Macrocyclic ligands bearing two 3-(hydroxy)-2-pyridinone moieties as side-arms. Conformational studies, synthesis, crystal structure, and alkali and alkaline earth complex formation

Article abstract of DOI:10.1039/b411165f

The synthesis and characterization of the new tetraazamacrocycle 4(N),10(N)-bis[2-(3-hydroxy-2-oxo-2H-pyridin-1-yl)ethyl]-1,7-dimethyl-1,4,7, 10-tetraazacyclododecane (L1) is reported. L1 shows two 3-(hydroxy)-1- (carbonylmethylen)-2(H)-pyridinone (HPO) m

Full text of DOI:10.1039/b411165f

View Article Online / Journal Homepage / Table of Contents for this issue
P A P E R
Macrocyclic ligands bearing two 3-(Hydroxy)-2-pyridinone
moieties as side-arms. Conformational studies, synthesis, crystal
structure, and alkali and alkaline earth complex formation
Gianluca Ambrosi,a Paolo Dapporto,b Mauro Formica,a Vieri Fusi,*a Luca Giorgi,a Annalisa
Guerri,b Simone Lucarini,c Mauro Micheloni,*a Paola Paoli,b Roberto Pontellini,a Patrizia
Rossib and Giovanni Zappiac
a Institute of Chemical Sciences, University of Urbino, P.za Rinascimento 6, I-61029 Urbino,
Italy. E-mail: vieri@chim.uniurb.it
b Department of Energy Engineering ‘Sergio Stecco’, University of Florence, Via S. Marta 3,
I-50139 Florence, Italy
c Institute of Pharmaceutical Chemistry, University of Urbino, P.za Rinascimento 6, I-61029
Urbino, Italy
Received (in Durham, UK) 21st July 2004, Accepted 17th August 2004
First published as an Advance Article on the web 18th October 2004
The synthesis and characterization of the new tetraazamacrocycle 4(N),10(N)-bis[2-(3-hydroxy-2-oxo-2H-
pyridin-1-yl)ethyl]-1,7-dimethyl-1,4,7,10-tetraazacyclododecane (L1) is reported. L1 shows two
3-(hydroxy)-1-(carbonylmethylen)-2(1H)-pyridinone (HPO) moieties linked to the macrocyclic base
4,7-dimethyl-1,4,7,10-tetrazacyclododecane, as side-arms. The acid–base and coordination properties towards
alkali, alkaline earth ions of L1 together with those of the structurally similar ligand 4-(N),10-(N)-bis[2-(3-
hydroxy-2-oxo-2H-pyridin-1-yl)acetamido]-1,7-dimethyl-1,4,7,10-tetraazacyclododecane (L2) are reported.
L2 binds all the metal ions investigated forming stable mono-nuclear complexes, while L1 binds only some
alkaline earth ions with stability constants lower than those of L2. Both compounds show selectivity towards
the Mg(II) ion with respect to the alkaline earth series. The binding area is formed by the four converging
oxygen atoms of the two HPO groups. L2 appears to be a more efficient ligand in the coordination of hard
metal ions because the two HPO binding groups are forced by the macrocyclic skeleton to stay on the same
part of the macrocyclic ring, while this conformation is unfavourable for L1. These data are supported by
molecular dynamics (MD) simulations performed on both ligands and by the crystal structure of the
[H2L1](ClO4)2 species which shows the two side-arms displaced opposite to the macrocyclic ring. The effects
of pH and metal ion coordination on the UV-Vis absorption and fluorescence emission properties of both
ligands were investigated, highlighting that the optical properties changes in acid–base reactions as well as in
the coordination of metal ions.
moieties had been driven to cooperate in the binding of organic
Introduction
guests by the macrocyclic skeleton of the 1,4-dimethyl-1,4,7,
10-tetrazacyclododecane (Me2[12]aneN4).17a
Molecular modeling highlighted the fundamental role of the
The properties of multifunctional ligands suitable to interact
with guests having different charge and nature are a challen-
ging field of study. For this purpose, macrocyclic molecules
show countless possibilities; these ligands and their complexes
are used as selective binders of metal ions, simple models for
biologically relevant systems, hosts for charged and uncharged
guests, catalysts, NMR contrast agents, photochemical de-
vices, selective extractants, analytical reagents, etc.1–8
A synthetic strategy which can widen the potential applica-
tions of macrocycles involves attaching side-arms on the
macrocyclic skeleton. In this way it is possible to combine
different chemical properties, exploiting the binding properties
of both macrocyclic and molecular fragments connected as
side-arms.9–12 Molecules containing many 3-hydroxy-2(1H)-
pyridinone (HPO) groups in their skeleton are chelating agents
that have shown peculiar binding and sequestering properties
towards metal cations with a high charge.13–16 Moreover, the
HPO group shows atoms able to form hydrogen bonds with
other species. This capability can be pH-modulated in solution,
producing donor or acceptor sites of hydrogen bonds.17 Re-
cently, we reported the coordination properties of ligand L2
(Fig. 1) towards ammonium guests in which two attached HPO
two N–C O groups, which link each HPO moiety to the
Q
twelve membered ring, in stiffening and preorganizing the host.
In fact, the two amide moieties force the two side arms to
occupy the same region with respect to the macrocycle leading
to the formation of an electron rich area able to recognize, via
hydrogen-bond, guests difficult to bind in aqueous solution,
such as ammonium cations. These results encouraged us to
extend the study with L2 addressing the binding of hard metal
cations, which are also difficult to bind in aqueous solution;
this investigation was part of the present study.
However a question still remains open, that is, how much
does the binding ability of L2 depend on the nature of the link
between the HPO side arms and the macrocyclic framework,
and therefore on the rigid topology of the host? Or does the
chelating nature of HPO suffice to account for the binding
properties of L2? By exploiting the molecular modelling tools
we decided to focus on this aspect by studying L1 (Fig. 1)
where the linkage between each HPO group and the
Me2[12]aneN4 moiety occurs by means of an amine N–CH2
function. The results obtained reveal that the hosts differ both
T h i s j o u r n a l i s & T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 4
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7
1359

View Article Online
Fig. 1 Schematic draws of ligands L1 and L2.
Fig. 2 Proposed models for: (a) the U shaped and (b) the S shaped
anion Hꢀ1Lꢀ.
in their overall molecular shape and in terms of flexibility. On
this basis L1 should be the ideal probe to test the interplay
between the host topology and the nature of the coordinating
moiety (HPO) in binding small charged species such as hard
metal cations. Thus we decided to proceed with the synthesis of
L1 (here reported) and the study of its acid–base and binding
properties which were compared with those of L2.
The binding efficiency of L1 and L2 was tested on the
families of alkali and alkaline earth ions, which play a relevant
role in biological systems and in medicine. A detailed under-
standing of the modes of complexation of these cations is
therefore of great importance for selection of optimal ligands
for example, to produce better sensors.
preferred ones (B20 kcal molꢀ1 more stable). Conversely
H
ꢀ1L1ꢀ was shown to be significantly flexible, allowing the
U - S conversion, being the preferred conformation the S
shaped. Thus not only is the dynamics behaviour of L1 and L2
appreciably different in terms of flexibility, the hosts also differ
in their preferred molecular topology.
Synthesis
Complexation equilibria in aqueous solution with both
ligands were studied by potentiometric and NMR experiments.
Given the optical properties shown by the HPO fragment, it
was hypothesized that both ligands could be attractive as
chemosensors, i.e., as molecular systems capable of binding
guests and at the same time signalling the interaction. Thus the
fluorescence and UV-Vis features of L1 and L2 were monitored
with and without the metal ions.
Scheme 1 outlines the synthesis of the new multidentate ligand
L1, which was based on the coupling of the alkyl-halide
reagent 3 with the 1,7-dimethyl-1,4,7,10-tetraazacyclodode-
cane 4, followed by the removal of the hydroxyl function
protection.
Reduction of the known carboxylic acid 1 with BH3/THF
afforded the corresponding alcohol 215b in 90% yield; 2 was
treated with TsCl/Et3N to produce the chloride derivative 3 in
high yield (86%). Treatment of 3 (2 equiv.) with the
1,7-dimethyl-1,4,7,10-tetraazacyclododecane 4 (1 equiv.) in
dry CH3CN in the presence of K2CO3 (4 equiv.) gave the
protected ligand 5 (40%). Finally, hydrogenation of 5 in the
presence of 10% Pd/C afforded the free ligand L1 in almost
quantitative yield. The compound was further purified as the
diperchlorate salt.
Results and discussion
Preliminary modeling studies
Ligand L1 appears to be an ideal candidate to assess, with
respect to L2, how much the rigidity of the host determines its
binding properties, given that the stiffness of the latter has to be
ascribed to its amide functions.17a The modeling approach,
devised to gain information on the molecular topology, based
on the conformational degree of freedom and dynamics of L1
with respect to L2, had as the ultimate goal the evaluation of
the coordinating ability of both ligands towards the hard metal
ions (vide infra). Thus at first our attention was devoted to the
mono-negative species of L1 (and L2 for comparative pur-
poses) whose conformational space was investigated by means
of molecular dynamics runs performed in vacuo, to compare
the rigidity of the hosts. Two main reasons led us to choose
H
ꢀ1Lꢀ as the chief species for our preliminary tests:
(i) the electrostatic nature of the metal ion-ligand interacꢀtion
which favours overall negatively charged ligands (i.e. Hꢀ1L or
H
2ꢀ
L
)
ꢀ2
(ii) the fact that in the Hꢀ1L2ꢀ species, fully formed at pH ¼
10, the two HPO groups are both present in solution in their
deprotonated form.17a
Two different starting conformations were considered for
each anion (Hꢀ1Lꢀ): the first one (U shaped) featuring the two
side arms staying on the same part with respect to the
Me2[12]aneN4 moiety (Fig. 2a), almost facing each other, while
in the other one (S in shape) the two HPO arms are pointing in
the opposite direction with respect to the mean plane contain-
ing the twelve membered ring (Fig. 2b). As expected, Hꢀ1L2ꢀ
appears quite rigid and no U - S conversion takes place,
conformations with the two side arms almost facing (U) are the
Scheme 1 Synthesis of L1.
1360
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7

View Article Online
.
Solid state structure of [H2L1] (ClO4)2
a careful examination of analogous structures deposited in the
Cambridge Structural Database (CSD, v 5.25)21 showed in all
the [3333] diprotonated rings that the four N–X hexocyclic
bonds have the same orientation (namely ‘‘up’’) with respect to
the best plane through the macrocycle. In each species the
values of the four distances between the adjacent nitrogen
atoms are very similar and not greater than 3.0 A. As a
consequence the trans-diprotonated cation is stabilised by the
N–Hꢁ ꢁ ꢁN interactions occurring within the ring cavity.22 The
[H2L1]21 cation is the first example of an up-up-down-down
orientation of the hexocyclic carbon atoms on N(2) and N(3)
and their symmetry-related counterpart, with respect to the
flattened twelve-membered ring. This orientation for the four
nitrogen substituents implies that the number of potential
intra-annular stabilising H-bond interactions is halved with
respect to the usual ‘‘all up’’ disposition of the N–X bonds. In
addition the distances between adjacent nitrogen atoms are
significantly larger than those observed for the usual [3333] C
corners conformation (3.19 and 3.28 A, for the up-up and up-
down pairs, respectively). In our case we can hypothesized that
the stabilisation of the biprotonated species is achieved by a
hydrogen bond involving the oxygen atom O(1) of the pyridi-
The diprotonated ligand possesses a symmetry centre; as a
consequence, the unit cell contains one (H2L1)21 cation and
two symmetry-related perchlorate anions (see Fig. 3).
The collected crystallographic data did not allow location of
the position of the acidic hydrogen atoms in the Fourier
difference map, however they can be easily localized by the
following considerations:
(1) the acidic hydrogens must be bound to two nitrogen
atoms of the (Me2[12]aneN4).
(2) due to the presence of the symmetry centre they are
bound to opposite nitrogen atoms, i.e. N(2), N(2)0 or N(3),
N(3)0 (0¼ ꢀx þ 1, ꢀy, ꢀz).
Steric hindrance allows us exclude the first possibility, in fact
if the two protonation hydrogens were bound to N(2) and
N(2)0 they would be too near to the hydrogen atom H(11a)
bound to the carbon atom C(11). As a consequence the two
acidic hydrogens must be bound to N(3) and N(3)0. The short
distance between N(3) and O(1)0 [2.72(1) A] nicely supports our
hypothesis. Finally this latter interaction could explain the
overall shape taken by the ligand (vide infra).
The hydrogen atom bound to O(2) was not found in the
Fourier difference map. However given the short distance
between O(2) and perchlorate oxygen atom O(4) [2.847(9) A],
the presence of a strong hydrogen bond18 between the hydro-
gen bound to O(2) and the oxygen atom O(4) may be inferred.
Due to the presence of these interaction between [H2L1]21 and
the perchlorate anions, the crystal lattice is made up of ionic
couples.
Thanks to the syn-clinal19 or gauche conformation shown by
the dihedral angles governing the orientation of the side arms,
each pyridinone moiety forms a sort of top on the opposite
sides of the 12-membered ring. As a consequence, the plane
containing the pyridinone and that defined by the twelve atoms
of the macrocycle form an angle of 35.3(3)1. The resulting
overall shape of the cation strictly resembles that outlined by
the MD simulations, and therefore appears to be a peculiar
feature of this host, irrespective of the degree of protonation
(namely, Hꢀ1L1ꢀ vs. H2L121).
0
none moiety [O(1)ꢁ ꢁ ꢁN(3)0 ¼ 2.72(1) A, ¼ ꢀx þ 1, ꢀy, ꢀz].
Solution studies
Basicity
Table 1 summarizes the basicity constants of L1 potentiome-
trically determined in 0.15 M NMe4Cl aqueous solution at
298.1 K; the basicity constants of L2 are reported for compar-
ison. The neutral species L1 behaves as a triprotic base and as a
monoprotic acid under the experimental conditions used. As
shown in Table 1, it can be present in solution as anionic
species Hꢀ1L1ꢀ indicating the permanence of one acidic hy-
drogen on the ligand at least at pH lower than 11. In fact,
taking into account that each HPO group shows an acidic
proton, the Hꢀ2L12ꢀ species could theoretically be achieved in
alkaline solution. This was demonstrated by the basicity
behavior of L2 (see Table 1) in which the species Hꢀ2L22ꢀ
,
which has lost all of the acidic hydrogen atoms, was detected.
In other words, Hꢀ2L12ꢀ is a stronger base than the same
species of L2, and it was not possible to evaluate its protona-
tion constant, under the experimental conditions used. This is
probably due to the higher number of protonation sites in L1;
in addition, MD simulations pointed out that, thanks to their
flexibility, the HPO side arms stabilize the acidic proton in
Given the sequence of the dihedral angles, the twelve-mem-
bered ring has the usual [3333] C corners conformation,20
which, however, deserves some additional comment. In fact,
H
from 1.61 to 2.88 A (mean value 1.86 A)).
ꢀ1L1ꢀ by N–Hꢁ ꢁ ꢁO interactions (the Hꢁ ꢁ ꢁO distance ranges
Analysis of the protonation constants starting from the
anionic species Hꢀ1L1ꢀ revealed that L1 behaves as a rather
strong base in the addition of three protons to the Hꢀ1L1ꢀ
species, with protonation constant values ranging from 9.70 to
7.74 logarithmic units, but with a sharp decrease in the last
Table 1 Logarithm of the protonation constants of L1 and L2
determined by means of potentiometric measurements in 0.15 M
NMe4Cl aqueous solutions at 298.1 K
log K
Reaction
L1
L2
2ꢀ
H
H
L
ꢀ2
þ H1 ¼ Hꢀ1Lꢀ
412
10.79(1)b
9.08(1)
8.33(1)
1.86(2)
ꢀ1Lꢀ þ H1 ¼ L
9.70(1)a
8.39(1)
7.74(1)
2.30(3)
L þ H1 ¼ HL1
HL1 þ H1 ¼ H2L21
H2L21 þ H1 ¼ H3L31
a
Fig. 3 ORTEP view of the diprotonated ligand together with the
interacting perchlorate anions. Ellipsoids are drawn at 40% probabil-
ity.
Values in parentheses are the standard deviations on the last sig-
b
nificant figure. From ref. 17a.
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7
1361

View Article Online
proton addition (log K ¼ 2.30). This trend suggests an easy
availability of the protonation sites up to the HL11 species, in
agreement with the topology of the ligand. For these reasons,
the four acidic protons in the H2L121 species are probably
located far from each other, two on each HPO moiety and the
other two on non-contiguous amine groups of the macrocyclic
base in a disposition closely resembling that inferred from the
crystal structure of this species (see Fig. 3). The undetectable
value of the first constant and the value of the second proto-
nation constant, which was similar to that found for the free
tetraaza-macrocyclic base and higher than that of L2,17a,23
suggest an involvement of the amine functions of the macro-
cyclic base in both protonation steps. In the other two steps,
the deprotonated HPO groups should instead be involved in
the protonation processes.
Spectroscopic studies were performed on L1 to ascertain
which of the protonation steps the HPO moieties are involved
in. To this aim, absorption UV-Vis and emission fluorescence
spectra were recorded for L1 at different pH values. Given its
response to light stimuli, the HPO group can be considered a
suitable chromophore not only in absorbance but also in
fluorescence.
These different spectral features in the opposite field of pH
can be ascribed to the presence in solution of the neutral form
of the pyridinone moieties at low pH values and of their
deprotonated form at alkaline pH values.17 The bands at
higher energy are attributed to the p - p* while those at
lower energy to the n - p* electronic transitions of protonated
or deprotonated HPO, for both spectra.24
By monitoring the spectra from acid to basic pH and
plotting the absorbance of one of the lmax (l ¼ 258 nm in
Fig. 5a, square blocks) versus pH, it is possible to determine
during which deprotonation steps the HPO groups lose the
acidic protons by coupling the trend of the absorbance with the
distribution diagram of the species obtained by potentiometric
measurements (lines in Fig. 5). Absorbance at 258 nm was
lowest at pH o 6, and increases at higher pH, reaching a
maximum value at pH ¼ 10. Considering the distribution
diagram of the species, absorbance increases upon the appear-
ance of the HL11 species and continues to increase in the
presence of the neutral species L1, reaching a plateau at
approximately pH 4 10. Taking into account that the change
in absorbance is due to the deprotonation of the HPO groups,
this profile can be attributed to the deprotonation of one
chromophore which occurs in the formation of the HL11
species and to a second deprotonation of the other HPO which
takes place in the formation of the neutral L1 species (see
Scheme 2). For these reasons, the acidic proton in the Hꢀ1L1ꢀ
species must be located on the macrocyclic base as above
suggested. Similar behavior had also been found for compound
L2, and is reported for clarity in Fig. 5b (l ¼ 258 nm, square
blocks). In this case, the deprotonation of HPO occurs in the
formation of neutral L2 and mononegative Hꢀ1L2ꢀ species.
Excitation of acid solutions of both L1 and L2 compounds
(lexc ¼ 301 nm), gave rise to a fluorescence emission band (lem
¼ 368 nm) attributed to the HPO fluorophore; similar
spectra were recorded for both compounds. The intensity
The UV-Vis spectra show different features in acid or basic
pH fields and the spectra obtained are similar to those of L2.17a
At pH ¼ 2, where the H3L31 species is prevalent in solution
(see Fig. 4a), two lmax at 232 (e ¼ 9600 cmꢀ1molꢀ1dm3) and
301 nm (e ¼ 13700 cmꢀ1molꢀ1dm3) can be observed; these data
are preserved also at pH less than 2. At pH ¼ 12, where the
H
l
ꢀ1Lꢀ species is prevalent in solution, two new bands with
max at 258 (e ¼ 10800 cmꢀ1molꢀ1dm3) and 313 nm (e ¼ 16700
cmꢀ1molꢀ1dm3) appear (see Fig. 4a). No further changes were
observed in the spectra at more strongly alkaline pH.
Fig. 5 Fluorescence emission (J) ([L] ¼ 5.0 ꢂ 10ꢀ6 M), absorption
titration (’) ([L] ¼ 6.8 ꢂ 10ꢀ5 M), and distribution curves of the
species (line) for L ¼ L1 (a) and L ¼ L2 (b) in aqueous solution at 298.1
K in 0.15 M NMe4Cl; lexc ¼ 301 nm, lem ¼ 368 nm, absorbance at l ¼
258 nm.
Fig. 4 Absorption spectra (a), and fluorescence emission spectra (b)
of L1 at different pH values ([L1] ¼ 6.8 ꢂ 10ꢀ5 M (a) and 5 ꢂ 10ꢀ6 M
(b); lexc ¼ 301 nm, lem ¼ 368 nm).
1362
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7

View Article Online
ing that the fluorescence intensity changes reflect only the
ground states acid–base equilibrium; thus, no indication of
excited state proton transfer reaction was found.24
The trend of fluorescence emission at 368 nm versus pH
(lexc ¼ 301 nm) is reported in Fig. 5a and 5b for L1 and L2,
respectively. In the emission spectra at 368 nm versus pH, a
quenching of the emission can be observed at alkaline pH
values for both L1 and L2 (Fig. 5a and 5b, circle blocks).
Taking into account the absorption data, this behavior can be
ascribed to the deprotonation of the HPO groups, the effect of
which is to dramatically decrease the emission intensity of
HPO. The first deprotonation of one of the HPO moieties also
affects the other group, giving rise to a significant drop in
emission for both compounds. This occurs during the depro-
tonation of H2L121 and of HL21 species yielding the HL11
and L2 species, respectively (see Fig. 5a and 5b). A drop in the
emission is also observed for the subsequent deprotonation of
these species suggesting, as above hypothesized, the deproto-
nation of the other HPO group.
In conclusion, ligands L1 and L2 behave as chemosensors
for H1, giving rise to the greatest fluorescence when HPO is in
its neutral form; the emission is quenched when the group loses
its acidic proton. The presence of amine functions, which in
many chemosensors can affect the fluorescence of the sensing
group, in this case seems to be unimportant.
Coordination of metal ions
The coordination properties of L1 and L2 were studied in 0.15
M NMe4Cl aqueous solution at 298.1 K. The stability con-
stants for the equilibrium reactions of the two macrocycles
with Li1, Na1, K1, Cs1 and Mg21, Ca21, Sr21 and Ba21 were
potentiometrically determined and are reported in Table 2.
Both ligands form only mononuclear species with the metal
ions examined; the main difference between L1 and L2 is that
the latter binds both alkali and alkaline earth metal ions, while
L1 binds only some of the alkaline earth metal ions. In
particular, L2 binds Li1, Na1 and K1 with the anionic
2ꢀ
H
L
ꢀ2
and Hꢀ1Lꢀ species while Cs1 is not bound; the values
of the addition constants are similar for all the three metals.
The absence of interaction with the bigger Cs1 suggests that
the binding zone is barely adaptable to the dimensions of the
cations. Considering that the coordination is essentially due to
non-covalent interactions and taking into account that under
these experimental conditions, the macrocyclic base of L2 with
only two amine donor groups displaced far from each other
cannot bind hard metal ions, the HPO moieties should be
mainly involved in the binding of the metal cations. Thus L2,
Scheme 2 Acidic hydrogen atoms location in the differently proto-
nated species of L1 and L2.
of the fluorescence emission spectra of the two compounds is
highly dependent on the protonation state of the HPO groups
(see Fig. 4b); however, the shape and position of the spectra is
substantially independent of pH in both ligands. Analogous
behavior was observed for similar fluorescent groups, suggest-
Table 2 Logarithms of the equilibrium constants (LogK) determined in 0.15 M NMe4Cl at 298.1 K for the complexation reactions of L1 and L2
with alkali and alkaline earth metal ions
L ¼ L1
Reaction
Mg21
Ca21
Sr21
Ba21
M21þHꢀ1Lꢀ¼MHꢀ2LþH1
M21 þ Hꢀ1Lꢀ ¼ MHꢀ1L1
ꢀ6.28(4)a
—
3.18(3)
—
2.57(1)
—
—
5.22(2)
L ¼ L2
Mg21
Ca21
Sr21
Ba21
2ꢀ
M21 þ H
L
¼ MHꢀ2L
7.08(2)
6.15(1)
5.88(2)
5.36(1)
4.30(3)
3.81(3)
3.96(4)
3.42(2)
ꢀ2
M21 þ Hꢀ1Lꢀ ¼ MHꢀ1L1
Li1
Na1
K1
2ꢀ
M1þH
L
¼MHꢀ2Lꢀ
2.39(4)
1.67(3)
2.66(4)
2.15(5)
2.15(5)
1.51(5)
ꢀ2
M1 þ Hꢀ1Lꢀ ¼ MHꢀ1L
Values in parentheses are the standard deviations on the last significant figure
a
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7
1363

View Article Online
whose side arms are already preorganised by the macrocyclic
skeleton (the two amide groups stiffen the whole molecule
which prefers a U shape, MD data), easily binds the alkali
metal cation. However, the four oxygen atoms provided by the
HPO moieties converge to shape a quite rigid binding area, as
already demonstrated in ammonium cation recognition.17a On
the contrary, the preferred S shape of L1, (MD data) featuring
the two side arms in opposite direction with respect to the
macrocycle, is very unfavorable for the uptake of the alkali
metal cations.
As already stated, both ligands bind alkaline earth metal
ions but, while L2 binds all the metal ions investigated, L1 does
not bind Ba21 (see Table 2). L2 binds the metals with its
2ꢀ
L
; moreover, L1 binds only Mg21 with
dianionic species H
ꢀ2
this species, producing the removal of the last acidic proton in
the presence of this alkaline earth ion. On the contrary, the
anion Hꢀ1Lꢀ is a coordinating species for both ligands.
Analysis of the stability constants of L2 revealed a trend of
the values with a peak of selectivity towards Mg21 with both
active binding species. The addition of Mg21exhibits a very
high values of Log K ¼ 7.08 and 6.15 to the Hꢀ2L22ꢀ and
tion of each M21 to the anionic Hꢀ1L2ꢀ species are similar to
those for the addition to the bi-negative Hꢀ2L22ꢀ species,
denoting that the acidic proton does not perturb the coordina-
tion of the metal ion; thus, we can safely assume that it is
located far from the binding area, i.e. in the macrocyclic base
bound to an amine group. The value of the constants, which
decreases when the dimensions of the cation increase, once
again supports the hypothesis that the two HPO shape a rigid
Fig. 6 Variation of the fluorescence emission intensity of L2 solution
([L2] ¼ 5 ꢂ 10ꢀ6 M) at pH ¼ 7.4, as a function of Mg21/L2 molar
ratio. Inset: distribution curves of the species (line) for the Mg/L2
system in molar ratio 1 : 1; fluorescence emission for the Mg/L2 system
(’) and for L2 (J) ([L2] ¼ 5 ꢂ 10ꢀ6 M) at different pH values in
aqueous solution at 298.1 K; lexc ¼ 301 nm, lem ¼ 368 nm, absorbance
at l ¼ 258 nm.
H
ꢀ1L2ꢀ species, respectively. The affinity values for the addi-
[MgHꢀ1L2]1 species in the system with M/L of 1 : 1 molar
ratio is 57%; on the contrary, the binding Hꢀ1L2ꢀ species is
substantially absent at this pH in the absence of Mg21 (see
distribution diagram of the species in Figs. 5b and 6). In other
words, the presence of magnesium contributes to deprotonate
the HPO side-arms by forming the binding species at pH values
lower than in the ligand alone. Keeping in mind that the
deprotonation of the HPO gives rise to the modifications in
the absorption and emission spectral features of the ligands,
the spectra at pH ¼ 7.4 have different absorptivity and
intensity of emission in the presence of Mg21 even though
the emission or absorption spectra of [MgHꢀ1L2]1 and
preorganized binding area formed by the fouꢀr oxygen atoms of
2ꢀ
the HPO groups in both H
L
ꢀ2
and Hꢀ1L species. Finally,
L2 shows higher stability constants with the alkaline earth than
with the alkali metal ion as expected when the coordination is
mainly due to electrostatic ligand–metal interactions.
H
ꢀ1L2ꢀ species are, for example, quite similar. This difference
Taking into account the addition of M21 to the Hꢀ1L1ꢀ
species, the value of the stability constants for the correspond-
ing [MHꢀ1L1]1 complexes once again decreases when the size
of the coordinated metal ions increases, reaching a point at
is more remarkable in the fluorescence emission spectra re-
corded at pH 7.4. In fact, while the free ligand shows a higher
emission at this pH due to the presence of the emitting species
HL21 (see Fig. 5b), in the presence of Mg21 there is the
formation of the [MgHꢀ1L2]1 species; this produces a quench-
ing of the fluorescence because the complexed species is formed
by the not emitting Hꢀ1L2ꢀ species. For this reason, the
relative intensity of the emission in the presence of Mg21 is
lower at pH 7.4 than that of the free ligand. This reduction can
also be reported as Job plot (Fig. 6) supporting the formation
of only adducts Mg/L2 of 1 : 1 stoichiometry at this pH. These
results suggest a possible application of L2 for use as chemo-
sensor for detecting and quantifying Mg21 in physiological
solution.
which it is no longer able to bind the bigger Ba21
.
Comparison of ꢀthe values of the addition constant of each
M21 to the Hꢀ1L species for both ligands revealed that L2
always has higher stability constants and thus has a greater
capability to bind hard metal ions. This behavior, as illustrated
above, can be ascribed to the higher level of preorganization of
the two HPO groups in L2 as compared to L1; in the case of
L1, this preorganization is sufficient to bind the bi-positively
charged alkaline earth Mg21, Ca21 and Sr21 but insufficient to
bind the mono-charged alkali metal ions.
Spectrophotometric experiments. UV/Vis spectra were re-
corded in the field of pH in which the complexed species of
NMR experiments. 1H and 13C NMR experiments were
performed for the M(II)/L2 system in D2O solution. The
1H NMR spectra of the HPO resonances of the Mg/L2 system
in 1 : 1 molar ratio were recorded at pH ¼ 8.5, where
the [MgHꢀ1L2]1 species is prevalent in solution (see Fig. 6),
is reported in Fig. 7 together with that of L2 in absence of
Mg21 recorded at the same pH value. Comparing the form and
the chemical shift of the spectra some changes are observable;
the resonances of the HPO groups become broad in the
presence of the metal ion, moreover, the resonances shift with
respect to those of the free ligand. In particular, the two
doublets attributed to the hydrogen atoms in 4- and 6-position
(see Fig. 7) shift upfield while the resonance of those in 5-
position shifts downfield. The upfield shift can be explained by
a deprotonation of the HPO groups due to the formation of the
[MgHꢀ1L2]1 species; the downfield shift can be attributed to
an involvement of the amide oxygen atom of each HPO group
in Mg21 stabilization. The latter hypothesis is supported by the
consideration that this resonance, in the absence of Mg21, did
not show a great shift for the several species of L2 existing in
L1 and L2 are present and compared with those of the Hꢀ1Lꢀ
2ꢀ
and H
L
species recorded under the same experimental
ꢀ2
conditions. No marked differences in absorbance or in lmax
were found by comparing the spectra of alkali and alkaline
earth complexes, thus indicating that the complexation of the
metal does not strongly perturb the optical properties of the
chromophores and thus the electronic states.
Even though the spectral profiles of the complexed species
2ꢀ
are similar to those of Hꢀ1Lꢀ and H
L
species in the
ꢀ2
absence of the metal ions, they are quite different when
comparing the spectra of the M/L system with that of L at
the same pH values. This mainly occurs in the case of Mg/L2
system, which shows the highest stability constants. The same
observations were made for the fluorescence emission spectra.
These spectral differences are due to the fact that the binding
species of the ligands appear at lower pH values in the presence
of the metal ions than in the absence of them (see Fig. 6).
In the Mg/L2 system, the main differences were observed at
physiological pH (pH ¼ 7.4). At this pH, the percentage of
1364
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7

View Article Online
point B 540 apparatus and are uncorrected. EI-MS spectra (70
eV) were recorded with a Fisons Trio 1000 spectrometer; ESI
mass spectra were recorded with a ThermoQuest LCQ Duo LC
/MS/MS spectrometer. UV absorption spectra were recorded
at 298 K with a Varian Cary-100 spectrophotometer equipped
with a temperature control unit. Fluorescence spectra were
recorded at 298 K with Varian Cary Eclipse spectrofluorimeter.
1H and 13C NMR spectra were recorded with a Bruker AC-200
instrument, operating at 200.13 and 50.33 MHz, respectively,
and equipped with a variable temperature controller. For the
spectra recorded in D2O, the peak positions are reported with
respect to HOD (4.75 ppm) for 1H NMR spectra, while
dioxane was used as reference standard in 13C NMR spectra
(d ¼ 67.4 ppm). For the spectra recorded in CDCl3, the peak
positions are reported with respect to TMS.
Fig. 7 1H NMR spectra of Mg/L2 system of 1 : 1 molar ratio and L2
recorded in aqueous solution at pH ¼ 8.5.
aqueous solution at different pH values.17a The observed
changes indicate the coordination of Mg21 by L2 mainly
perturbing the HPO groups, thus supporting the hypothesis
that the area of L2 in which the M(II) are stabilized is formed
by the four oxygen atoms of the HPO groups. Aliphatic
resonances resulted quite similar in both systems and thus they
were not reported. These data are also supported by those from
the 13C NMR experiments. The NMR experiments performed
in the presence of the other alkaline earth metal ions gave
similar results for both ligands.
Synthesis
Ligand L1 was obtained following the synthetic procedure
reported in Scheme 1. 1,4,7,10-tetrazacyclo-dodecane (4)25, 4-
(N),10-(N)-bis[2-(3-hydroxy-2-oxo-2H-pyridin-1-yl)acetamido]-
1,7-dimethyl-1,4,7,10-tetraazacyclododecane (L2)17a and of
3-(benzyloxy)-1-(carboxymethyl)-2-(1H)-pyridinone (1)15a,17a
were prepared as previously described.
3-Benzyloxy-1-(2-hydroxyethyl)-1H-pyridin-2-one (2)
BH3 (1 M in THF, 15.44 mmol) was carefully added under
nitrogen to a stirred solution of 3-(benzyloxy)-1-(carboxy-
methyl)-2-(1H)-pyridinone 1 (2 g, 7.72 mmol) in dry THF
(15 ml) at 0 1C. The reaction was stirred for 2 h at room
temperature, then quenched with ice and extracted with
CHCl3. Standard work-up afforded the alcohol 2 (1.70 g,
90%) as a white solid which was used without purification;
mp 74–75 1C (Lit.15b: 75–76 1C); 1H and 13NMR spectra as
reported.15b
Conclusions
The synthesis of the new macrocyclic ligand L1 containing two
pyridinone side-arms attached on a tetraaza macrocyclic scaf-
fold is reported. L1 differs from a previously synthesized ligand
L2 in its mode of linking the two HPO units to the macrocyclic
Me2[12]aneN4 base. While in L2 the two HPO moieties are
linked by a rigid amide N–C O group, in L1 both linkages
Q
occur by a more flexible amine N–CH2 group. This change
affects the acid–base but above all the coordination properties
of the two ligands towards hard metal ions. In fact, in aqueous
solution L2 binds all the metal ions investigated, with the
exception of Cs1, and shows a peak of selectivity towards
Mg21. Instead L1 binds only some of the alkaline earth ions
examined, while the alkali ions are not bound at all. L1 exhibits
lower stability constant values than those of L2 with the same
metal ions. In other words, L2 appears to be more preorga-
nized for the coordination of hard metal ions with respect to L1
although the latter shows more coordination sites. These
considerations are supported by coordination data in aqueous
solution, MD evidences and the crystal structure of the
[H2L1]21 species. In the latter species, the two HPO are in
opposite position with respect to the mean macrocyclic plane,
in contrast to the structure found for the [HL2]1 cation.17a
Absorbance, fluorescence and NMR experiments support
the hypothesis that both HPO groups are involved in the
stabilization of all the metal ions coordinated. The fluorescence
emission properties of both ligands are ascribed to the HPO
groups which are fluorescent as pyridinone or silent, losing the
acidic proton, as pyridinonate. Although the complexed spe-
cies do not strongly affect the spectral behavior of the sensors,
an interesting application for the Mg/L2 system can be hy-
pothesized. In fact, a decrease in fluorescence was observed in a
solution of L2 in the presence of Mg21 at physiological pH
with respect to the free ligand. This suggests that L2 could be
used as chemosensor for Mg21 in physiological media to detect
and quantify this metal ion in a biological system.
3-Benzyloxy-1-(2-chloroethyl)-1H-pyridin-2-one (3)
Et3N (2.55 ml, 18.33 mmol) followed by tosyl chloride (1.46 g,
7.64 mmol) were added to a stirred solution of 2 (1.5 g, 6.11
mmol) in dry CH2Cl2 (50 ml) at room temperature. The
reaction was stirred overnight, diluted with CH2Cl2 (100 ml),
washed with 5% aqueous NaHCO3 (3 ꢂ 50 ml) and brine (1 ꢂ
50 ml). The organic solution was dried over Na2SO4 and
concentrated in vacuo. The yellow oil was purified by flash
chromatography (cyclohexane/AcOEt 70 : 30) and to give 3 in
86% yield (1.37 g, 5.22 mmol) as a white solid; mp 84–85 1C;
IR (film): 3421, 1650, 1600, 1258 cmꢀ1; MS m/z (EI): 263 [M]1,
186 [M ꢀ Ph]1, 172 [M ꢀ CH2-Ph]1; 1H NMR: (CDCl3) d 7.48
–7.24 (m, 5H), 6.97 (dd, J ¼ 1.7 Hz, J ¼ 7.0 Hz, 1H), 6.69 (dd,
J ¼ 1.5 Hz, J ¼ 7.3 Hz, 1H), 6.05 (t, J ¼ 7.1 Hz, 1H), 5.13 (s,
2H), 4.28 (t, J ¼ 5.7 Hz, 2H), 3.93 (t, J ¼ 5.7 Hz, 2H); 13C
NMR: (CDCl3) d 158.04, 148.83, 136.14, 130.19, 128.60,
128.06, 127.34, 115.76, 104.39, 70.77, 52.23, 41.93.
4(N),10(N)-Bis[2-(3-Benzyloxy-2-oxo-2H-pyridin-1-yl)ethyl]-
1,7-dimethyl-1,4,7,10-tetraazacyclododecane (5)
A solution of 3-benzyloxy-1-(2-chloroethyl)-1H-pyridin-2-one
3 (1.44 g, 5.45 mmol) in dry CH3CN (50 ml) was added
dropwise at 0 1C to a stirred solution of 1,7-dimethyl-
1,4,7,10-tetraazacyclododecane 4 (0.55 g, 2.73 mmol) and
K2CO3 (1.50 g, 10.09 mmol) in dry CH3CN (50 ml) under
nitrogen. The reaction was stirred at room temperature for 48
h, after which the reaction mixture was filtered and the solution
concentrated in vacuo. The residue was purified by silica gel
chromatography (CH2Cl2 saturated with NH3/MeOH 90 : 10)
to give the ligand 5 (0.71 g, 40%) as a white solid; mp 116–119
1C dec.; IR (film): 3415, 1648, 1592, 1221 cmꢀ1; MS m/z (ESI):
Experimental section
General methods
IR spectra were recorded with a Shimadzu FTIR-8300 spectro-
meter. Melting points were determined with a Buchi melting
¨
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7
1365

View Article Online
677.3 ([M þ Na]1), 655.3 ([M þ 1]1); 1H NMR: (CDCl3) d 7.46
ꢀ 7.25 (m, 12H), 6.65 (dd, J ¼ 1.4 Hz, J ¼ 7.3 Hz, 2H), 5.95 (t,
J ¼ 7.3 Hz, 2H), 5.30 (s, 4H), 3.99 (t, J ¼ 5.5 Hz, 4H), 2.71 (t,
Table 3 Crystal data and structure refinement parameters for com-
pound [H2L1]21 ꢁ (ClO4
)
2
ꢀ
Empirical formula
Formula weight
Temperature (K)
Wavelength (A)
C24H40Cl2N6O12
675.52
298
J ¼ 5.5 Hz, 4H), 2.56 (m, 8H), 2.34 (m, 8H), 2.06 (m, 6H); 13
C
NMR: (CDCl3) d 158.19, 148.48, 136.48, 131.41, 128.52,
127.91, 127.32, 116.04, 103.77, 70.75, 56.36, 54.71, 53.62,
48.16, 43.68.
0.71069
Triclinic, P1
ꢀ
Crystal system, space group
Unit cell dimensions (A, 1)
a ¼ 8.917(6); a ¼ 90.08(5)
b ¼ 9.765(3); b ¼ 111.32(4)
c ¼ 10.260(5); g ¼ 106.14(5)
794.2(7)
4(N),10(N)-Bis[2-(3-Hydroxy-2-oxo-2H-pyridin-1-yl)ethyl]-1,
7-dimethyl-1,4,7,10-tetraazacyclododecane (L1)
Volume (A3)
Z, Calculated density (g cmꢀ3
Absorption coefficient (mmꢀ1
F(000)
Crystal size
)
1, 1.412
0.273
356
10% Pd/C (0.15 g) was added to a solution of protected ligand
5 (0.50 g, 0.76 mmol) in dry MeOH (30 ml) and dry CH2Cl2 (10
ml) and the mixture was hydrogenated (5 atm.) for 48 h at
room temperature. The catalyst was filtered through a Celite
pad, and the purple solution was concentrated. The ligand L1
was obtained in quantitative yield (0.35 g) as purple solid
without any further purification; mp: 125–127 1C dec.; MS
m/z (ESI): 497.3 ([M þ Na]1), 475.3 ([M þ 1]1); IR (film):
)
0.3 ꢂ 0.4 ꢂ 0.5 mm
2.57 to 25.01
2791
Full-matrix least-squares on F2
1227/141
y range for data collection (1)
Reflections collected
Refinement method
Data/parameters
Final R indices [I 4 2s(I)]
R indices (all data)
R1 ¼ 0.0927, wR2 ¼ 0.2571
R1 ¼ 0.2043, wR2 ¼ 0.3469
1
3455, 1652, 1585, 1232 cmꢀ1; H NMR: (D2O) d 6.94 (d, J ¼
6.5 Hz, 2H), 6.58 (d, J ¼ 7.5 Hz, 2H), 6.37 (t, J ¼ 7.0 Hz, 2H);
4.01 (m, 4H), 3.00–2.50 (m, 26H); 13C NMR: (D2O) d 163.14,
154.45, 124.69, 119.29, 112.20, 57.06, 56.05, 51.65, 46.76, 43.30.
[H2L1](ClO4)2. L1 (47.5 mg, 0.1 mmol) was dissolved in H2O
(10 ml) and NaClO4 (122 mg, 1 mmol) was added to the
solution and the pH adjusted to 4 with HClO4 0.1 M. Colorless
crystals suitable for X-ray analysis formed in a day at room
temperature (38 mg, 80%). Anal. for [H2L1](ClO4)2,
C24H40Cl2N6O12 (675.52): Calcd. C 42.67, H 5.97, N 12.44;
Found C 42.6, H 6.0, N 12.3%.
bound. The protonation hydrogens and the acidic protons of
the HPO moieties were not found in the Fourier synthesis and
were not introduced (see ‘‘results and discussion’’ paragraph).
Crystallographic data and refinement parameters are reported
in Table 3w.
EMF measurements
Equilibrium constants for protonation and complexation reac-
tions with L1 and L2 were determined by pH-metric measure-
ments (pH ¼ ꢀlog [H1]) in 0.15 M NMe4Cl at 298.1 ꢃ 0.1 K,
using the fully automatic equipment that has already been
described; the EMF data were acquired with the PASAT
computer program.33 The combined glass electrode was cali-
brated as a hydrogen concentration probe by titrating known
amounts of HCl with CO2-free NMe4OH solutions and deter-
mining the equivalent point by Gran’s method,34 which gives
the standard potential E1 and the ionic product of water (pKw
¼ 13.83(1) at 298.1 K in 0.15 M NMe4Cl, Kw ¼ [H1][OHꢀ]).
At least three potentiometric titrations were performed for
each system in the pH range 2–11, using different molar ratio of
M/L ranging from 1 : 1 to 3 : 1. All titrations were treated
either as single sets or as separate entities, for each system; no
significant variations were found in the values of the deter-
mined constants. The HYPERQUAD computer program was
used to process the potentiometric data.35
Molecular dynamics (MD) simulation protocol
The species studied were the mono-negative charged anions of
L1 and L2, namely: Hꢀ1L1ꢀ, and Hꢀ1L2ꢀ. The acidic proton
in both Hꢀ1Lꢀ anions was located on the macrocycle accord-
ing to the experimental evidence obtained for L2.17a
The MSI software programs InsightII and Discover version
98.026 were used for all the molecular modelling studies. The
force field used was AMBER. MD calculations were per-
formed, in vacuo at 300 K and the Verlet leapfrog algorithm,
with a time step of 1 fs, was used for integration of equations of
motion in all simulations.
Before starting the MD procedures the studied species were
optimised until the convergence criterion was met (o0.01 kcal
molꢀ1). For every MD run the system was allowed to equili-
brate for 80 ps and then a 800 ps dynamic study was performed
(snapshot conformations were saved every ps).
X-ray diffraction
Intensity data concerning the [H2L1](ClO4)2 compound were
collected on a MACH3 Nonius diffractometer using a Mo Ka
radiation (l ¼ 0.71069 A). Cell constants were determined by
using 25 reflections. Intensity data were corrected for Lorentz
and polarisation effects, an absorption correction was applied
once the structure was solved using the Walker and Stuart
method.27
The structure was solved by using the package SIR-9728 and
subsequently refined on the F2 values by the full-matrix least-
squares programs SHELXL-97.29
Acknowledgements
The authors thank the Italian Ministero dell’Istruzione dell’U-
niversita e della Ricerca (MIUR), PRIN2002 for financial
support and CRIST (Centro Interdipartimentale di Cristallo-
grafia Strutturale, University of Florence) where the X-ray
measurement was carried out.
References
Atomic scattering factors and anomalous dispersion correc-
tions for all the atoms were taken from ref. 30.
Geometrical calculations were performed by PARST97,31
and molecular plots were produced by the program OR-
TEP3.32
1
(a) C. J. Pedersen, J. Am. Chem. Soc., 1967, 89, 7017; (b) D. J.
Cram and J. M. Cram, Science, 1984, 183, 4127; (c) J.-M. Lehn,
Angew. Chem., Int. Ed. Engl., 1988, 27, 89; (d) R. M. Izatt, K.
Pawlak and J. S. Bradshaw, Chem. Rev., 1995, 95, 2529.
(a) L.-F. Lindoy, The chemistry of macrocyclic ligand complexes,
Cambridge University Press, Cambridge, 1989; (b) J.-S. Bradshaw,
2
All the non-H atoms were refined anisotropically. All the
hydrogen atoms bound to carbon atoms were introduced in
calculated positions and refined isotropically with a tempera-
ture factor depending on that of the atom to which they are
w CCDC reference number 235708. See http://www.rsc.org/suppdata/
nj/b4/b411165f/ for crystallographic data in .cif or other electronic
format.
1366
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7

View Article Online
Aza-crown Macrocycles, Wiley, New York, 1993; (c) P. Dietrich,
P. Viout and J.-M. Lehn, Macrocyclic Chemistry, VCH, Wein-
heim, 1993.
(a) N. V. Gerbeleu, V. B. Arion and J. Burgess, Template synthesis
of macrocyclic compounds, Wiley-VCH, Weinheim, 1999; (b) D.
Parker, Macrocycle synthesis: a practical approach, Oxford Uni-
versity Press, Oxford, 1996.
G. W. Gokel, Crown Ethers and Cryptands (Monographs in
Supramolecular Chemistry), ed. J. F. Stoddart, The Royal Society
of Chemistry, Cambridge, 1992.
S. Patai, A. Rapport and E. Weber, Crown Ethers and Analogs,
Wiley, New York, 1998.
(a) Y. A. Zoltov, Macrocyclic compounds in analytical chemistry,
Wiley-Interscience, New York, 1997; (b) M. Formica, V. Fusi, M.
Micheloni, R. Pontellini and P. Romani, Coord. Chem. Rev., 1999,
1; (c) E. Bardazzi, M. Ciampolini, V. Fusi, M. Micheloni, N.
Nardi, R. Pontellini and P. Romani, J. Org. Chem., 1999, 64, 1335;
(d) P. Dapporto, M. Formica, V. Fusi, M. Micheloni, P. Paoli, R.
Pontellini, P. Romani and P. Rossi, Inorg. Chem., 2000, 39, 4663.
(a) E. Kimura, T. Gotoh, S. Aoki and M. Shiro, Inorg. Chem.,
2002, 41, 3239; (b) C. Bazzicalupi, A. Bencini, E. Berni, A.
Bianchi, V. Fedi, V. Fusi, C. Giorgi, P. Paoletti and V. Valtancoli,
Inorg. Chem., 1999, 38, 4115; (c) V. Fusi, A. Llobet, J. Mahia, M.
Micheloni, P. Paoli, X. Ribas and P. Rossi, Eur. J. Inorg. Chem.,
2002, 4, 987.
(a) Calixarene in Action, eds. L. Mandolini and R. Ungaro,
Imperial College Press, London, England, 2000; (b) A. Bianchi,
K. Bowman-James and E. Garcia-Espana, Supramolecular Chem-
istry of Anions, Wiley-VCH, New York, 1997.
S. Mizukami, T. Nagano, Y. Urano, A. Odani and K. Kikuchi,
J. Am. Chem. Soc., 2002, 124, 3920.
119, 10093; (c) P. W. Durbin, B. Kullgren and K. N. Raymond,
Health Phys., 1997, 72, 865.
15 (a) M. Streater, P. D. Taylor, R. C. Heider and J. Porter, J. Med.
Chem., 1990, 33, 1749; (b) T. N. Lambert, L. Dasaradhi, V. J.
Huber and A. S. Gopalan, J. Org. Chem., 1999, 64, 6097; (c) M. A.
Santos, Coord. Chem. Rev., 2002, 228, 187; (d) V. B. Di Marco, G.
G. Bombi, A. Tapparo, A. K. Powell and C. E. Anson, J. Chem.
Soc., Dalton Trans., 1999, 2427.
16 (a) G. Xiao, D. Van der Helm, R. C. Hider and J. B. Porter,
J. Med. Chem., 1996, 100, 2345; (b) A. Shanzer, J. Libman, S.
Lifson and C. E. Felder, J. Am. Chem. Soc., 1986, 108, 7609.
17 (a) M. Formica, V. Fusi, L. Giorgi, A. Guerri, S. Lucarini, M.
Micheloni, P. Paoli, R. Pontellini, P. Rossi, G. Tarzia and G.
Zappia, New J. Chem., 2003, 27, 1575; (b) B. L. Rai, H. Khodr and
R. C. Hider, Tetrahedron, 1999, 55, 1129; (c) J. A. Berson, W. M.
Jones and S. L. F. O’Challagan, J. Am. Chem. Soc., 1956, 78, 622.
18 G. R. Desiraju and T. Steiner, The weak hydrogen bond, Oxford
University Press, 1999.
3
4
5
6
19 W. Klyne and V. Prelog, Experentia, 1960, 16, 521.
20 I. Bernal, Stereochemical and Stereophysical Behaviour of Macro-
cycles, Vol. 2, Elsevier, 1987.
7
21 F. H. Allen, Acta Cryst., 2002, B58, 380.
22 P. J. A. Ribeiro-Claro, A. M. Amado, P. M. M. Marques and J. J.
C. Teixeira-Dias, J. Chem. Soc., Perkin Trans., 1996, 2, 1161.
23 R. Hancock, P. Wade and M. Ngwenya, Inorg. Chem., 1990, 29,
1968.
24 (a) A. Testa, J. Photochem. Photobiol., A: Chem., 1992, 64, 73; (b)
G. Wenska, B. Skalski, Z. Gdaniec, R. W. Adamiak, J. Matulic-
Adamic and L. Beigelman, J. Photochem. Photobiol., A: Chem.,
2000, 133, 169–176.
25 M. Ciampolini, P. Dapporto, M. Micheloni, N. Nardi, P. Paoletti
and F. Zanobini, J. Chem. Soc., Dalton Trans., 1984, 1357.
26 Biosym/MSI, 9685 Scranton Road, San Diego, CA 92121-3752,
USA.
27 N. Walker and D. D. Stuart, Acta Crystallogr., Sect. A., 1983, 39,
158.
8
9
10 (a) P. Dapporto, V. Fusi, M. Micheloni, P. Palma, P. Paoli and R.
Pontellini, Inorg. Chim. Acta, 1998, 168, 275–276; (b) C. Bazzica-
lupi, A. Bencini, A. Bianchi, V. Fedi, V. Fusi, C. Giorgi, P.
Paoletti, L. Tei and B. Valtancoli, J. Chem. Soc., Dalton Trans.,
1999, 1101.
11 G. Xue, J. S. Bradshaw, N. K. Dalley, P. B. Savage, K. E.
Krakowiak, R. M. Izatt, L. Prodi, M. Montalti and N. Zaccher-
oni, Tetrahedron, 2001, 57, 7623.
28 A. Altomare, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, M.
C. Burla, G. Polidori and M. Camalli, J. Appl. Cryst., 1999, 32,
115.
12 G. Ambrosi, P. Dapporto, M. Formica, V. Fusi, L. Giorgi, M.
Micheloni, P. Paoli, R. Pontellini and P. Rossi, Chem. Eur. J.,
2003, 9, 800.
13 (a) J. Xu, D. W. Whisenhunt, A. C. Veeck, L. C. Uhlir and K. N.
Raymond, Inorg. Chem., 2003, 42, 2665; (b) J. Xu, B. O’Sullivan
and K. N. Raymond, Inorg. Chem., 2002, 41, 6731; (c) C. J.
Sunderland, M. Botta, S. Aime and K. N. Raymond, Inorg. Chem,
2001, 40, 6746; (d) J. Xu, E. Radkov, M. Ziegler and K. N.
Raymond, Inorg. Chem., 2001, 39, 4156.
29 G. M. Sheldrick, SHELX 97, University of Gottingen, Germany,
1997.
30 International Tables for X-ray Crystallography, Kynoch Press,
Birmingham, UK, Vol. 41974.
31 M. Nardelli, Comput. Chem., 1983, 7, 95.
32 L. J. Farrugia, J. Appl. Cryst., 1997, 30, 565.
33 M. Fontanelli and M. Micheloni, I Spanish-Italian Congr. Ther-
modynamics of Metal Complexes, Peniscola, Univ. of Valencia
Spain, June 3–6, 1990, 41.
¨
14 (a) J. Xu, B. Kullgren, P. W. Durbin and K. N. Raymond, J. Med.
Chem., 1995, 38, 2606; (b) M. Meyer, J. R. Telford, S. M. Cohen,
D. J. White, J. Xu and K. N. Raymond, J. Am. Chem. Soc., 1997,
34 (a) G. Gran, Analyst, 1952, 77, 661; (b) F. J. Rossotti and
H. Rossotti, J. Chem. Educ., 1965, 42, 375.
35 P. Gans, A. Sabatini and A. Vacca, Talanta, 1996, 43, 1739.
N e w J . C h e m . , 2 0 0 4 , 2 8 , 1 3 5 9 – 1 3 6 7
1367