of perfectly regular conformation of C4
symmetry (positioned on a crystallographic 4-fold axis; Figure
5). Phenyl rings are at an angle of 67.29° to the macrocycle
plane. Amide nitrogen atoms from [NaL]+ moieties form
hydrogen bonds with perchlorate ions, which, in turn, bridge to
further [NaL]+ ions, thus forming a 2D network perpendicular
to the [100] direction.
Unfortunately, the hexyl groups disobeyed the C4 crystallo-
graphic symmetry, causing severe and irresolvable disorder. In
addition, the tert-butyl groups, the perchlorate anion, and the
solvent water present in the structure are also in disorder,
reducing the quality of the structural model. To overcome this
problem, a second complex was prepared, using sodium 5,5-
diethylbarbiturate instead of perchlorate. It was presumed that
such a complex will not crystallize in the tetragonal system, thus
allowing the ordering of the hexyl groups. The crystallization of
this complex from a mixture of MeCN, MeOH, and water
yielded [NaLMeCN]barb MeOH monosolvate tetrahydrate.
With the expected lowering of the symmetry, an unexpected
change in the geometry of [NaL]+ also occurred. The
coordination number of the sodium atom was no longer 8
but 7, with one of the amide groups rotated by approximately
180°, so that instead of carbonyl oxygen bonding to the central
sodium, the amide nitrogen forms a hydrogen bond (N1−
H1···O6) with a coordinated carbonyl oxygen. The coordina-
tion polyhedron of the sodium atom can be described as a
square-face monocapped trigonal prism. The calixarene cone is
of regular conformation (approximate C4 symmetry), and the
values of the angles between the phenyl rings and the
macrocycle plane of 64.47°, 64.67°, 66.32°, and 66.67°
approach the value of 67.29° found in the structure of
[NaL]ClO4 (Figure 6). Like in the structure of LMeCN, the
MeCN molecule is situated in the calixarene cone, with the
methyl group of the MeCN forming a weak C−H···π hydrogen
bond with the aromatic rings of the cone (C···centroid distances
Structural Studies of L in Solution. The structure of the
investigated calix[4]arene derivative in solution was studied by
1
means of H NMR spectroscopy. Three solvents were used,
namely, deuterated chloroform, methanol, and acetonitrile. The
1H NMR spectrum of L in CDCl3 (Figure S6 and Table S3,
Supporting Information) showed two singlets at 1.08 and 6.77
ppm due to tert-butyl groups and calixarene aromatic protons
(Ar−H), respectively. In addition, two sets of doublets due to
the bridging methylene protons (3.23 and 4.47 ppm) were
observed along with the singlet at 4.75 ppm, which corresponds
to ArOCH2 protons. Such a pattern is characteristic for the cone
conformation with C4v symmetry or time-averaged C4v
conformation of tetrasubstituted calix[4]arenes.1e A rather
high chemical shift of NH protons (7.46 ppm) indicated the
presence of intramolecular NH...OC hydrogen bonds.5a,b The
spectrum of L in CD3OD (Figure S6 and Table S3, Supporting
Information) was basically similar (apart from the absence of
the NH signal). On the other hand, the 1H NMR spectrum of L
in CD3CN was quite different, being characterized by two sets
of signals, i.e., two signals for each of the ligand protons with
unequal integrals (Figure 7 and Table S3, Supporting
Information). This interesting finding can be explained by
taking into account the possibility of inclusion of a MeCN
molecule into the calixarene hydrophobic cavity, which is well
documented in the literature4a−f,5k,7 and observed in the crystal
structure of LMeCN. Thus, the presence of two sets of signals
indicates the existence of an equilibrium between the “free”
ligand and the one with a bound MeCN molecule. For example,
two singlets assigned to ArH protons appeared at 7.33 ppm
(3.24 H) and 7.12 ppm (4.76 H). Because the downfield shift
of the ArH signal can be taken as an indication of the
interaction of the MeCN molecule with the calix[4]arene
aromatic rings,7a,8a the lower field resonance can be attributed
F
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
determined by microcalorimetric, spectrophotometric (KL+),
conductometric (KL+), and potentiometric (NaL+) titrations,
and the obtained values are given in Table 2.
The UV spectral changes observed upon the addition of
LiClO4, NaClO4, or KClO4 to the solution of L in MeCN were
qualitatively similar. In all cases, the ligand UV spectrum
exhibited a hypochromic shift of its larger part, accompanied by
the occurrence of a well-defined isosbestic point. The
spectrophotometric titration of L with K+ is shown in Figure
8, and the stability constant of the KL+ complex computed by
1
Figure 7. H NMR spectrum of L in CD3CN. Signals corresponding
to the LMeCN species are labeled by asterisks.
to the LMeCN species and the higher field one to the free
ligand. The complete assignment of the signals in the 1H NMR
spectrum of L in CD3CN was carried out with the aid of the
results of computational studies, temperature dependence of
the L NMR spectrum, and NMR titrations of L with metal salt
solutions (see below).
The equilibrium constant for the reaction of inclusion of the
MeCN molecule in the L hydrophobic cavity (L + MeCN ⇄
LMeCN) can be calculated as a ratio of integrals of signals
corresponding to L and LMeCN. At 25 °C, it amounts to 0.68
(estimated from the integrals of signals corresponding to ArH
and tert-butyl protons).
Temperature dependences of 1H NMR spectra of L in
CDCl3 and CD3CN are shown in Figures S7 and S8
(Supporting Information). Upon heating from −50 to +50
°C, the only considerable change observed in chloroform was
that of the chemical shift of the amide NH protons, which was
shifted upfield. That was expected as a consequence of
intramolecular NH···OC hydrogen-bond cleavage.5b The
situation was much more interesting in the case of MeCN. The
intensities of the signals belonging to the one set of L signals
(denoted by asterisks in Figure 7) gradually decreased with an
increase in the temperature, whereas those of the other set
increased. Because inclusion of the MeCN molecule in the
hydrophobic cavity of calix[4]arene is known to be
exothermic,4c,d the observed changes can be explained by
taking into account that with increasing temperature the
equilibrium between L and LMeCN species is shifted toward
the free ligand. This finding can serve as an additional
Figure 8. (a) Spectrophotometric titration of L (c = 2.57 × 10−4 mol
dm−3 and V0 = 2.0 cm3) with KClO4 (c = 3.3 × 10−3 mol dm−3) in
MeCN. Ic = 0.1 mol dm−3 (Et4NClO4); t = 25.0 0.1 °C. Spectra are
corrected for dilution. (b) Dependence of the absorbance at 282 nm
+
■
on the n(K )/n(L) ratio: ( ) experimental; () calculated.
1
corroboration of the assignment of the ligand H NMR signals
presented in Figure 7.
processing these data is given in Table 2. The stability constants
of the LiL+ and NaL+ complexes were too large to be
determined spectrometrically because of the limited sensitivity
of this method. However, the strong complexation and 1:1
Studies of Cation Complexation in Solution. Stability
constants of the LiL+, NaL+, and KL+ complexes in MeCN were
Table 2. Stability Constants of Complexes of Alkali-Metal Cations with L in MeCN at 25 °C
log K SE
cation
Li+
Na+
K+
Rb+
Cs+
calorimetry
spectrophotometry
conductometry
potentiometry
6.93 0.03
6.04 0.05
>5
>5
>5
3.56
a
6.88 0.05
>5
3.61 0.02
3.54 0.02
a
a
a
a
a
a
No complexation was observed. SE = standard error of the mean (N = 3−4).
G
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(M+/L) stoichiometry of these complexes were undoubtly
proven by the almost linear dependence of the absorbance on
the n(M+)/n(L) ratio (n denotes the total amount), followed
by a break in the titration curve at a molar ratio of
approximately 1 (Figure S14, Supporting Information). The
addition of Rb+ and Cs+ nitrates in the MeCN solution of L did
not cause any significant changes in its UV spectrum, indicating
that complexation of larger cations by the ligand was very weak
or nonexistent.
The above results were confirmed by conductometric
measurements. In the cases of LiClO4, NaClO4, and KClO4
solutions, the conductivity (corrected for dilution) decreased
upon the addition of the ligand (due to a lower electric mobility
of the larger ML+ complex compared to that of the free cation),
whereas during titrations of RbNO3 and CsNO3 solutions with
L, almost no change of the conductivity was observed. As can
be seen in Figure S15 (Supporting Information), in the
titrations of LiClO4 and NaClO4 solutions, the conductivity
decreased almost linearly with the amount of ligand added, and
at the molar ratio n(L)/n(M+) ≈ 1, a break in the titration
curve occurred. On the other hand, in the case of KClO4, the κ
versus n(L) dependence was curved (Figure 9), which allowed
Figure 10. Potentiometric titration of NaClO4 (c = 1 × 10−4 mol dm−3
and V0 = 30.24 cm3) with L (c = 7.68 × 10−4 mol dm−3) in MeCN:
( ) experimental; () calculated. Ic = 0.01 mol dm−3 (Et4NClO4); t
■
= 25.0 0.1 °C.
quite high (Table 2), in agreement with the results of the other
measurements conducted in this work.
The thermodynamic parameters obtained by microcalori-
metric titrations are summarized in Table 3. The stepwise
Table 3. Thermodynamic Parameters for Complexation of
Alkali-Metal Cations with L in MeCN at 25 °C Obtained by
a
Calorimetry
(ΔrG° SE)/kJ
(ΔrH° SE)/kJ
(ΔrS° SE)/J
cation
mol−1
mol−1
K−1 mol−1
Li+
Na+
K+
−34.5 0.2
−39.3 0.1
−20.63 0.04
−12.9 0.1
−33.8 0.2
−19.7 0.4
72.3 0.8
18.5 1.2
3.1 1.3
a
SE = standard error of the mean (N = 3−5).
addition of a LiClO4, NaClO4, or KClO4 solution to the MeCN
solution of L resulted in exothermic enthalpy changes. Standard
reaction enthalpies and equilibrium constants were computed
by least-squares nonlinear analysis of the calorimetric data. As
an example, the experimental and fitted data corresponding to
the titration of L with Na+ are shown in Figure 11 (the results
of other calorimetric measurements are given in Figures S14
and S15, Supporting Information). Standard reaction Gibbs
energies and entropies were obtained by means of the well-
known equations ΔrG° = −RT ln K and ΔrG° = ΔrH° − TΔrS°,
respectively.
Figure 9. Conductometric titration of KClO4 (c = 9.7 × 10−5 mol
dm−3 and V0 = 21.0 cm3) with L (c = 6.51 × 10−4 mol dm−3 and 7.26
−3
× 10−4 mol dm ) in MeCN: ( ) experimental; () calculated. t =
■
25.0 0.1 °C. Conductivities are corrected for dilution.
for calculation of the KL+ stability constant by nonlinear
regression analysis of the titration data (nonspecific, i.e.,
electrostatic, interactions affecting the conductivity were not
taken into account). The obtained value was in excellent
agreement with those determined by microcalorimetric and
spectrophotometric measurements (Table 2). It should be
noted that the assessed molar conductivities of the [LiL]ClO4
(136 S cm2 mol−1), [NaL]ClO4 (136 S cm2 mol−1), and
[KL]ClO4 (129 S cm2 mol−1) species were similar. That could
be expected if one neglects the ionic interactions and by taking
into account that the size of the conducting complex is mainly
determined by the size of the ligand involved.5i
The stability constant of the NaL+ complex might seem to be
too large to be determined accurately from a microcalorimetric
direct titration experiment. However, its value obtained in that
way is in very good agreement with the one measured by
potentiometric titrations (Table 2). In addition, it should be
noted that the validity of our experimental and fitting procedure
was also checked by performing the simulations suggested by
Backman et al.31
̈
The addition of RbNO3 or CsNO3 to the investigated
calixarene solution did not cause any significant heat effects.
These findings suggested that no complexation took place
under the conditions used or that the values of the
corresponding reaction enthalpies were close to zero. However,
by taking into account the spectrophotometric and conducto-
metric findings, it can be concluded that the investigated
calix[4]arene ion-binding site was not suitable for Rb+ and Cs+
cations.
The stability constant of the NaL+ complex was determined
by direct potentiometry using a sodium-selective glass
electrode. The corresponding titration curve exhibited an
inflection point at n(L)/n(Na+) ≈ 1 (Figure 10). The
equilibrium constant for the complexation reaction between
Na+ and L obtained by processing the potentiometric data was
H
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
complexed (e.g., Ar−H, 7.33 ppm) and uncomplexed ligand
(Ar−H, 7.12 ppm) were at practically the same chemical shifts
as those corresponding to the LMeCN and L species,
respectively. In other words, the positions of these signals
remained almost the same during titration, whereby their
intensities changed (Figure S9, Supporting Information).
Above a molar ratio of approximately 1:1, the spectrum
consisted of a single set of signals (chemical shifts being almost
equal to those of LMeCN) that did not change with an increase
in the amount of NaClO4. The findings described strongly
suggest that conformations of LMeCN and the sodium
complex with L in MeCN are very similar. Moreover, one
can conclude that the Na+−L complex species predominantly
possesses a MeCN molecule included in its hydrophobic cavity.
To examine in more detail the interactions of the
investigated calixarene and its sodium complex with the
1
MeCN molecule, H NMR titrations of these species with
MeCN in deuterated chloroform were carried out. The
formation of the LMeCN adduct could not be detected
because, during titration of L with MeCN (Figure S12,
Supporting Information), the chemical shift of CH3CN protons
in the MeCN/L molar ratio range 3:1 showed no difference
with respect to the corresponing signal of MeCN in CDCl3 (2.0
ppm). Likewise, no significant change of the chemical shifts of
L protons could be observed. That is most likely a consequence
of the rather low extent of MeCN inclusion in the L
hydrophobic cavity under the stoichiometric conditions
applied. On the other hand, binding of MeCN with the NaL+
complex was clearly indicated. At the lowest MeCN/NaL+
molar ratio examined (0.27), the signal of MeCN protons
appeared at 1.80 ppm (upfield shift) and then gradually shifted
toward the position of the MeCN signal in CDCl3 (Figure S13,
Supporting Information). Moreover, the signals of the L
protons also exhibited shifts with increasing MeCN concen-
tration. It should be noted that in chloroform an extensive ion
pairing between NaL+ and ClO4− ions is likely to occur and that
can have an impact on the process of MeCN molecule
inclusion. However, the extent of this impact is hard to assess.
The above results show that the affinity of the NaL+ complex
toward MeCN is much higher than that of the free ligand. The
reason lies in the fact that in the case of NaL+ the ligand
hydrophobic cavity is better preorganized to accept the MeCN
molecule.4f,5k,8c
Figure 11. (a) Microcalorimetric titration of L (c = 4.0 × 10−5 mol
dm−3 and V = 1.4182 cm3) with NaClO4 (c = 4.0 × 10−4 mol dm−3) in
MeCN: t = 25 °C. (b) Dependence of successive enthalpy changes on
+
■
the n(Na )/n(L) ratio. ( ) experimental; () calculated.
The data presented in Table 3 indicate that both enthalpic
and entropic contributions to the reaction Gibbs energy for
complexation of Li+, Na+, and K+ are favorable. However, the
mutual relations of these contributions are substantially
different. The complexation entropy is most favorable for Li+
binding with L, whereas the −ΔrH° value is the lowest for this
reaction. As described elsewhere,5k this could be partly
explained by a stronger solvation of the Li+ cation compared
to the other alkali-metal ions. The enthalpic and overall
thermodynamic stability is the largest in the case of the NaL+
complex. The value of the reaction enthalpy for K+ complex-
ation with L is between those for Li+ and Na+, whereas the
entropic contribution to the binding Gibbs energy is in this case
considerably lower (almost negligible at the standard state of 1
mol dm−3; Table 3). That is reflected in the lowest stability of
the KL+ species compared to NaL+ and LiL+ in MeCN.
In addition to the experiments described above, the reactions
of Li+, Na+, and K+ with L in MeCN were also followed by 1H
NMR titrations. The complexation-induced changes of the L
spectrum were different for each cation studied. In the case of
Na+, below a stoichiometric 1:1 (NaClO4/L) ratio, two distinct,
separate resonances for the NaL+ complex and the free ligand
were present simultaneously (Figure S9, Supporting Informa-
tion). Interestingly, the positions of signals belonging to the
Formation of the complex of L with Li+ was accompanied by
the shifts and broadening of the resonances attributed to L
(without MeCN) species up to the n(Li+)/n(L) ratio ≈1,
whereas the positions of the signals belonging to LMeCN
remained practically unchanged (Figure S10, Supporting
Information). This indicated that, compared to the reaction
of Na+ with L, the exchange kinetics of complexation of Li+ by
L was faster. Beyond the molar ratio of approximately 1, no
further spectral changes occurred, and two distinct sets of
signals were observed (Figure S10, Supporting Information).
The latter was likely to be a consequence of the existence of an
equilibrium between two types of Li+L complexes, i.e., one with
(LiLMeCN+) and the other without (LiL+) a specifically bound
MeCN molecule. The corresponding complexation processes
are presented in Scheme 2.
K1 and K4 denote the equilibrium constants of the MeCN
binding reactions with the free ligand and LiL+ complex,
respectively. The value of K4 was assessed to be 0.88 from the
ratio of integrals of 1H NMR peaks assigned to the LiLMeCN+
and LiL+ species, whereas K1 = 0.68 was estimated earlier (see
I
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Chemical Equilibria Corresponding to the
Binding of Li+ Cation by L
above). It can be easily shown that these constants are related
to the microscopic complexation equilibrium constants K2 and
K3 in the following way:
K1
K4
K2
K3
=
(2)
The macroscopic stability constant K determined calori-
metrically (Table 2) can be expressed as
+
[LiL ] + [LiLMeCN+]
Figure 12. Molecular structures of the square cone (A; C4 point group)
and flattened cone (B; C2 point group) conformations of the model
calix[4]arene L′ optimized at the B3LYP/6-311++G(d,p) level of
theory.
K =
[Li+]{[L] + [LMeCN]}
(3)
or
K2 + K1K3
K =
The DFT calculations showed that the difference in energy is
approximately 4 kJ mol−1 in favor of the flattened cone
conformation. That difference is probably due to the reduced
steric repulsion between tert-butyl groups in the flattened cone
conformation. The minimal distance between protons of
neighboring tert-butyl groups is 2.849 Å for the square cone
conformation, whereas in the case of the flattened cone
conformation, these distances are 3.197 and 3.257 Å, thus
giving a more relaxed structure. This could also explain why L
adopts this conformation in the crystal structures of both L and
its MeOH solvate. In the square cone conformation, there are
four equivalent hydrogen bonds between each carbonyl oxygen
atom and the nitrogen atom with the distance rON = 2.949 Å
and the angle αOHN = 155.5°. In the flattened cone
conformation, there are two equivalent sets of these hydrogen
bonds with the distances and angles of 2.908 Å (148.4°) and
3.086 Å (167.2°).
In order to investigate the effect that binding of the MeCN
molecule has on the structure of calix[4]arene, an adduct with a
MeCN molecule in the hydrophobic cavity was formed and
optimized (Figure 13).
The structure of this species is similar to the square cone
conformation of free L′ (distances between protons of
neighboring tert-butyl groups are in the range from 2.908 to
3.075 Å, while the hydrogen bonds remain essentially the
same). The position of the MeCN molecule inside the cone is
slightly tilted from the centerline (a similar structure was
obtained using the B97D functional). That position indicates a
possibility of active movement of the solvent molecule inside
the calix[4]arene molecule cavity. This conclusion is also
supported by the crystal structure of the MeCN solvate of L,
1 + K1
(4)
By combining eqs 2 and 4, one can calculate the values of
microscopic constants K2 and K3, which were found to be 9.8 ×
105 and 1.3 × 106, respectively.
Like with lithium, the equilibrium kinetics of the L complex
with K+ was found to be fast on the NMR time scale. That can
be easily concluded by inspecting Figure S11 (Supporting
Information), which shows changes of the ligand spectrum
occurring during its titration with KClO4 in deuterated MeCN.
The peaks corresponding to the macrocycle without MeCN in
its hydrophobic cavity exhibited broadening and shifting, while
the positions of the LMeCN peaks did not change, but their
intensities decreased with an increase in the amount of K+ ion.
Complete calixarene complexation could not be achieved
experimentally because of the insufficiently high stability
constant of the K+L complex(es) and insufficient solubility of
KClO4 in MeCN. However, the 1H NMR spectrum of L in the
solution saturated with KClO4 (Figure S11, Supporting
Information) indicates that most probably the complex species
almost exclusively comprises a MeCN molecule in its aromatic
basket. Like in the case of the complex with Na+, a single peak
of ArH protons appeared at 7.33 ppm and the one attributed to
the tert-butyl protons at 1.21 ppm (chemical shifts that almost
completely correspond to those observed for LMeCN,
NaLMeCN+, and LiLMeCN+ species).
Quantum-Chemical Computations. Geometry optimiza-
tions were performed separately for square cone and flattened
cone conformations (C4 and C2 point groups, respectively;
Figure 12) of model calix[4]arene molecule L′ with hexyl
groups replaced by methyl groups.
J
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
coordination of Na+ has been noticed in the crystal structure of
[NaL]ClO4. NH groups are pointing outside and present free
positions for the possible formation of hydrogen bonds with the
solvent molecules in a fashion similar to that seen in the crystal
structure of [NaL]ClO4, where the NH groups are hydrogen-
bonded with the perchlorate anions.
Calculations of NMR Spectra. A comparison between
experimental and theoretical NMR chemical shifts provides
practical information on the chemical structure and con-
formation of the investigated macrocycle and its complexes.
The results of the calculations for the square cone conformation
of calix[4]arene (C4 point group; Figure 12) confirmed our
initial assignation (Tables S4 and S5, Supporting Information).
MD Simulation of Free Ligand in MeCN. The geometry
of the L cone can be represented by the distances between the
opposing phenyl carbon atoms that are directly connected to
the tert-butyl group. In that way, it can be seen (Figure 15) that,
Figure 13. Molecular structure of L′ with an explicit molecule of
MeCN in the hydrophobic cavity optimized at the B3LYP/6-311+
+G(d,p) level of theory.
where together with the sloping of the MeCN molecule an
elongation of the thermal motion ellipsoids of the MeCN cyano
group is noticed (Figure 4), which indicates a motion of the
molecule with the methyl group fixed in the macrocycle cone
and the cyano group oscillating perpendicularly to the
calixarene molecular axis. Therefore, in order to investigate in
more detail the solvent effects on the molecular structure of
calix[4]arene, MD simulations have been carried out.
The optimized structure of the sodium complex with L′ is
given in Figure 14. The minimal distance between protons of
neighboring tert-butyl groups is 2.855 Å, which is again similar
to the square cone conformation of the free ligand. In this
complex, intramolecular hydrogen bonds do not exist because
the Na+ is octacoordinated with the oxygen atoms. Such
Figure 15. Distances between the opposing phenyl carbon atoms that
are directly connected to the tert-butyl group during the MD
simulation of L in MeCN.
immediately after the beginning of simulation, L changed the
conformation from an initial square cone, as in the crystal
structure of NaL+, to a flattened cone, which was similar to the
one obtained by crystallographic and DFT investigations.
After 1.2 ns, a molecule of MeCN entered the L cone. That
was reflected in the distances between the aromatic carbon
atoms, which became similar (Figure 15), making the cone
symmetry approximately C4. The MeCN methyl group pointed
inside the L cone, in accordance with the solid-state structure. A
similar result was observed earlier by MD simulations of
systems comprising the other calixarene derivatives.25,32 The
solvent molecule in the LMeCN adduct left the calixarene
cavity after 22 ns, resulting again in a flattened cone
conformation of L. A total of 4 ns later, another MeCN
molecule entered the cone and remained there until the end of
the simulation. At the lower rim of L, the formation of two
kinds of intramolecular NH···OC hydrogen bonds was
observed, namely, two-center and three-center ones. In the
latter case, two hydrogen atoms are bound to one carbonyl
oxygen atom. The number of intramolecular hydrogen bonds
ranges from 0 to 3, with the average for L being 2.1 and that for
LMeCN 1.5. Criteria applied for an interaction to be
characterized as a hydrogen bond were that the N−H−O
angle was between 100° and 180° and that the distance
between nitrogen and oxygen atoms was not larger than 3.5 Å.
Figure 14. Molecular structure of the Na+ complex with L′ optimized
at the B3LYP/6-311++G(d,p) level of theory.
K
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
When two or more hydrogen bonds were formed, in 52% of the
cases the three-center bond was found for L and only 20% for
LMeCN. A similar number and type of hydrogen bonds were
observed in the crystal structures of L (one three-center
intramolecular hydrogen bond; Figure 3a,b) and LMeCN (one
two-center intramolecular hydrogen bond; Figure 3c). Dis-
tribution of the hydrogen-bond lengths obtained by MD
simulation has a maximum between 2.8 and 2.9 Å, which is in
agreement with the hydrogen bond lengths in the solid state
(Table S2 and Figure S18, Supporting Information). In
addition, the average structure of L obtained by MD simulation
agrees very well with its molecular structure in the crystal. That
can be taken as a corroboration of our choice of the force-field
parameters.
phenyl carbon atoms bound to the tert-butyl group were
practically the same, indicating the existence of C4 cone
conformation. The Na+ cation was on average coordinated by
4 phenolic and 3.3 carbonyl oxygen atoms. The latter
corresponded to the time average of the structure, with all
four carbonyl oxygen atoms oriented toward Na+ and that with
one carbonyl oxygen atom oriented toward the solvent. This
was in accordance with the crystal molecular structure of
NaLMeCN+, where the Na+ cation is heptacoordinated (Figure
4b). It should also be mentioned that the average cation−
oxygen atom distances obtained by MD simulation were similar
to those found in the crystal.
The average number of carbonyl oxygen atoms that
coordinate Na+ is greater than in the case of the LiLMeCN+
complex. That can be explained by taking into account the
difference in ionic radii of the free cations. Because Li+ is
considerably smaller than Na+ [r(Li+) = 0.76 Å; r(Na+) = 1.02
Å for coordination number 6],33b it can be reasonably expected
that the ligand binding sites will be more easily accommodated
around the latter than the former cation. Nevertheless, the
cation−calixarene interaction is weaker in the case of the
NaLMeCN+ complex (−431 kJ mol−1) compared to
LiLMeCN+.
The stabilization potential energy of the LMeCN species
relative to L is −74 kJ mol−1. If one subtracts the MeCN
vaporization internal energy (30.5 kJ mol−1)33a from this value,
the overall stabilization effect amounts to −43.5 kJ mol−1,
which is comparable with the enthalpies of MeCN inclusion in
the tert-butyl calix[4]arene cone measured by Arena et al. (−35
to −39 kJ mol−1).4c,d Such a stabilization energy indicates that
the LMeCN adduct is likely to exist in a MeCN solution of L,
1
as confirmed by H NMR spectroscopy.
In the simulation of the KL+ complex, a molecule of MeCN
entered the calixarene hydrophobic cavity after 0.5 ns. During
the simulation (50 ns), five different MeCN molecules entered/
left the calixarene cone, whereby the complex without the
MeCN molecule included existed for very short periods of time.
The average K+ coordination number was 7.9 in the case of
KLMeCN+ (3.9 carbonyl oxygens), whereas it was somewhat
lower in the KL+ complex (7.7). The higher coordination
number compared to Li+ and Na+ can be accounted for by the
reasoning similar to that described above. However, the
cation−L interaction in KLMeCN+ (−353 kJ mol−1) and
KL+ (−360 kJ mol−1) is weaker than that in the corresponding
lithium and sodium complexes. The average distances between
the cation and oxygen atoms of L obtained for KLMeCN+
species agree with those found in the crystal structure of the K+
ion complex with a similar calixarene ligand having the MeCN
molecule included in the hydrophobic basket.4e
MD Simulations of Complexes of L with Alkali-Metal
Cations in MeCN. MD simulation of the LiL+ complex in
MeCN was carried out. The complex was initially without a
MeCN molecule included and was in a flattened cone
conformation (less flattened than in the case of free L). After
62 ps of simulation, the solvent molecule entered the LiL+ cone
and distances between the opposing phenyl carbon atoms
became approximately the same. After 4 ns, the MeCN
molecule exited the LiLMeCN+ complex, leaving the cone
empty. In a very short period of time (180 ps), another solvent
molecule entered the calixarene hydrophobic cavity. During the
whole simulation (50 ns), six different MeCN molecules were
found to be included in the calixarene cone. Throughout the
periods when there was no solvent molecule bound, the
calixarene basket showed more flexibility and streamed toward a
flattened cone conformation. These findings can serve as an
additional confirmation of the 1H NMR-based conclusions
concerning the appearance of two forms of L complexes with
Li+, i.e., LiL+ and LiLMeCN+, in a MeCN solution.
When there was no MeCN molecule specifically bound in
the complex, the average coordination number of the Li+ cation
was found to be about 6. In this case, the cation was
coordinated by four phenolic oxygen atoms and on average 2.3
carbonyl oxygen atoms. On the other hand, when a MeCN
molecule was included in the hydrophobic cavity, the average
number of carbonyl oxygen atoms in the coordination sphere of
Li+ increased to 2.8, which was reflected in the energetically
more favorable Li+−L interaction in LiLMeCN+ (−507 kJ
mol−1) compared to LiL+ (−492 kJ mol−1). The slight increase
in the coordination number was likely due to the allosteric
effect of MeCN inclusion, which reduced the flexibility of the
complexed ligand.
The MD simulations showed that the energy of the MeCN−
calixarene interaction in LMeCN, LiLMeCN+, NaLMeCN+,
and KLMeCN+ species is almost the same and amounts to −50
kJ mol−1.
CONCLUSION
■
The comprehensive structural (solution and solid state),
thermodynamic, and computational studies were undertaken
in order to get a detailed insight into the reactivity of the
calix[4]arene derivative L. Complexation of this compound
with alkali-metal cations was thoroughly investigated, as was the
inclusion of the MeCN molecule in the hydrophobic cavity of
the free and complexed form of the ligand. On the basis of the
results of the thermodynamic investigations, it was concluded
that for complexation of Li+, Na+, and K+ with L in MeCN,
both enthalpic and entropic contributions to the corresponding
reaction Gibbs energies were favorable. However, in the case of
the Li+ cation, the entropic contribution was dominant, whereas
for Na+ and K+ cations, the complexation reactions were mostly
enthalpically controlled. The stability constants of the
complexes were determined by means of spectrophotometric,
conductometric, potentiometric, and microcalorimetric titra-
tions. The values obtained by different methods agreed very
Immediately after the beginning of the simulation of the
NaL+ complex, a MeCN molecule entered the calixarene basket
and remained there until the end of the simulation. This result,
along with the other studies conducted, confirmed the
1
assignation of the H NMR spectra of L and Na+ solutions at
n(Na+)/n(L) ≥ 1 condition. In other words, it corroborated the
conclusion that the observed signals corresponded to the
NaLMeCN+ species. The distances between the opposing
L
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
well, indicating their reliability. The binding affinity of L for
Na+ was found to be the highest. The selectivity of the ligand
toward the Na+ cation with respect to the others can be
expressed as the ratio of the corresponding stability constants.
From the data in Table 2, it follows that Na+/Li+ selectivity is
about 7, whereas the Na+/K+ one amounts to ≈2200.
conformation (C2 point group) of L is more favorable than the
square cone conformation (C4 point group). The structures of
the LMeCN adduct and NaL+ complex were found to be
similar to the square cone conformation of free calixarene.
However, in the case of LMeCN species, the sloping of the
MeCN molecule inside the ligand cone caused a slight
distortion from C4 symmetry. The inclined position of the
MeCN molecule in the cavity implied the possibility of its
active movement. To investigate in more detail this movability,
classical MD simulations were performed. The results indicate
that free L and the LMeCN adduct coexist in the MeCN
solution of the investigated macrocycle. The MeCN molecule
included in the calixarene hydrophobic cone has only a slight
effect on the structure of the lower-rim hydrophilic cavity and,
consequently, on its metal-ion-binding ability. That can be seen
from the average number of intramolecular NH···OC
hydrogen bonds present in L (2.1) compared to LMeCN
(1.5). In the ML+ complexes, the ligand is in the appropriate
square cone conformation to bind the MeCN molecule,
allowing it to easily enter the hydrophobic cavity. However,
the dynamics of MeCN binding by L rather strongly depends
on the cation comprising the complex. In NaL+, the MeCN
molecule remains in the cavity until the end of the simulation
(50 ns), while in the case of LiL+ and KL+, it frequently enters
and leaves the cavity. To test the hypothesis that preorganiza-
tion of the calixarene basket into the square cone conformation
is of great importance for its filling with the MeCN molecule,
we have performed a MD simulation in which the distances
between the opposing phenyl carbon atoms that are directly
connected to the tert-butyl groups were restricted to those in
the NaL+ complex. It was found that the MeCN molecule
immediately enters the calixarene cone and remains there during
most of the simulation time. From the above considerations, it
can be concluded that cation binding by L has a rather strong
allosteric effect on inclusion of the MeCN molecule in the
calixarene cone.25 That, in turn, has an important impact on the
solution thermodynamic stability of the metal-ion complexes
with the investigated macrocycle.
1
The H NMR spectrum of L in deuterated acetonitrile was
characterized by an existence of two signals for each of the
ligand protons with unequal integrals. These signals were
assigned to the free calixarene and its adduct with acetonitrile,
and the equilibrium constant for the CD3CN inclusion reaction
was calculated. The inclusion of the solvent molecule in the
ligand hydrophobic cavity in the NaL+ complex was found to be
much more favorable compared to the free ligand. That can be
explained by taking into account the fact that coordination of
the Na+ cation results in the preorganization of the calixarene
upper-rim cavity into a rigid cone conformation, which provides
its stronger interactions with the MeCN molecule. This
conclusion was in agreement with the results of molecular
simulations. Unlike NaL+, the specific interaction of LiL+ with
MeCN is not so strong (although stronger than that with
uncomplexed L). The reason probably lies in the more flexible
conformation that calixarene adopts upon binding with Li+
1
compared to Na+. The H NMR spectrum of L completely
complexed by Li+ in CD3CN showed two sets of signals
corresponding to the species with and without the MeCN
molecule included. That allowed for the determination of the
equilibrium constant for the reaction of LiLMeCN+ complex
formation from LiL+ and MeCN. By knowing this value,
together with the one corresponding to the MeCN binding
with the free ligand and the macroscopic Li+ + L binding
constant, it was possible to calculate the microscopic
equilibrium constants for the formation of LiL+ and LiLMeCN+
species (Scheme 2).
The structures of L and its complexes with Na+ and MeCN
were also explored by single-crystal X-ray diffraction, as well as
by DFT and MD computational studies. The conformations of
L and its MeCN adduct in the solid state corroborate the
existence of the specific MeCN−calixarene interaction observed
in the MeCN solution of L. Crystallization of L from a mixture
of solvents containing MeCN yielded a MeCN−calixarene
supramolecular complex where the MeCN molecule is inserted
in the calixarene cone so that the methyl group forms weak C−
H···π hydrogen bonds with the aromatic rings of the cone. The
insertion of MeCN into the calixarene hydrophobic cavity
changes its conformation from a flattened cone of free L to a
regular one in LMeCN. Binding of Na+ to L has a similar effect
on the conformation of the ligand, as can be seen from the
crystal structure of [NaL]ClO4, where the calixarene cone is of
C4 symmetry. When MeCN was added to a solution of the
sodium complex, insertion of the MeCN molecule into the
ligand cone occurred. This yielded NaLMeCN+ species with the
cone conformation virtually identical with those found in both
LMeCN and NaL+. It therefore follows that no change of the
molecular conformation is needed to allow entry of the MeCN
molecule in the NaL+ complex, whereas that is not the case for
the inclusion of MeCN in L. This is in accordance with the
more favorable inclusion of MeCN in NaL+ compared to L, as
observed in the solution.
ASSOCIATED CONTENT
■
S
* Supporting Information
X-ray crystallographic data in CIF format, ORTEP plots and
packing diagrams, experimental and calculated H and 13C
1
NMR chemical shifts of compound L, temperature dependence
1
1
of H NMR spectra of L in CDCl3 and CD3CN, results of H
NMR titrations of L with LiClO4, NaClO4 and KClO4 in
1
CD3CN and H NMR titrations of L and [NaL]ClO4 with
MeCN in CDCl3, additional results of spectrophotometric,
conductometric, and microcalorimetric titrations, and addi-
tional results of MD simulations. This material is available free
of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: vtomisic@chem.pmf.hr.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Geometry optimizations and calculation of isotropic
shieldings for L and its complexes with Na+ and the MeCN
molecule were performed using DFT at the B3LYP/6-311+
+G(d,p) level. Calculations showed that the flattened cone
■
This work was supported by the Ministry of Science, Education
and Sports of the Republic of Croatia (Projects 119-1191342-
2960, 119-1193079-3069, and 098-0982904-2912).
M
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(7) (a) Danil de Namor, A. F.; Apaza de Sueros, N.; McKervey, M.
A.; Barrett, G.; Arnaud-Neu, F.; Schwing-Weill, M. J. J. Chem. Soc.,
Chem. Commun. 1991, 21, 1546−1548. (b) Bonechi, C.; Donati, A.;
Martini, S.; Rossi, C.; Arduini, A.; Pochini, A.; Lonetti, B.; Baglioni, P.
J. Phys. Chem. B 2004, 108, 7603−7610.
REFERENCES
■
(1) (a) Gutsche, C. D. Calixarenes: An Introduction, 2nd ed.; The
Royal Society of Chemistry: Cambridge, U.K., 2008. (b) Calixarenes
2001; Asfari, Z., Bohmer, V., Harrowfield, J., Vicens, J., Eds.; Kluwer
̈
Academic Publishers: Dordrecht, The Netherlands, 2001. (c)
Calixarenes in the Nanoworld; Vicens, J., Harrowfield, J., Eds.; Springer:
Dordrecht, The Netherlands, 2001. (d) Danil de Namor, A. F.;
Cleverley, R. M.; Zapata-Ormachea, M. L. Chem. Rev. 1998, 98, 2495−
(8) (a) Danil de Namor, A. F.; Chahine, S.; Castellano, E. E.; Piro, O.
E. J. Phys. Chem. A 2005, 109, 6743−6751. (b) Danil de Namor, A. F.;
Gil, E.; Llosa Tanco, M. A.; Pacheco Tanaca, D. A.; Pulcha Salazar, L.
E.; Schulz, R. A.; Wang, J. J. Phys. Chem. 1995, 99, 16781−16785.
(c) Danil de Namor, A. F.; Zapata-Ormachea, M. L.; Hutcherson, R.
G. J. Phys. Chem. B 1999, 103, 366−371. (d) Danil de Namor, A. F.;
Chahine, S.; Kowalska, D.; Castellano, E. E.; Piro, O. E. J. Am. Chem.
Soc. 2002, 124, 12824−12836. (e) Danil de Namor, A. F.; Aparacio-
Aragon, W. B.; Goitia, M. T.; Castal, R. Supramol. Chem. 2004, 16,
423−433.
2525. (e) Bohmer, V. Angew. Chem., Int. Ed. Engl. 1995, 34, 713−745.
̈
(2) (a) Schule, D. T.; Peters, J. A.; Schatz, J. Coord. Chem. Rev. 2011,
̈
255, 2727−2745. (b) Mokhtari, B.; Pourabdollah, K.; Dalali, N. J.
Inclusion Phenom. Macrocyclic Chem. 2011, 69, 1−55. (c) Sliwa, W.;
Girek, T. J. Inclusion Phenom. Macrocyclic Chem. 2010, 66, 15−41.
(d) Creaven, B. S.; Donlon, D. F.; McGinley, J. Coord. Chem. Rev.
2009, 253, 893−962. (e) Sliwa, W. J. Inclusion Phenom. Macrocyclic
Chem. 2005, 52, 13−37. (f) Ludwig, R.; Dzung, N. T. K. Sensors 2002,
2, 397−416.
(9) Armarego, W. L. F.; Chai, C. L. L. Purification of Laboratory
Chemicals, 6th ed.; Butterworth-Heinemann (Elsevier): Oxford, U.K.,
2009.
(3) (a) Matthews, S. E.; Beer, P. D. Supramol. Chem. 2005, 17, 411−
(10) Arnaud-Neu, F.; Fanni, S.; Guerra, L.; McGregor, W.; Ziat, K.;
Schwing-Weill, M.-J.; Barrett, G.; McKervey, M. A.; Marrs, D.; Seward,
E. M. J. Chem. Soc., Perkin Trans. 2 1995, 113−118.
435. (b) Lhotak
́
, P. Top. Curr. Chem. 2005, 255, 65−95. (c) Matthews,
S. E.; Beer, P. D. In Calixarenes 2001; Asfari, Z., Bohmer,V.,
̈
Harrowfield, J., Vicens, J., Eds.; Kluwer Academic Publishers:
Dordrecht, The Netherlands, 2001; pp 421−439. (d) Hamdi, A.;
Abidi, R.; Vicens, J. J. Inclusion Phenom. Macrocyclic Chem. 2008, 60,
193−196. (e) Okunola, O. A.; Seganish, J. L.; Salimian, K. J.; Zavaliy,
P. Y.; Davis, J. T. Tetrahedron 2007, 63, 10743−10750. (f) Sun, X. H.;
Li, W.; Xia, P. F.; Luo, H.; Wei, Y.; Wong, M.; Cheng, Y.; Shuang, S. J.
Org. Chem. 2007, 72, 2419−2426.
(11) (a) Gampp, H.; Maeder, M.; Meyer, C. J.; Zuberbuhler, A. D.
̈
Talanta 1985, 32, 95−101. (b) Gampp, H.; Maeder, M.; Meyer, C. J.;
Zuberbuhler, A. D. Talanta 1985, 32, 251−264. (c) Maeder, M.;
̈
Zuberbuhler, A. D. Anal. Chem. 1990, 62, 2220−2224.
̈
(12) Gans, P.; Sabatini, A.; Vacca, A. Talanta 1996, 43, 1739−1753.
(13) Briggner, L. E.; Wadso, I. J. Biochem. Biophys. Methods 1991, 22,
̈
101−118.
(4) (a) Arduini, A.; Pochini, A.; Secchi, A.; Ugozzoli, F. In Calixarenes
(14) Oxford Diffraction. CrysAlis CCD and CrysAlis RED, version
1.170; Oxford Diffraction Ltd.: Wroclaw, Poland, 2003.
(15) Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112−122.
(16) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837−838.
(17) Allen, F. H. Acta Crystallogr., Sect. B 2002, 58, 380−388.
2001; Asfari, Z., Bohmer, V., Harrowfield, J., Vicens, J., Eds.; Kluwer
̈
Academic Publishers: Dordrecht, The Netherlands, 2001; pp 457−
475. (b) Curí
̌ ́ ́
nova, P.; Pojarova, M.; Budka, J.; Lang, K.; Stibor, I.;
Lhotak
́
, P. Tetrahedron 2010, 66, 8047−8050. (c) Arena, G.; Contino,
A.; Longo, E.; Spoto, G.; Arduini, A.; Pochini, A.; Secchi, A.; Massera,
C.; Ugozzolli, F. New J. Chem. 2004, 28, 56−61. (d) Arena, G.;
Continoa, A.; Magrì, A.; Sciottoa, D.; Arduini, A.; Pochini, A.; Secchi,
A. Supramol. Chem. 2001, 13, 379−386. (e) Arduini, A.; Giorgi, G.;
Pochini, A.; Secchi, A.; Ugozzolli, F. Tetrahedron 2001, 57, 2411−
2417. (f) Smirnov, S.; Sidorov, V.; Pinkhassik, V.; Havlicek, J.; Stibor,
I. Supramol. Chem. 1997, 8, 187−196. (g) Wang, J.; Bodige, S. G.;
Watson, W. H.; Gutsche, C. D. J. Org. Chem. 2000, 65, 8260−8263.
(5) (a) Nomura, E.; Takagaki, M.; Nakaoka, C.; Uchida, M.;
(18) Stilinovic,
441−445.
́
V.; Kaitner, B. Acta Crystallogr., Sect. A 2010, 66,
(19) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J.
B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J.
J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;
Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision
D.01; Gaussian, Inc.: Wallingford, CT, 2004.
Taniguchi, H. J. Org. Chem. 1999, 64, 3151−3156. (b) Frkanec, L.;
̌
Visn
̌
jevac, A.; Kojic-
́
Prodic,
́
B.; Zinic,
́
M. Chem.Eur. J. 2000, 6, 442−
453. (c) Cho, Y. L.; Rudkevich, D. M.; Shivanyuk, A.; Rissanen, K.;
Rebek, J., Jr. Chem.Eur. J. 2000, 6, 3788−3796. (d) Mareque Rivas,
J.; Schwalbe, C. H.; Lippard, S. J. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,
9478−9483. (e) Stibor, I.; Ruzick
̌
ova, M.; Krat
́
ky, R.; Vindys,
̌
M.;
̌
́
Havlícek, J.; Pinkhassik, E.; Lhotak, P.; Mustafina, A. R.; Morozova, Y.
̌
́
E.; Kazakova, E. Kh.; Gubskaya, V. P. Collect. Czech. Chem. Commun.
2001, 66, 641−662. (f) No, K.; Lee, J. H.; Yang, S. H.; Yu, S. H.; Cho,
M. H.; Kim, M. J.; Kim, J. S. J. Org. Chem. 2002, 67, 3165−3168.
(g) No, K.; Lee, J. H.; Yang, S. H.; Noh, K. H.; Lee, S. W.; Kim, J. S.
Tetrahedron 2003, 59, 2403−2407. (h) Xu, H.; Kinsel, G. R.; Zhang, J.;
̌
(20) (a) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55,
̌ ́ ́
Li, M. Tetrahedron 2003, 59, 5837−5848. (i) Tomisi V.; Galic, N.;
c,
̌
̌ ́
Bertosa, B.; Frkanec, L.; Simeon, V.; Zinic, M. J. Inclusion Phenom.
117−129. (b) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027−
2094.
Macrocyclic Chem. 2005, 53, 263−268. (j) Baldini, L.; Sansone, F.;
Scaravelli, F.; Casnati, A.; Ungaro, R. Supramol. Chem. 2010, 22, 776−
(21) (a) Cances
(b) Cances, E.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1997, 107,
3032−3041. (c) Mennucci, B.; Cances, E.; Tomasi, J. J. Phys. Chem. B
̀
, E.; Mennucci, B. J. Math. Chem. 1998, 23, 309−326.
̀
788. (k) Pozar, J.; Preoca
̌
nin, T.; Frkanec, L.; Tomisi
N.; Buric, N.; Tomas,
́
V. Supramol. Chem. 2011, 23, 389−397.
̌
c,
́
V. J. Solution
̌
̀
Chem. 2010, 39, 835−848. (l) Galic,
́
́
̌
R.;
1997, 101, 10506−10517. (d) Mennucci, B.; Cammi, R.; Tomasi, J. J.
Chem. Phys. 1998, 109, 2798−2807.
Frkanec, L.; Tomisic,
̌
(6) (a) Danil de Namor, A. F.; Hutcherson, R. G.; Sueros Velarde, F.
J.; Zapata-Ormachea, M. L.; Pulcha Salazar, L. E.; Al Jammaz, I.; Al
Rawi, N. Pure Appl. Chem. 1998, 70, 769−778. (b) Danil de Namor, A.
(22) (a) Cammi, R.; Mennucci, B.; Tomasi, J. J. Chem. Phys. 1999,
110, 7627−7638. (b) Cammi, R. J. Chem. Phys. 1998, 109, 3185−3196.
(23) Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. J. Chem.
Theory Comput. 2008, 4, 435−447.
(24) Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J. J. Am. Chem.
Soc. 1996, 118, 11225−11236.
(25) de Araujo, A. S.; Piro, O. E.; Castellano, E. E.; Danil de Namor,
A. F. J. Phys. Chem. A 2008, 112, 11885−11894.
F. In Calixarenes 2001; Asfari, Z., Bohmer, V., Harrowfield, J., Vicens,
̈
J., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands,
2001; pp 346−364. (c) Arnaud-Neu, F.; McKervey, M. A.; Schwing-
Weill, M.-J. In Calixarenes 2001; Asfari, Z., Bohmer, V., Harrowfield, J.,
Vicens, J., Eds.; Kluwer Academic Publishers: Dordrecht, The
Netherlands, 2001; pp 385−406.
̈
N
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(26) Swope, W. C.; Andersen, H. C.; Berens, P. H.; Wilson, K. R. J. J.
Chem. Phys. 1982, 76, 637−649.
(27) (a) Darden, T.; York, D.; Pedersen, L. J. Chem. Phys. 1993, 98,
10089−10092. (b) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden,
T.; Lee, H.; Pedersen, L. G. J. Chem. Phys. 1995, 103, 8577−8593.
́
(28) (a) Nose, S. Mol. Phys. 1984, 52, 255−268. (b) Hoover, W. G.
Phys. Rev. A 1985, 31, 1695−1697.
(29) Martyna, G. J.; Tuckerman, M. E.; Tobias, D. J.; Klein, M. L.
Mol. Phys. 1996, 87, 1117−1157.
(30) Ugozzoli, F.; Andreetti, G. J. Inclusion Phenom. Mol. Recognit.
Chem. 1992, 13, 337−348.
(31) Backman, P.; Bastos, M.; Hallen
Biochem. Biophys. Methods 1994, 28, 85−100.
́
, D.; Lonnbro, P.; Wadso, I. J.
̈
̈
̈
(32) Guilbaud, P.; Varnek, A.; Wipff, G. J. Am. Chem. Soc. 1993, 115,
8298−8312.
(33) CRC Handbook of Chemistry and Physics, 87th ed.; Lide, D. R.,
Ed.; CRC Press: Boca Raton, FL, 2006; (a) pp 6−98; (b) pp 12-11
and 12-12.
NOTE ADDED AFTER ASAP PUBLICATION
■
Due to a production error Table 2 was incorrect in the version
published ASAP May 9, 2012; the correct version reposted May
11, 2012.
O
dx.doi.org/10.1021/ic300474s | Inorg. Chem. XXXX, XXX, XXX−XXX