Synthesis, protonation and CuII complexes of two novel isomeric pentaazacyclophane ligands: Potentiometric, DFT, kinetic and AMP recognition studies

Article abstract of DOI:10.1002/ejic.200800576

The synthesis and coordination chemistry of two novel ligands, 2,6,9,12,16-pentaaza[17]metacyclophane (L¹) and 2,6,9,12,16- pentaaza[17]paracyclophane (L²), is described. Potentiometric studies indicate that L¹ and L²

Full text of DOI:10.1002/ejic.200800576

FULL PAPER
DOI: 10.1002/ejic.200800576
Synthesis, Protonation and CuII Complexes of Two Novel Isomeric
Pentaazacyclophane Ligands: Potentiometric, DFT, Kinetic and AMP
Recognition Studies
Andrés G. Algarra,[a] Manuel G. Basallote,*[a] Raquel Belda,[b] Salvador Blasco,[b]
C. Esther Castillo,[a] José M. Llinares,[c] Enrique García-España,*[b] Laura Gil,[b]
M. Ángeles Máñez,[a] Conxa Soriano,[c] and Begoña Verdejo[b]
Keywords: Cyclophanes / Macrocyclic ligands / Copper / Kinetics / Density functional calculations
The synthesis and coordination chemistry of two novel li-
gands, 2,6,9,12,16-pentaaza[17]metacyclophane (L1) and
2,6,9,12,16-pentaaza[17]paracyclophane (L2), is described.
Potentiometric studies indicate that L1 and L2 form a variety
of mononuclear complexes the stability constants of which
reveal a change in the denticity of the ligand when moving
from L1 to L2, a behaviour that can be qualitatively explained
by the inability of the paracyclophanes to simultaneously use
both benzylic nitrogen atoms for coordination to a single
metal centre. In contrast, the formation of dinuclear hydrox-
ylated complexes is more favoured for the para L2 ligand.
nearby water molecules. The kinetics of the acid promoted
decomposition of the mono- and dinuclear CuII complexes of
both cyclophanes have also been studied. For both ligands,
dinuclear complexes convert rapidly to mononuclear species
upon addition of excess acid, the release of the first metal ion
occurring within the mixing time of the stopped-flow instru-
ment. Decomposition of the mononuclear [CuL2]2+ and
[CuHL2]3+ species occurs with the same kinetics, thus show-
ing that protonation of [CuL2]2+ occurs at an uncoordinated
amine group. In contrast, the [CuL1]2+ and [CuHL1]3+ species
show different decomposition kinetics indicating the exis-
DFT calculations have been carried out to compare the geo- tence of significant structural reorganisation upon proton-
metries and relative energies of isomeric forms of the [CuL]2+
complexes of L1 and L2 in which the cyclophane acts either
as tri- or tetradentate. The results indicate that the energy
cost associated with a change in the coordination mode of
the cyclophane from tri- to tetradentate is moderate for both
ligands so that the actual coordination mode can be deter-
mined not only by the characteristics of the first coordination
sphere but also by the specific interactions with additional
ation of the [CuL1]2+ species. The interaction of AMP with
the protonated forms of the cyclophanes and the formation
of mixed complexes in the systems Cu–L1-AMP, Cu–L2-AMP,
and Cu–L3-AMP, where L3 is the related pyridinophane con-
taining the same polyamine chain and 2,6-dimethylpyridine
as a spacer, is also reported.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
of these compounds is their ability to form metal complexes
containing coordinatively unsaturated metal sites. In each
case, the actual details of the coordination environments are
affected by different factors such as the kind of aromatic
spacer, the number of nitrogen atoms and the size and flexi-
bility of the polyamine chain.
Introduction
In the last years some of us have focused on the synthesis
and study of new families of cyclic ligands formed by link-
ing together the ends of a polyamine and an aromatic
spacer through methylene carbon atoms.[1–5] Perhaps one of
the most interesting aspects of the coordination chemistry
The role of the aromatic spacer, in particular the type of
substitution (ortho, meta or para), becomes fundamental in
the coordination chemistry of synthetic cyclophanes, espe-
cially in those with polyamine chains of small size. For ex-
ample, the rigidity of the p-xylyl unit prevents the simulta-
neous participation of both benzylic nitrogen atoms in the
coordination of a single metal ion which results in a higher
tendency for the formation of dinuclear complexes than
that exhibited by analogous ligands with o- or m-xylyl spa-
cers.[2,6] In contrast, small and medium-sized cyclophanes
with one o-xylyl spacer usually form only mononuclear
complexes. The larger the size of the polyamine, the less
drastic these changes are so that the coordination chemistry
[a] Departamento de Ciencia de Materiales e Ingeniería Met-
alúrgica y Química Inorgánica, Facultad de Ciencias, Uni-
versidad de Cádiz,
Apartado 40, Puerto Real, 11510 Cádiz, Spain
E-mail: manuel.basallote@uca.es
[b] Departamento de Química Inorgánica, Instituto de Ciencia
Molecular, Universidad de Valencia,
Apartado Correos 22085, 46071 Valencia, Spain
E-mail: enrique.garcia-es@uv.es
[c] Departamento de Química Orgánica, Facultad de Farmacia,
Universidad de Valencia,
Avda. Vicente Andrés Estellés s/n, 46100 Burjassot (Valencia),
Spain
Supporting information for this article is available on the
WWW under http://www.eurjic.org or from the author.
62
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 62–75

CuII Complexes of Pentaazacyclophane Ligands
of these isomers becomes more similar. In the present work, and the corresponding pyridine derivative L3.[19] In all cases
we decided to analyse the case of the isomeric cyclophanes the five expected protonation steps can be identified within
L1 and L2 (see Figure 1), paying special attention to their the pH range of the study (2–11) and the trend in the values
behaviour towards protonation and to their ability to form of the stepwise constants can be easily rationalised on the
mono- and dinuclear CuII complexes. In addition, the ki- basis of minimisation of electrostatic repulsion between the
netic properties of the latter complexes have been also ex- positive charges generated at the protonated amine groups.
amined in order to obtain additional information about the Thus, the new L1 cyclophane shows large values of the con-
lability of closely related metal complexes differing only in stants for the first two protonation steps whereas the third
their protonation states. In this sense, recent reports have and fourth protonation steps occur with constants with in-
shown that the different CuHxLz+ species of a given ligand termediate values and the last step occurs with a much
can decompose with different kinetics which reveals the more reduced protonation constant as a consequence of the
existence of extensive structural reorganisation of the CuII fact that the incoming proton is forced to bind to the cen-
complexes upon protonation.[7] The paper also includes tral nitrogen of the polyamine chain while being surrounded
some exploratory work on the stability of the AMP species by two protonated sites separated by ethylenic chains. These
formed by the cyclophanes and their copper complexes. The results are quite similar to those previously reported for the
nucleotide-recognition capability of protonated noncyclic analogous pyridinophane L3[19] for which the proposed pro-
and macrocyclic polyamines was discovered decades tonation sequence was confirmed by 1H and 13C NMR
ago[8–13] and it has been exploited for nucleotide sensing spectroscopy at variable pH values. For these two ligands
and separation.[14–17] The interaction of the ammonium (L1 and L3), the protonation constants are always smaller
groups in the HxLx+ species with the phosphate anions of than those corresponding to the open-chain L4 ligand. In
adenosine-5Ј-monophophate (AMP), adenosine-5Ј-di- contrast, although the results obtained for L2 are quite par-
phophate (ADP) and adenosine-5Ј-triphophate (ATP) leads allel, a greater basicity is observed for all the protonation
to the formation of HxLA(x–z)+ (Az– = nucleotide anion) steps in this ligand, probably because the para substitution
species with increased stability as a consequence of π stack- at the aromatic spacer facilitates the separation of the posi-
ing interactions with aromatic rings in the polyamine. To tive charges generated at the cyclophane upon protonation.
gain insight into the stability of this kind of species, an ini- As a consequence of this capability, the protonation con-
tial approach to the nucleotide-recognition capability of the stants for L2 become closer to those obtained for the open-
L1–L3 cyclophanes and their CuII complexes was made by chain polyamine L4.[18]
studying their interaction with AMP.
Table 1. Stepwise protonation constants for the L1 and L2 ligands.[a]
Literature values[18,19] for the protonation constants of L3 and L4
are also included for comparison.
Reaction[b]
L1
L2
L4
L3
H + L = HL
H + HL = H2L
H + H2L = H3L
H + H3L = H4L
H + H4L = H5L
log β5
9.70(2)
9.37(2)
7.81(2)
6.99(2)
2.80(4)
36.67
10.69(2)
9.66(1)
8.32(2)
7.21(2)
3.03(3)
38.91
10.55
9.89
8.69
7.55
3.55
40.23
9.65
9.32
7.62
6.62
2.86
36.07
[a] At 298.1 K in the presence of 0.15  NaClO4. The values in
parentheses correspond to the standard deviations in the last sig-
nificant figure. [b] Charges omitted for clarity.
The stability constants for the formation of Cu2+ com-
plexes with the L1 and L2 cyclophanes are included in
Table 2 together with those previously reported for L3 and
L4.[18,19] All three cyclophanes allow for the formation of
both mono- and dinuclear metal complexes whereas the
open-chain L4 ligand only forms mononuclear species. As
usually occurs in these kinds of ligands, the nuclearity of
the species formed is strongly dependent on the M/L molar
ratio so that while for L1 at a 1:1 ratio only mononuclear
species are detected throughout the whole pH range
studied, for a 2:1 ratio the dinuclear hydroxylated complex
is the only species in solution above pH 6 (Figure 2). As
expected from the change in the substitution at the aromatic
spacer, the stability of the dinuclear species increases for the
Figure 1. Ligands and abbreviations used in this work.
Results and Discussion
Equilibrium Studies on Ligand Protonation and CuII
Complexation
The stepwise protonation constants for the L1 and L2 para-substituted L2 cyclophane. For a 2:1 molar ratio sev-
ligands are included in Table 1 and compared with those eral dinuclear species can be detected which also dominate
previously reported for the open-chain counterpart L4[18] the species distribution curves above pH 6 (Figure 3).
Eur. J. Inorg. Chem. 2009, 62–75
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
63

M. G. Basallote, E. García-España et al.
FULL PAPER
Figure 2. Distribution diagram for the system Cu2+-L1. (A) [Cu+2] = [L1] = 10–3 moldm–3. (B) [Cu+2] = 2ϫ10–3 moldm–3 [L1] =
10–3 moldm–3.
Figure 3. Distribution diagram for the system Cu2+–L2. (A) [Cu+2] = [L2] = 10–3 moldm–3. (B) [Cu+2] = 2ϫ10–3 moldm–3 [L2] =
10–3 moldm–3.
Table 2. Stepwise stability constants for the formation of Cu2+ reveals that the first protonation constant for the [CuL1]2+
,
[CuL2]2+ and [CuL3]2+ complexes has values that compare
well with the third protonation step of the free ligands seen
in Table 1, a step in which there are the same charge
changes. This comparison suggests that not all the nitrogen
atoms in the ligand are involved in metal coordination, as
confirmed by the crystal structure for the case of L3 com-
mented on above. However, the formation constant of
[CuL2]2+ is much smaller than that obtained for the equiva-
lent L3 species. This fact, and the high value for the second
protonation constant for [CuL2]2+ strongly suggests that in
this case there are only three nitrogen atoms tightly bound
to the metal ion which contrasts with the tetradentate coor-
dination observed for the L3–Cu complex.[19] Therefore, p-
substitution at the aromatic ring of L2 appears to prevent
the simultaneous participation of both benzylic nitrogen
atoms in the coordination to a single metal ion, a conclu-
sion supported by several previous reports of crystal struc-
tures involving para-azacyclophanes.[2–4]
complexes with the L1 and L2 cyclophanes.[a] Literature values[18,19]
for the systems Cu2+-L3 and Cu2+-L4 are also included for compar-
ative purposes.
Reaction[b]
L1
L2
L3
L4
Cu + L = CuL
CuL + H = CuHL
CuHL + H = CuH2L
2 Cu + L = Cu2L
2 Cu + L + H2O = Cu2L(OH) + H
19.05 (1) 15.70(4) 20.44
21.28
8.86
3.39
–
–
–
7.25 (3)
10.00(4) 6.96
3.33 (2)
5.12(2)
2.75
–
–
22.00(6)
15.19(2) 20.65
6.04(3) 10.84
–
2 Cu + L + 2 H2O = Cu2L(OH)2 + 2 H 9.05 (1)
[a] At 298.1 K in the presence of 0.15  NaClO4. The values in
parentheses correspond to the standard deviations in the last sig-
nificant Figure. [b] Charges omitted for clarity.
With regards to the actual values of the stability con-
stants, the major observation made from the values in
Table 2 is that the stability of the [CuL2]2+ species is signifi-
cantly lower than that found for the same species with the
other ligands suggesting a lower coordination number of
the cyclophane in this case. The previously reported crystal
In attempts to obtain additional information about the
structure of the [CuL3]2+ complex reveals a very distorted denticity of the L1 and L2 cyclophanes in their CuII com-
octahedral coordination sphere about the metal centre so plexes by using experimental techniques different from
that the coordination number of the ligand can be best de- potentiometric data, were carried out instead EPR and UV/
scribed as being four.[19] A coordination number of four has Vis spectroscopic studies. EPR spectra recorded at the pH
also been established for the CuII complex of the acyclic values where the mononuclear species [CuHL1]3+, [CuL1]2+
,
L4 ligand.[19,20] Although deriving coordination numbers by [CuHL2]3+ and [CuL2]2+ predominate in solution (see some
taking into account only free energy terms can be mislead- representative examples in the Supporting Information) are
ing, a careful consideration of the stability constants of the in all cases very similar to each other and they do not show
different protonated and nonprotonated complex species, in any superhyperfine structure due to interaction with 14N
comparison with the protonation of the free ligands, can nuclei so that no direct information about the number of
provide some valuable information. In this sense, Table 2 coordinated nitrogen atoms could be obtained. However,
64
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 62–75

CuII Complexes of Pentaazacyclophane Ligands
since it has been reported that there is an approximate addi- above pH 7 and [Cu2L2(OH)2]2+ above pH 9. The low pKa
tivity of the influence of the number of nitrogen donors on value for the reaction of [Cu2L2]4+ to give [Cu2L2(OH)]3+
the parameters in the EPR spectra of Cu-polyamine com- (pKa = –6.82) suggests that the hydroxo ligand might be
plexes,[21] a more detailed analysis of the EPR spectra at bridging both metal centres. From the behaviour of related
100 K was made. The only species with an EPR spectrum compounds, it can be anticipated that these dinuclear spe-
significantly different from the other species is [CuHL2]3+ cies can show interesting properties in the assistance of hy-
and this shows an isotropic spectrum with a g value of 2.09 drolytic processes with different electrophilic substrates
which suggests that protonation of [CuL2]2+ leads to a such as carboxylates or phosphates,[5,22] a possibility that
change in the geometry towards a tetrahedral structure. will be explored in future work.
However, it was found that the spectra of the [CuHL1]3+
,
[CuL1]2+ and [CuL2]2+ complexes are similar in all cases
with values of gЌ = 2.08, g = 2.21 and A = 190 G. These
values compare well with those reported for Cu(en)22+, Cu-
(trien)2+ and Cu(cyclam)2+[21] but no conclusions about dif-
ferences in the denticity of the ligand in the L1 and L2 com-
DFT Calculations on the Formation of CuL2+ Species with
the L1 and L2 Ligands
The most striking result in the previous section is proba-
plexes could be established. Nevertheless, further evidence bly the large difference in the stability constants of the
for the change in the denticity of the L1 and L2 ligands in [CuL1]2+ and [CuL2]2+ complexes which strongly suggests a
their Cu complexes is provided by the electronic spectra change in the number of donor atoms used by the cy-
which show an absorption maximum centred at 570 nm in clophane in both complexes with L1 acting as tetradentate
the case of [HCuL1]+3 (ε = 194 –1 cm–1), 565 nm for and L2 as tridentate. In addition, potentiometric results do
[CuL1]2+ (ε = 191 –1 cm–1), 585 nm for [HCuL2]3+ (ε = not provide any evidence of coordination of these ligands
85 –1 cm–1), 585 nm for [CuL2]2+ (ε = 110 –1 cm–1) and through the whole set of five potential nitrogen donors in
610 nm for [Cu2L2(OH)]3+ (ε = 215 –1 cm–1). A maximum any of the species formed in solution. While the latter find-
at wavelengths close to 575 nm has been observed for com- ings can be rationalised by invoking the steric constraints
plexes with related polyamines acting as tetradentate li- imposed by the aromatic spacer, there is no apparent reason
gands with four coordinated nitrogen atoms disposed at the to think that any of these cyclophanes could not behave as a
vertices of a square around the metal ion whereas com- tetradentate ligand in their CuII complexes and we therefore
plexes with tridentate polyamines show absorption bands at think that the reasons for the tridentate behaviour of L2
larger wavelengths, close to that found for [CuL2]2+. Thus, must be related to the stability difference between species
the Cu(dien)2+ complex shows a maximum at 605–615 nm with the ligand acting as tri- and tetradentate. For this
(ε = 73–82 –1 cm–1)[20] and the H3CuL complex of reason, we carried out DFT calculations aimed at determin-
4,7,10,13-tetraazahexadecane-1,16-diamine shows its maxi- ing the geometries and relative energies of [CuL]2+ species
mum at 590 nm (ε not given in the literature).[18] Both of containing both tridentate and tetradentate L1 or L2. As
these complexes contain a polyamine coordinated in a tri- expected from the equilibrium measurements, all attempts
dentate manner and coordination of an additional amine to optimise the geometries of species with the ligand acting
group leads to a shift of the maximum to shorter wave- as pentadentate were unsuccessful because the optimisation
lengths by 15–60 nm relative to the 3  coordinated species: procedure leads to dissociation of at least one Cu–N bond.
Cu(dien)(NH3)2+ has λmax = 576 nm (ε = 84 –1 cm–1),[20] In addition, the calculations confirmed the expectations in
for the CuL2+ complex of 4,7,10,13-tetraazahexadecane- the sense that the possibility of a first coordination sphere
1,16-diamine λmax = 578 nm (ε = 178 –1 cm–1)[18] and for involving both benzylic nitrogen atoms is, in the case of L2,
the HCuL3+ and CuL2+ species with 4,7,10-triazatridecane- prevented due to geometric constraints.
1,13-diamine the values are λmax
=
573 nm (ε
=
The optimised geometries and relative energies of the
187 –1 cm–1) and λmax = 585 nm (ε = 214 –1 cm–1),[20] most stable species found for the [CuL1]2+ and [CuL2]2+
respectively. Interestingly, increased molar absorptivities complexes are shown in Figure 4 and Table 3 which include
were also observed in the tetracoordinated species (ε = 170– a set of structures that can be classified in all cases as dis-
200 –1 cm–1) with respect to the corresponding tricoor- torted square pyramidal or tetragonally distorted octahe-
dinated species (ε = 70–150 –1 cm–1), something which was dral, with Cu–N distances that fall within the 2.0–2.2 Å
attributed to distortion from an ideal square planar struc- range. This kind of distorted geometry has been found in
ture.[20]
the crystal structures of many CuII–polyamine complexes,
Another important point to consider is the formation of the experimental Cu–N distances being in all cases ca. 2.0–
dinuclear hydroxy complexes. Since these ligands do not 2.1 Å.[23,24] Another common feature of the calculated
saturate the first coordination sphere of two metal ions, the structures in Figure 4 is the existence of relatively large Cu–
metal centres must bind other ligands which should most O distances with the axially coordinated water molecules.
likely be water molecules in the absence of better entering This elongation of the axial bonds is also quite common in
ligands. However, because of the electrostatic repulsion be- the crystal structures of Cu-polyamine complexes, although
tween the metal centres, hydrolysis of coordinated water to the experimental distances tend to be smaller than those
form µ-hydroxo species is favoured, the major species in calculated by the DFT procedures.[25] However, it must be
solution being [Cu2L1(OH)2]3+ above pH 7, [Cu2L2(OH)]3+ pointed out that the DFT calculations indicate that the in-
Eur. J. Inorg. Chem. 2009, 62–75
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
65

M. G. Basallote, E. García-España et al.
FULL PAPER
Figure 4. Optimised geometries calculated for the [CuL1]2+ and [CuL2]2+ complexes with the cyclophane acting as tri- or tetradentate.
Colour code: green Cu, red O, blue N, grey C, white H.
teraction of the metal ion with the axial ligand is very weak Table 3. Relative energy (kcalmol–1) of the different species calcu-
lated for the CuII complexes with the L1 and L2 ligands (see op-
so that relatively large changes in the bond lengths lead to
small energy changes. Under these conditions, it must be
timised structures in Figure 4).
L1 (m)
Gas
phase
L2 (p)
Aqueous Gas
solution phase
expected that the actual bond lengths in solution and in
the crystals will be strongly dependent on factors such as
solvation, hydrogen bonding and crystal packing forces.
Optimised geometries calculated for the [CuL1]2+ or
[CuL2]2+ complexes with the cyclophanes acting as tri- or
tetradentate ligands are collected in Figure 4. We will first
discuss the structures of the pentacoordinated complexes
with three central nitrogens of the bridge and two water
molecules (structures m-3N-2H2Oa and p-3N-2H2Oa in
Figure 4). For both cyclophanes, the complexes display a
distorted square pyramidal coordination with the three co-
ordinated amine groups located in the basal plane and one
Species
Aqueous
solution
3N-2H2Oa + H2O 0.0
3N-2H2Ob + H2O 0.9
0.0
6.4
–2.0
1.4
6.5
15.2
4.7
9.8
0.0
0.0
1.2
5.1
–3.8
5.6
25.0
4.0
22.4
–6.0
–15.3
–20.4
5.9
14.6
–3.7
12.6
3N-3H2Oa
3N-3H2Ob
–11.6
–14.8
4N-2H2Oa + H2O 9.6
4N-2H2Ob + H2O 15.4
4N-2H2Oa·H2O
4N-2H2Ob·H2O
0.2
0.8
of the coordinated waters forming a strong hydrogen bond structures (m-3N-3H2Oa and p-3N-3H2Oa) that show a
with one of the uncoordinated benzylic nitrogen atoms. very distorted octahedral geometry with one of the axial
From these structures, a third water molecule was included water ligands placed significantly closer to the metal ion
in the calculations and located in the vacant coordination than the other (Cu–O distances of 2.48 and 3.20 Å for m-
site opposed to the apical water. The calculations lead to 3N-3H2Oa, and 2.54 and 2.84 Å for p-3N-3H2Oa). As in
66
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 62–75

CuII Complexes of Pentaazacyclophane Ligands
the previous structures, the water in the equatorial plane
With regards to the relative energies of the optimised
forms a strong hydrogen bond with one benzylic nitrogen. geometries containing tri- and tetradentate cyclophanes, the
As expected from its weak interaction with the metal ion, results of the DFT calculations indicate in all cases a higher
the stabilisation associated with the third water molecule is stability of the structures with the tridentate cyclophane,
not large, so that the structures m-3N-3H2Oa and p-3N- the energy difference between the 3N-3H2O and 4N-2H2O
3H2Oa are only 2.0 and –5.1 kcalmol–1 more stable in structured (plus a free additional H2O) being 8.5 (L1) or
aqueous solution than a separate water molecule and m- 0.5 (L2) kcalmol–1 in aqueous solution (20.0 and
3N-2H2Oa or p-3N-2H2Oa, respectively (11.6 and 21.2 kcalmol–1 in the gas phase, respectively) if the a series
15.3 kcalmol–1 in the gas phase). These differences are ex- is considered in both cases. These differences are sure to be
pected to be even smaller in real systems because of the overestimates because of the possibility of hydrogen bond-
possibility of hydrogen bonding between the third water ing between the third water molecule and the 4N-2H2O sys-
and the m-3N-2H2Oa or p-3N-2H2Oa species.
tems and, for this reason, the systems were optimised whilst
Figure 4 also includes the optimised geometries for allowing for this interaction. The resultant geometries are
pentacoordinate complexes with tridentate cyclophane but also included in Figure 4 and labelled as m- or p-4N-
with a coordination environment that includes one of the 2H2O·H2O. The species with tridentate cyclophanes are still
benzylic nitrogen donors. These species are named follow- more stable even when this additional interaction is consid-
ing the procedure in the previous paragraph except that ered in the tetradentate systems, although the energy differ-
they are labelled as b instead of a. Their optimised struc- ence is reduced. A similar conclusion can be drawn by com-
tures can be described in a parallel way to that used for the paring the energies of the 3N-2H2O and 4N-2H2O pairs
a family and the relative energies are also included in of structures which also indicates a higher stability of the
Table 3. It is interesting to note that the species in the b tridentate form.
family are less stable than their analogues with the other
coordination mode (a series) in the case of the L1 complexes tions that the energy difference between the different coor-
but they are more stable in the case of L2.
dination modes is small for both cyclophanes and the actual
In any case, it must be concluded from the DFT calcula-
Finally, Figure 4 includes the optimised geometries for values in real systems will be strongly dependent on the
species containing tetradentate L1 and L2. The only way of extent of hydrogen bonding with additional water mole-
obtaining stable structures with this coordination was by cules, something not considered in the calculations. From
keeping one of the benzylic nitrogen atoms uncoordinated, this point of view, both coordination modes are expected
all attempts with all the nitrogen atoms coordinated except to coexist in solution. Moreover, since there are no large
one of the central amine groups being unsuccessful. For this differences between the results obtained for the L1 and L2
coordination, two different geometries close in energy could ligands, it must also be concluded that the experimental ob-
be optimised for each of the cyclophanes. In the structures servation of tridentate coordination in [CuL2]+2 and tetra-
of m-4N-2H2O-a and p-4N-2H2O-a, one of the water mole- dentate coordination in [CuL2]+2 cannot be attributed prin-
cules is displaced from the hypothetical sixth coordination cipally to changes in the first coordination sphere but must
site and forms a hydrogen bond with one of the coordinated be better related to differences in the network of hydrogen
amines in the equatorial plane. The distance from this water bonds formed with uncoordinated water molecules. At this
molecule to the metal centre is so large (4.0 Å for L1 and point, it is important to remember that the analysis of
3.9 Å for L2) and the H2O–Cu–OH2 angle deviates so much potentiometric results only provides direct information on
from 180° (actual values of 148.5 and 156.7° for L1 and L2, the stoichiometry, CuϪLϪH in the present case, and the
respectively) that the structures can be described as being stability of the different species but it does not allow a dis-
pentacoordinate. However, these structures are not far in tinction to be made between isomeric species with the same
energy from the alternative m-4N-2H2O-b and p-4N-2H2O- stoichiometry, i.e. the stability constant derived for a species
b structures which show coordination environments with such as [CuL]2+ does not discriminate between the possible
the four nitrogen atoms in the equatorial plane, one axial relative contributions of microscopic species with the same
water located at 2.24 Å (L1) or 2.16 Å (L2) and a weakly stoichiometry but differing in the denticity of the ligand or
interacting water molecule located at 3.41 Å (L1) or 3.70 Å in the number of coordinated water molecules. The results
(L2). The H2O–Cu–OH2 angle is now closer to 180° (actual of the present DFT studies indicate that mixtures of species
values of 163.2 and 157.5° for L1 and L2, respectively). The with different structures and even with a different denticity
energy difference between the a-b pairs of structures is sig- of the ligand can be formed in solution and that the relative
nificantly different for both ligands, those labelled as a be- amounts of each species are strongly dependent not only
ing more stable by 8.7 (L1) or 19.4 (L2) kcalmol–1 in solu- on the characteristics of the first coordination sphere but
tion (5.8 or 8.7 kcalmol–1 in the gas phase). As a whole, also on factors as difficult to quantify as the specific inter-
these results indicate that the stabilisation associated with actions with solvent molecules. In any case, despite the fact
this weak axial coordination of the second water molecule that calculations fail to predict the tetradentate behaviour
is lower than that associated with hydrogen bonding with of the cyclophane in [CuL1]2+, the calculations indicate that
one of the coordinated amine groups so that it is not ex- a higher stability of the tridentate form is expected to occur
pected to be coordinated in solution, especially in the case when changing to the para L2 ligand. Inspection of the data
of L2.
in Table 3 reveals that changing from L1 to L2 does not
Eur. J. Inorg. Chem. 2009, 62–75
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
67

M. G. Basallote, E. García-España et al.
FULL PAPER
change very much the energy of the most stable tetradentate
form but it leads to stabilisation of all the tridentate forms,
especially those involving coordination of one benzylic ni-
trogen, in agreement with the tridentate coordination mode
found in the crystal structure of the HgII complex with a
related cyclophane.[2b]
Kinetics of Decomposition of the CuII Complexes with the
L1 and L2 Ligands
The species distribution curves in Figure 2 and Figure 3
show that upon addition of an excess of acid, the Cu–L1
and Cu–L2 complexes decompose with formation of Cu2+
and the fully protonated ligands, as indicated in Equa-
tion (1) for the case of a mononuclear species.
Figure 6. Plot of the observed rate constants vs. the acid concentra-
tion for the decomposition of the different Cu–L1 complexes
(25.0 °C, 0.15  NaClO4): [CuHL1]3+ (circles), [CuL1]2+ (triangles)
and [Cu2L1(OH)2]2+ (squares).
CuHxL(2+x)+ + H+ Ǟ Cu2+ + H5L5+
(1)
exc
The spectra of the reaction mixture immediately after
mixing in the stopped-flow instrument (ca. 2 ms) and at the
end of the reaction were also recorded in all cases from the
analysis of kinetic experiments using a diode-array detector
and they were compared with those of the complex species
before addition of the excess acid. Some representative
spectra are shown in Figure 7 which includes the spectrum
of the dinuclear [Cu2L2(OH)]3+ species in the absence of
added acid (a) and the initial (b) and final (c) spectra in the
kinetic experiments. These spectra clearly reveal the exis-
tence of a rapid absorbance change that occurs within the
mixing time of the stopped-flow instrument. This initial fast
absorbance change is only observed when the starting solu-
tion contains a dinuclear species, no rapid changes of sim-
ilar characteristics being observed for the mononuclear spe-
cies. Actually, the analysis of the spectroscopic changes ob-
served during the decomposition of the mononuclear
[CuHL2]3+ and [CuL2]2+ species leads to initial and final
spectra similar to those in Figure 7 (b) and Figure 7 (c). It
is also interesting to note that dissociation of the first metal
ion from the dinuclear [Cu2L2(OH)]3+ species does not
cause any significant change in the position of the absorp-
tion maximum despite the fact that the hydroxyl bridge
must be destroyed in the process. Nevertheless this observa-
tion is in agreement with the conclusions derived from the
analysis of the spectra obtained at different pH and Cu/L
ratios. As the rate constants obtained for the different Cu–
L1 species are significantly different, special care was taken
in this case to compare the spectra obtained for each species
immediately after mixing with the acid excess. A rapid re-
lease of one CuII ion within the mixing time of the stopped-
flow instrument was also observed during the decomposi-
tion of the dinuclear [Cu2L1(OH)2]2+ species and in all cases
an absorption maximum centred at wavelengths close to
570 nm was observed during the early stages of the stopped-
flow experiment which suggests a similar coordination envi-
ronment for the metal ion in all the three mononuclear Cu–
L1 species, the decomposition kinetics of which lead to the
rate constants in Figure 6.
Stopped-flow experiments showed that the acid pro-
moted decomposition of the CuII complexes with both cy-
clophanes occurs in all cases with a single measurable ki-
netic step with rate constants that change with the acid con-
centration according to the rate law in [Equation (2)] (Fig-
ures 5 and 6). However, whereas for the L2 complexes the
values of the observed rate constants are independent of the
nature of the species in the starting solution (Figure 5) and
the whole set of data can be well fitted by Equation 2 to
obtain values of a = 17.5Ϯ0.3 s–1 and b = 114Ϯ9 –1 s–1,
the kinetic data for the L1 complexes change significantly
for the different species studied (Figure 6). The values of
the a and b parameters for each species are: a =
24.7Ϯ0.1 s–1 and b = 119Ϯ2 –1 s–1 for [CuHL1]3+, a =
17.0Ϯ0.5 s–1 and b = 114Ϯ14 –1 s–1 for [CuL1]2+, and a =
13.3Ϯ0.3 s–1 and b = 23Ϯ8 –1 s–1 for [Cu2L1(OH)2]2+
.
kobs = a + b[H+]
(2)
Figure 5. Plot of the observed rate constants vs. the acid concentra-
tion for the decomposition of the CuII complexes with L2 (25.0 °C,
0.15  NaClO4). The solid line corresponds to the best fit of all the
data in the plot.
68
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 62–75

CuII Complexes of Pentaazacyclophane Ligands
the electronic spectra for the different species indicates a
similar coordination environment of the metal ion, i.e. the
number of donor groups used by the ligand must be the
same in all cases independent of the protonation state of
the complex. From the position of the absorption band, a
tetradentate binding mode for L1 can be deduced[20] for the
three mononuclear species, the decomposition kinetics of
which were monitored: [CuHL1]3+, [CuL1]2+ and the one
resulting from the rapid release of one CuII ion from the
dinuclear complex. Although it is evident that a species
such as [CuHL1]3+ contains a protonated amine group that
does not exist in [CuL1]2+, if the cyclophane is coordinated
to the metal ion through the same donor groups in both
complexes, the addition of the excess acid to a solution of
[CuL1]2+ would lead to the formation of [CuHL1]3+ within
the mixing time of the stopped-flow instrument and the
same kinetics of decomposition should be observed. This
reasoning leads to the conclusion that L1 coordinates to the
metal ion through the same number of donor atoms in both
Figure 7. Comparison of the electronic spectrum of the [Cu2-
L2(OH)]3+ species (a) with the spectra obtained from the analysis
of the spectroscopic changes recorded during the acid-promoted
decomposition of this species: initial spectrum (b) and final product
(c).
As a summary, the kinetics results indicate that decom- species (four according to the spectra and the equilibrium
position of the mononuclear CuII complexes with both cy- data) but the actual donor groups differ from one complex
clophanes occurs in a single step, whereas the dinuclear spe- to the other. As protonation of the uncoordinated amine
cies decompose in two steps, the first metal ion being re- groups is faster than the breaking of Cu–N bonds, a species
leased much faster than the second. Unfortunately, dissoci- of composition [CuHL1]3+ is rapidly formed when an excess
ation of the first CuII occurs within the mixing time of the of acid is added to a solution of [CuL1]2+ but this
instrument and no kinetic data could be obtained. It is im- [CuHL1]3+ species is an isomer of the most thermodynami-
portant to note that the decomposition of the dinuclear cally stable species with the same composition and so its
complexes formed with larger polyaza macrocycles and kinetics of decomposition will be different. A similar rea-
cryptands occurs with a lower rate and statistically con- soning can be applied to the mononuclear species formed as
trolled kinetics, i.e. both metal centres behave independently an intermediate during the decomposition of the dinuclear
and the rate of release of the first metal ion doubles the [Cu2L1(OH)2]2+ complex. It appears unlikely that L1 coor-
rate corresponding to the second.[26–29] Thus, it appears that dinates to one of the metal centres in the dinuclear species
the smaller size of the L1 and L2 cyclophanes creates impor- through more than three amine groups but the lability of
tant electrostatic repulsions between the metal centres in the CuII ion again makes possible the formation of a
the dinuclear species and this results in the faster release of [CuHL1]3+ species following the dissociation of the first
one metal ion upon treatment with acid.
metal ion. As this species decomposes with kinetics dif-
The similarity of the spectroscopic changes and rate con- ferent from those observed in the study of the mononuclear
stants for the only measurable kinetic step for all the mono- species, it must be concluded that it is actually a different
and dinuclear Cu–L2 species strongly suggests that the same isomer. Thus, the different values of the a and b kinetics
process is monitored in all cases which is surely the decom- parameters obtained for solutions containing the
position of the most acidic mononuclear species. These re- [CuHL1]3+, [CuL1]2+ and [Cu2L1(OH)2]2+ complexes would
sults for the L2 complexes can be easily rationalised by con- actually correspond to the decomposition of three isomeric
sidering that the cyclophane does not use all the donor forms of the [CuHL1]3+ species.
groups in the mononuclear complexes and the different spe-
An attractive hypothesis is that the three species show
cies studied only differ by the protonation/deprotonation of the coordination environment depicted schematically in
some of the uncoordinated amine groups. In that case, ad- Figure 8. Independent of their precise structures, it can be
dition of an excess of acid leads to the rapid protonation of anticipated that these isomers would have Cu–N bonds with
the amine groups which remain uncoordinated and depro- different steric constrains that would lead to different kinet-
tonated. As a consequence, the species formed within the ics of decomposition.[23,30–34] We have observed recently[7]
mixing time of the stopped-flow instrument is always the these kinds of differences in the kinetics of decomposition
same and so the same kinetics of decomposition are ob- for the mononuclear CuII complexes of cyclophanes with
served independent of the starting species.
larger polyamine chains containing an additional amine
In contrast, the results obtained for the L1 complexes group and suggested that these kinetic measurements are a
cannot be explained in the same way. The three different useful test to detect changes in the actual coordination
Cu–L1 mononuclear species, including that resulting from mode of the ligand. However, extensive DFT calculations
dissociation of the first ion from the dinuclear complex, de- indicated that the only way of getting a tetradentate coordi-
compose with different kinetics despite the fact that their nation for L1 in mononuclear CuII complexes is by keeping
spectra are also very similar in all cases. This similarity of one of the benzylic nitrogen atoms uncoordinated, i.e. as
Eur. J. Inorg. Chem. 2009, 62–75
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
69

M. G. Basallote, E. García-España et al.
FULL PAPER
shown in the structure at the right hand side of Figure 8.
For the decomposition of the CuII complexes with the
Thus, the experimental observation of different kinetics of open-chain polyamine L4,[19] the value of a is negligible and
decomposition for three species with the same set of donor b = 217 –1 s–1. Thus, although the macrocyclic nature of
atoms would reveal that the complex does not reorganise the L1 and L2 ligands only causes small changes in the val-
rapidly to the most stable conformation of [CuHL1]3+ upon ues of b, the appearance of a significant a term indicates
protonation of [CuL1]2+ or CuII dissociation from [Cu2- that the contribution of the solvent-attack pathway be-
L1(OH)2]2+. In the latter cases, the rapid processes occur- comes more important for the complexes with both cy-
ring within the mixing time of the stopped-flow instrument clophanes. However, for the CuII complexes with the pyridi-
appear to lead to different conformations of the [CuHL1]3+ nophane L3,[19] a becomes negligible again, b is somewhat
species. If decomposition is faster than reorganisation to larger (520–700 –1 s–1) and the approximation kH[H+] ϽϽ
the most stable conformation, these transient species will (k–1 + kH O) is not valid, the plots of kobs vs. [H+] showing
2
decompose with different kinetics because of the different a clear curvature that allows the calculation of a new pa-
steric constraints of the Cu–N bonds in their coordination rameter c = kH/(k–1 + kH O) = 27 –1. Thus, the cyclic na-
2
environments.
ture of the polyamine is not enough to cause the appear-
ance of a significant a term in the kinetics of decomposition
of the CuII complexes. The relative importance of both par-
allel attacks must then be better related to differences in the
nature of the activated intermediate “Cu–N” which facili-
tates or hinders attacks by solvent and H+, depending of
the degree of bond dissociation. Solvent attack will be facil-
itated when the Cu–N bond becomes more elongated in the
intermediate, whereas the assistance of H+ is necessary
when there is a small degree of bond dissociation. The ex-
tent to which the Cu–N bond is elongated in the intermedi-
ate appears to be determined by subtle changes in the steric
constraints imposed by the ligand in the different species
more than by the cyclic/acyclic nature of the polyamine.
Thus, for the phenanthroline analogue of L1 and L2, the
[CuHL]3+ species decomposes mainly through the solvent-
dependent pathway (a = 18 s–1, b = 1.34 –1 s–1), whereas
[CuL]2+ decomposes exclusively through the H+-pathway (a
= 0, b = 4264 –1 s–1, c = 52 –1).[35]
Figure 8. Schematic drawings showing three hypothetical isomeric
forms of the [CuHL1]3+ species with the L1 cyclophane acting as a
tetradentate ligand. The actual coordination spheres can be more
complex because of the presence of additional water ligands.
The comparison of the actual values of the kinetic pa-
rameters obtained in the present work with those previously
reported for related complexes requires a previous consider-
ation of the intimate mechanism of decomposition. The
mechanism widely accepted in the literature[30–32] for the
decomposition of polyamine complexes of CuII assumes
that the acid attack on the Cu–N bond that is broken in
the rate-determining step does not occur directly because it
requires the previous formation of an activated intermediate
“Cu–N” with partial breaking of the Cu–N bond [Equa-
tion (3)]. The complete dissociation of the metal–ligand
bond can be achieved through two parallel pathways that
involve solvent or acid attacks on this intermediate [Equa-
tions (4) and (5)]. If the activated intermediate is assumed
to be formed under steady-state conditions the rate law for
this mechanism is given by Equation (6) which can be sim-
plified to the form of Equation (2) with the equivalences a
= k1kH O/(k–1 + kH O) and b = k1kH/(k–1 + kH O), assuming
Interaction with AMP and Formation of Ternary CuII–
AMP-Cyclophane Complexes
To further explore the coordination properties of these
new cyclophanes, additional studies were carried out on
AMP recognition. The formation of mixed [CupHrL-
(AMP)](2p+r–2)+ complexes and especially their stability is of
relevance in oligonucleotide binding,[12,36–38] and the pres-
ent Cu–L systems (L = L1, L2, L3) offer an excellent oppor-
tunity for exploring the recognition capability in closely re-
lated mono- and dinuclear species.
2
2
that kH[H+] ϽϽ (2k–1 + kH O). Thus, the b/a quotient is
2
A prerequisite for the determination of the constants of
the mixed AMP–CuII complexes with L1, L2 and L3 is the
knowledge of the stability constants of all the binary sys-
tems implicated. As polyammonium receptors can interact
by themselves with nucleotide species, the equilibrium con-
stants for the systems AMP–L1, AMP–L2 and AMP–L3
were determined potentiometrically and the values obtained
for the global stability constants are included in Table 4.
However, as both AMP and the L receptors participate in
overlapping protonation processes, translating those cumu-
lative constants into representative stepwise constants re-
quires consideration of the basicity of AMP[39] and the dif-
equivalent to kH/kH O and indicates the relative rates of at-
2
tack by H+ and H2O on the activated intermediate.
Cu–N p “Cu–N”; k1, k–1
(3)
(4)
(5)
“Cu–N” + H2O Ǟ Cu2+ + HN+; kH O
2
“Cu–N” + H+ Ǟ Cu2+ + HN+; kH
(6) ferent cyclophanes. If it is assumed that the interaction be-
70
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 62–75

CuII Complexes of Pentaazacyclophane Ligands
Figure 9. Species distribution curves for the L-A-H systems (L = L1, L2, L3; A = AMP–2; [L]0 = [A]0 = 1ϫ10–3 ).
tween both species does not alter very much in the pH
range in which the different protonated forms of AMP and
L exist, the stepwise constants shown in the last part of
Table 4 can be considered representative of the equilibria
occurring in solution. Species distribution curves calculated
from the equilibrium data in Table 4 are shown in Figure 9
which reveals that HxLx+-AMP species dominate at most
pH values for the case of the L1 and L3 ligands, the interac-
tion of the para L2 ligand being significantly lower. This is
clearly reflected in a plot of the amount of complexed AMP
(A) by the different receptors (A/L, 1:1 mole ratio) shown
in Figure 10.
Table 4. Logarithms of the stability constants for the L1–AMP, L2–
AMP and L3–AMP species (A
=
AMP–2) determined at
298.1Ϯ0.1 K in 0.15 moldm–3 NaClO4.
Figure 10. Plot of the percentage of complexed AMP for the three
systems ([AMP] = 10–3 ).
Reaction[a]
L1
L2
L3
A + L+ H = HLA
–
14.65(4)[b]
23.78(4)
31.24(9)
–
44.65(6)
48.87(6)
–
14.07(5)
23.07(5)
30.80(4)
37.67(4)
43.59(4)
47.22(5)
–
A + L+ 2 H = H2LA
A + L+ 3 H = H3LA
A + L+ 4 H = H4LA
A + L+ 5 H = H5LA
A + L+ 6 H = H6LA
A + L+ 7 H = H7LA
22.37(1)
30.89(1)
38.47(1)
44.84(1)
48.89(1)
51.46(2)
cycles[36,38] and significantly higher than those found for
smaller amines and for biogenic amines such as spermidine
or putrescine.[8,40] It is also interesting to note that the sta-
bility of the AMP–HxL species is also lower with polyaza
macrocycles that are larger than the present cyclophanes.[41]
This appears to show that the interaction is optimised for
macrocycles of intermediate size in which the phosphate
group better matches the positive charges of the macro-
cyclic cavity. In addition, the interaction probably becomes
stronger for the case of cyclophanes because of additional
π stacking interactions and actually, the stability constants
found in this work are larger than those found for macro-
cycles of a similar size but lacking aromatic spacers.[8,36–40]
Another interesting conclusion from the data in Table 4 is
A + LH = HLA
–
3.96
3.43
2.57
–
2.71
3.90
3.03
–
4.42
4.10
4.21
4.46
4.32
5.09
4.05
–
A + LH2 = H2LA
A + LH3 = H3LA
A + LH4 = H4LA
AH + LH4 = H5LA
AH + LH5 = H6LA
AH2 + LH4 = H6LA
AH2 + LH5 = H7LA
3.30
4.01
4.59
4.90
6.16
5.05
4.83
[a] Charges omitted for clarity. [b] Values in parentheses are stan-
dard deviations in the last significant Figure.
In general, the values of the constants in Table 4 are of that the strength of the interaction between AMP2– and the
the same order as those found for the interaction of AMP different HxLx+ forms does not appear to be affected signif-
with the protonated forms of some other polyaza macro- icantly when the positive charge on the cyclophane is in-
Eur. J. Inorg. Chem. 2009, 62–75
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
71

M. G. Basallote, E. García-España et al.
FULL PAPER
creased which strongly suggests that the strength of the in- observed for the protonated forms of the ligand, except for
teraction is not dominated only by charge-charge electro- the case of the mononuclear complexes of L3, the stability
static factors. Other factors like hydrogen bonding should constants of which for the interaction with AMP are up to 4
also significantly affect the interaction. In this sense, it must log units higher than those found for the protonated ligand.
be pointed out that although the stability of HxLx+-AMP However, it must be pointed out that the nature of the inter-
adducts usually tends to increase with the protonation state action between AMP and the receptor is presumably dif-
of L because of the increased coulombic attractions and the ferent in both cases because Cu–L complexes are coordina-
larger number of possible hydrogen bonds between the host tively unsaturated and so they are susceptible to AMP coor-
and the guest,[38,40,41] there are also literature reports show- dination. At this point, it is interesting to note that all the
ing similar stability constants for the interaction of AMP log K values in Table 5 indicate that all the Cu–L complexes
with the different protonated forms of some tripodal polya- interact with AMP more strongly than Cu2+ (log KCu–AMP
mines.[42]
= 3.0–3.2)[40,43–45] although the values can be considered to
The determination of the stability constants for the Cu–
L-AMP systems (L = L1, L2 and L3) through the analysis
of the titration curves is much more complex because of the
large number of species present in solution. The analysis
was carried out by fixing the values of the protonation con-
Table 5. Logarithms of the stability constants for the formation of
mixed Cu–L-AMP complexes (AMP–2
298.1Ϯ0.1 K in 0.15 moldm–3 NaClO4.
= A) determined at
Reaction[a]
L1
L2
L3
stants of AMP,[39] the formation constants of the Cu2+
–
Cu + L + A = CuLA
23.5(1)[b]
–
–
–
AMP complexes[43] and the previously discussed proton-
ation constants of the different L cyclophanes (Table 1), the
equilibrium constants for the AMP–L system (Table 4) and
those for the Cu–L complexes (Table 2). The results so ob-
tained are included in Table 5 and representative species dis-
tribution curves are shown in Figure 10. Because of the
complex nature of the equilibrium mixtures, the quality of
Cu + H + L + A = CuHLA
Cu + 2 H + L + A = CuH2LA
Cu + 3 H + L + A = CuH3LA
Cu + 4 H + L + A = CuH4LA
2 Cu + L + A = Cu2LA
31.2(1)[b]
34.08(6)
37.89(1) 36.00 (3) 40.55(6)
41.56(2) 40.41(10) 44.53(7)
–
45.18 (4) 47.88(7)
27.78 (4) 33.73(7)
–
2 Cu + H + L + A = Cu2HLA
2 Cu + 2 H + L + A = Cu2H2LA
2 Cu + L + A = Cu2LA
–
–
–
38.80(1)
–
–
39.93(4)
–
29.09(2)
these results was then checked by fitting together the data 2 Cu + H2O + L + A =
corresponding to titrations of binary Cu–L and ternary
22.00(2) 20.61(5) 24.80(1)
Cu2LA(OH) + H
Cu–L-AMP systems including, as parameters to be refined,
all the constants for the formation of Cu–L and Cu–L-
AMP complexes. This analysis leads to results similar to
those previously described, the differences being in all cases
within the limits of the estimated errors.
CuH3L + AH = CuH4LA
CuH2L + AH2 = CuH4LA
CuH2L + AH = CuH3LA
CuHL + AH2 = CuH3LA
CuHL + AH = CuH2LA
CuHL + A = CuHLA
–
–
5.87
5.30
5.53
4.90
–
3.98
4.07
3.53
4.75
4.24
–
–
7.77
8.32
7.17
7.09
6.68
4.10
Cu2L(OH) + A = Cu2LA(OH)
5.42
Comparison of the data in Tables 4 and 5 indicates that
the CuII complexes of the L cyclophanes show stability con-
[a] Charges omitted for clarity. [b] Values in parentheses are stan-
stants for the interaction with AMP quite similar to those dard deviations in the last significant Figure.
Figure 11. Species distribution diagrams for the mixed Cu–L-A-H systems (L = L1, L2, L3; A = AMP–2; [Cu+2] = [L]0 = [A]0 = 1ϫ10–3 ).
72
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 62–75

CuII Complexes of Pentaazacyclophane Ligands
2,6,9,12,16-Pentakis(p-tolylsulfonyl)-2,6,9,12,16-pentaza[17]paracy-
clophane (L2·5Ts): This compound was obtained using the pro-
cedure described for L1·5Ts except that α,αЈ-dibromo-p-xylene was
used instead of α,αЈ-dibromo-m-xylene. The product was also iso-
lated as a white solid (5.53 g, yield 67%); m.p. 187–189 °C.
C53H63N5O10S5 (1090.41): calcd. C 58.38, H 5.82, N 6.42; found C
58.3, H 5.9, N 6.5. 1H NMR (CDCl3): δ = 1.57–1.63 (m, 4 H), 2.42
(s, 12 H), 2.46 (s, 3 H), 2.91–3.11 (m, 16 H), 4.19 (s, 4 H), 7.17 (s,
4 H), 7.29 (d, J = 8 Hz, 8 H), 7.33 (d, J = 8 Hz, 4 H), 7.63 (d, J =
8 Hz, 8 H), 7.72 (d, J = 8 Hz, 4 H) ppm. 13C NMR (CDCl3): δ =
be similar to those found for other Cu–polyamine com-
plexes.[40,41,46] The stability of the interactions with mono-
nuclear Cu–L3 complexes are among the most stable. These
results, and especially the large stabilisation observed for
the L3 complexes, strongly suggest that selectivity in re-
cognition is probably determined more by other interac-
tions (hydrogen bonding, π stacking) than by direct coordi-
nation to the metal centre.
However, because of the existence of multiple competi-
tive equilibria, we have plotted the percentages of com- 19.2, 26.8, 45.4, 46.2, 46.6, 48.0, 51.2, 124.9, 125.0, 125.8, 127.5,
127.6, 132.8, 133.6, 134.1, 138.6, 141.2, 141.7 ppm. MS (FAB): m/z
plexed AMP as a function of pH for both systems in order
to better compare the affinity for AMP of the free ligands
and their protonated complexes with that that displayed by
the CuII–L complexes (see Figure 11 and Supporting Infor-
mation). While for the systems AMP–L1 and AMP–L3 CuII
addition does not yield large increases in the amount of
= 1089 [M – H]+.
2,6,9,12,16-Pentaza[17]metacyclophane Pentahydrochloride (L1·5HCl):
The tosyl groups of L1·5Ts (2.79 g, 2.56 mmol) were removed by
reductive cleavage with a mixture of HBr/HAc (150 mL) and PhOH
(15.20 g, 16.0 mmol) by heating at 90 °C for 24 h. The solid ob-
complexed AMP, in the case of the para derivative L2 the tained was then filtered and washed with a mixture of EtOH/
amount of complexed AMP in the presence of CuII is
CH2Cl2 (1:1 v/v). The macrocycle was obtained as its hydrobromide
salt and was then converted into the free amine by using an ionic
clearly augmented over a broad pH range (Figure 12).
exchange resin (Amberlite IRA 402). The free amine was purified
by chromatography on neutral alumina using methanol as the elu-
ent. The resultant oil was dissolved in ethanol and treated with
37% HCl until complete precipitation of a white solid which was
filtered to give L1 as its pentahydrochloride salt (0.45 g, yield 35%);
m.p. 185–187 °C. C18H33N5·5HCl·3H2O (555.84): calcd. C 38.90,
1
H 7.98, N 12.60; found C 39.0, H 8.0, N 12.6. H NMR (D2O): δ
= 1.85–1.95 (m, 4 H), 2.95–2.98 (m, 8 H), 3.22 (s, 8 H), 4.04 (s, 4
H), 7.28 (s, 3 H), 7.37 (s, 1 H) ppm. 13C NMR (D2O): δ = 23.2,
43.9, 44.1, 44.4, 44.9, 51.1, 130.5, 131.3, 131.6, 132.8 ppm.
2,6,9,12,16-Pentaza[17]paracyclophane Pentahydrochloride (L2·5HCl):
This compound was obtained from L2·5Ts using the same pro-
cedure described above for the analogous metacyclophane (0.68 g,
yield 33%); m.p. 248–250 °C. C18H33N5·5HCl·3H2O (555.84):
1
calcd. C 38.90, H 7.98, N 12.60; found C 39.3, H 8.0, N 12.6. H
NMR (D2O): δ = 1.66–1.75 (m, 4 H), 2.68–2.70 (m, 4 H), 2.75–
Figure 12. Percentages of complexed AMP for the system CuII–
2.85 (m, 12 H), 4.04 (s, 4 H), 7.41 (s, 4 H) ppm. 13C NMR (D2O):
AMP-L calculated for 1:1:1 molar ratio ([AMP] = 1ϫ10–3 ).
δ = 23.2, 43.1, 44.6, 45.2, 50.6, 131.4, 131.6 ppm.
EMF Measurements: The potentiometric titrations were carried out
at 298.1Ϯ0.1 K using 0.15  NaClO4 as the supporting electrolyte.
Experimental Section
The experimental procedure (burette, potentiometer, cell, stirrer,
microcomputer, etc.) has been fully described elsewhere.[47] The ac-
Syntheses
quisition of the emf data was carried out with the computer pro-
2,6,9,12,16-Pentakis(p-tolylsulfonyl)-2,6,9,12,16-pentaza[17]metacy-
clophane (L1·5Ts): A solution of α,αЈ-dibromo-m-xylene (2.03 g,
7.67 mmol) in a dry CH3CN/CH2Cl2 (1:1 v/v, 150 mL) mixture was
added dropwise to a suspension of 1,5,8,11,15-pentakis(p-tolylsul-
fonyl)-1,5,8,11,15-pentazapentadecane (7.60 g, 7.67 mmol) and
K2CO3 (10.55 g, 76.3 mmol) in dry CH3CN (250 mL). The suspen-
sion was heated to reflux for a further 24 h and then filtered. The
resultant solution was vacuum-evaporated to dryness and the resi-
due suspended in ethanol at reflux to give L1·5Ts as a white solid
(2.79 g, yield 37 %); m.p. 108–110 °C. C53H63N5O10S5 (1090.41):
calcd. C 58.38, H 5.82, N 6.42; found C 58.1, H 5.9, N 6.1. 1H
NMR (CDCl3): δ = 1.74–1.76 (m, 4 H), 2.42 (s, 6 H), 2.44 (s, 6 H),
gram PASAT.[48] The reference electrode was an Ag/AgCl electrode
in saturated KCl solution. The glass electrode was calibrated as a
hydrogen-ion concentration probe by titration of previously stan-
dardised amounts of HCl with CO2-free NaOH solutions and de-
termination of the equivalent point by the Gran’s method[49] which
gives the standard potential E0 and the ionic product of water [pKw
= 13.73(1)].
The computer program HYPERQUAD was used to calculate the
protonation and stability constants from the titration curves.[50]
The protonation constants for AMP were taken from ref.[12]. The
pH range investigated was 2.5–10.5 and the concentration of CuII,
2.46 (s, 3 H), 3.03–3.09 (m, 8 H), 3.15–3.20 (t, J = 8 Hz, 4 H), AMP and ligands ranged from 3ϫ10–4 to a maximum value of
3.26–3.29 (t, J = 8 Hz, 4 H), 4.25 (s, 4 H), 7.25–7.28 (d, J = 8 Hz, 1.5ϫ10–3 moldm–3. The different titration curves for each system
4 H), 7.28–7.31 (d, J = 8 Hz, 4 H), 7.32–7.35 (d, J = 8 Hz, 2 H),
(at least two) were treated either as a single set or as separated
7.37 (s, 3 H), 7.48 (s, 1 H), 7.61–7.63 (d, J = 8 Hz, 4 H), 7.68–7.71 curves without significant variations in the values of the stability
(d, J = 8 Hz, 4 H), 7.73–7.76 (d, J = 8 Hz, 2 H) ppm. 13C NMR
(CDCl3): δ = 19.1, 26.9, 45.8, 46.2, 46.4, 47.2, 51.5, 124.8, 125.1,
127.5, 127.6, 135.7, 141.2 ppm. MS (FAB): m/z = 1090 [M+].
constants. The sets of data were merged together and treated simul-
taneously to give the final stability constants. Moreover, in the case
of the AMP–L systems several measurements were made both in
Eur. J. Inorg. Chem. 2009, 62–75
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
73

M. G. Basallote, E. García-España et al.
FULL PAPER
formation and in dissociation (from acid to alkaline pH and vice
versa) to check the reversibility of the reactions.
[1]
[2]
a) A. Bencini, M. I. Burguete, E. García-España, S. V. Luis,
J. F. Miravet, C. Soriano, J. Org. Chem. 1993, 58, 4749–4753;
b) M. I. Burguete, B. Escuder, E. García-España, S. V. Luis,
J. F. Miravet, J. Org. Chem. 1994, 59, 1067–1071; c) B. Altava,
M. I. Burguete, B. Escuder, E. García-España, M. C. Muñoz,
Tetrahedron 1997, 53, 2629–2640.
a) A. Andrés, M. I. Burguete, E. García-España, S. V. Luis,
J. F. Miravet, C. Soriano, J. Chem. Soc. Perkin Trans. 2 1993,
749–755; b) E. García-España, J. Latorre, S. V. Luis, J. F. Mira-
vet, P. Pozuelo, J. A. Ramírez, C. Soriano, Inorg. Chem. 1996,
35, 4591–4596; c) B. Altava, A. Bianchi, C. Bazzicalupi, M. I.
Burguete, E. García-España, S. V. Luis, J. F. Miravet, Supra-
mol. Chem. 1997, 8, 287–299; d) J. A. Aguilar, P. Díaz, A. Dom-
énech, E. García-España, J. M. Llinares, S. V. Luis, J. A. Ramí-
rez, C. Soriano, J. Chem. Soc. Perkin Trans. 2 1999, 1159–1168;
e) M. I. Burguete, P. Díaz, E. García-España, S. V. Luis, J. F.
Miravet, M. Querol, J. A. Ramírez, Chem. Commun. 1999, 649–
650.
A. Andrés, C. Bazzicalupi, A. Bianchi, E. García-España, S. V.
Luis, J. F. Miravet, J. A. Ramírez, J. Chem. Soc., Dalton Trans.
1994, 2995–3004.
M. I. Burguete, B. Escuder, E. García-España, J. Latorre, S. V.
Luis, J. A. Ramírez, Inorg. Chim. Acta 2000, 300, 970–977.
B. Altava, M. I. Burguete, S. V. Luis, J. F. Miravet, E. García-
España, V. Marcelino, C. Soriano, Tetrahedron 1997, 53, 4751–
4762.
M. I. Burguete, E. García-España, S. V. Luis, J. F. Miravet, C.
Soriano, An. Quím. 1993, 89, 99–100.
J. Aguilar, M. G. Basallote, L. Gil, J. C. Hernández, M. A.
Máñez, E. García-España, C. Soriano, B. Verdejo, Dalton
Trans. 2004, 94–103.
E. Kimura, M. Kodama, T. Yatsunami, J. Am. Chem. Soc.
1982, 104, 3182–3187.
M. W. Hosseini, J. M. Lehn, M. P. Mertes, Helv. Chim. Acta
1983, 66, 2454–2466.
B. Dietrich, M. W. Hosseini, J. M. Lehn, R. B. Sessions, J. Am.
Chem. Soc. 1981, 103, 1282–1283.
M. W. Hosseini, J. M. Lehn, Helv. Chim. Acta 1987, 70, 1312–
1319.
DFT Calculations: All DFT calculations were performed using the
Gaussian03 package[51] and the unrestricted Becke three-parameter
hybrid functional combined with the Lee–Yang–Parr correlation
functional (UB3LYP).[52,53] The Pople style 6-31G(d,p) basis set
was used on the C, H, O and N atoms, and the SDD basis set[54]
and effective core potential (ECP) on Cu (as implemented in
Gaussian03, SDD is D95V up to Ar and Stuttgart/Dresden ECPs
for the remaining elements of the periodic Table). All geometry
optimisations were performed without any symmetry constrains
and efforts were made to find the lowest energy conformation by
comparing the structures optimised from different starting geome-
tries. Harmonic frequency calculations were performed to confirm
that the calculated structures were minima. Aqueous solution ener-
gies were calculated using the CPCM formalism[55,56] (water sol-
vent, ε = 78.39 as implemented in Gaussian03). The gas phase ge-
ometry was used for all of these calculations because it has been
demonstrated in many previous studies that the change of geometry
due to solvation effects is usually insignificant.[57,58]
[3]
[4]
[5]
Kinetic Experiments: The kinetics of decomposition of the CuII
complexes with the ligands L1 and L2 were studied at 298.1Ϯ0.1 °C
using an Applied Photophysics SX17MV stopped-flow spectropho-
tometer. The Cu/L ligand ratio and the pH of the starting solutions
of the metal complexes were selected from the species distribution
curves in order to achieve the maximum concentration for one of
the complex species while maintaining a very low concentration of
the others. For this reason, only those species which represent at
least 80% of the total ligand under some conditions were studied.
These solutions were mixed in the stopped-flow instrument with
solutions containing HClO4 at the desired concentration. The ionic
strength of both solutions was adjusted to 0.15  with NaClO4.
The decomposition of the L1 and L2 complexes was monitored at
575 nm and 610 nm, respectively, because the absorbance changes
observed in preliminary experiments using a diode-array detector
were at a maximum at these wavelengths. In all cases, the data
for the acid-promoted decomposition of the complexes could be
satisfactorily fitted by a single exponential using the software of
the stopped-flow instrument. The data from the preliminary diode-
array experiments were analysed with the GLINT program[59] and
yielded rate constants similar to those derived from the single-wave-
length experiments in addition to the spectra of the reaction mix-
ture immediately after mixing and at the end of the decomposition
process.
[6]
[7]
[8]
[9]
[10]
[11]
[12]
A. Bencini, A. Bianchi, E. García-España, E. C. Scott, L. Mo-
rales, B. Wang, T. Deffo, F. Takusagawa, M. P. Mertes, K. B.
Mertes, P. Paoletti, Bioorg. Chem. 1992, 20, 8–29.
For a recent review on ATP coordination by polyazamacro-
cycles see for instance: E. García-España, P. Díaz, J. M. Llin-
ares, A. Bianchi, Coord. Chem. Rev. 2006, 250, 2952–2987.
E. Martell, R. J. Motekaitis, Q. Lu, D. A. Nation, Polyhedron
1999, 18, 3203–3218.
Szymanska, H. Radecka, J. Radecki, M. Pietraszkiewicz, O.
Pietraszkiewicz, Comb. Chem. High Throughput Screening
2000, 3, 509–517.
S. Y. Lin, W. H. Chen, C. Y. Liu, Electrophoresis 2002, 23,
3550–3557.
[13]
[14]
[15]
Supporting Information (see footnote on the first page of this arti-
cle): Additional species distribution curves, EPR spectra, Cartesian
coordinates, geometries and energies of all the species optimized
by DFT procedures.
[16]
[17]
M. T. Albelda, J. Aguilar, S. Alves, R. Aucejo, P. Díaz, C. Lod-
eiro, J. C. Lima, E. García-España, F. Pina, C. Soriano, Helv.
Chim. Acta 2003, 86, 3118–3135.
[18]
[19]
J. A. Aguilar, P. Díaz, F. Escartí, E. García-España, L. Gil, C.
Soriano, B. Verdejo, Inorg. Chim. Acta 2002, 339, 307–316.
P. D íaz, M. G. Basallote, M. A. Máñez, E. García-España, L.
Gil, J. Latorre, C. Soriano, B. Verdejo, S. V. Luis, Dalton Trans.
2003, 1186–1193.
H. Gampp, D. Haspra, M. Maeder, A. D. Zuberbuehller, Inorg.
Chem. 1984, 23, 3724–3730.
S. Siddiqui, R. E. Shepherd, Inorg. Chem. 1986, 25, 3869–3876.
See for instance: a) E. Kimura, I. Nakamura, T. Koike, M.
Shionoya, Y. Kodama, T. Ikeda, M. Shiro, J. Am. Chem. Soc.
1994, 116, 4764–4771; b) T. Koike, S. Kajitani, I. Nakamura,
E. Kimura, M. Shiro, J. Am. Chem. Soc. 1995, 117, 1210–1219;
c) C. Bazzigalupi, A. Bencini, E. Berni, S. Ciattini, A. Bianchi,
Acknowledgments
We would like to acknowledge Ministerio de Ciencia y Tecnología
(MCYT) and the Federación Española de Enfermedades Raras
(FEDER) (Grants CTQ2006-14909-C02-01 and CTQ2006-15672-
CO5-01) as well as the Junta de Andalucía (FQM-137) for financial
support. Computer facilities by the Centro de Supercomputación
de la Universidad de Cádiz are also acknowledged. Predoctoral
grants are also acknowledged to Ayuntamiento de Valencia (to
B. V.), Universidad de Cádiz (to C. E. C.), and MCYT (to S. B. and
A. G. A.).
[20]
[21]
[22]
74
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 62–75

CuII Complexes of Pentaazacyclophane Ligands
C. Giorgi, P. Paoletti, B. Valtancoli, Inorg. Chim. Acta 2001,
317, 259–267.
[39]
[40]
[41]
A. E.Martell, R. M. Smith, R. J. Motekaitis, NIST Critically
Selected Stability Constants of Metal Complexes Database,
NIST Standard Reference Database, version 4, (1997).
a) A. Gasowska, L. Lomozik, R. Jastrzab, J. Inorg. Biochem.
2000, 78, 139–147; b) L. Lomozik, A. Gasowska, G. Krzysko,
J. Inorg. Biochem. 2006, 100, 1781–1789.
a) C. Anda, A. Llobet, V. Salvado, J. Reibenspies, R. J. Mote-
kaitis, A. E. Martell, Inorg. Chem. 2000, 39, 2986–2999; b) C.
Anda, A. Llobet, V. Salvado, A. E. Martell, R. J. Motekaitis,
Inorg. Chem. 2000, 39, 3000–3008.
M. T. Albelda, E. García-España, H. R. Jiménez, J. M. Llin-
ares, C. Soriano, A. Sornosa-Ten, B. Verdejo, Dalton Trans.
2006, 4474–4481.
E. M. Bianchi, S. A. A. Sajadi, B. Song, H. Sigel, Chem. Eur.
J. 2003, 9, 881–892.
H. Sigel, S. S. Massoud, N. A. Corfú, J. Am. Chem. Soc. 1994,
116, 2958–2971.
S. S. Massoud, R. Tribolet, H. Sigel, Eur. J. Biochem. 1990,
187, 387–393.
A. E. Martell, R. J. Motekaitis, Q. Lu, D. A. Nation, Polyhe-
dron 1999, 18, 3203–3218.
[23] L. H. Chen, C. S. Chung, Inorg. Chem. 1989, 28, 1402–1405.
[24] See for example: a) T. G. Fawcett, S. M. Rudich, B. H. Toby,
R. A. Lalancette, J. A. Potenza, H. J. Schugar, Inorg. Chem.
1980, 19, 940–945; b) Z. D. Georgousis, P. C. Christidis, D.
Hadjipavlou-Litina, C. A. Bolos, J. Mol. Struct. 2007, 837, 30–
37; c) M. Chadim, P. Díaz, E. García-España, J. Hodacova,
P. C. Junk, J. Latorre, J. M. Llinares, C. Soriano, J. Zavada,
New J. Chem. 2003, 27, 1132–1139; d) A. T. Chaviara, P. C.
Christidis, A. Papageorgiou, E. Chrysogelou, D. J. Hadjipav-
lou-Litina, C. A. Bolos, J. Inorg. Biochem. 2005, 99, 2102–2109;
e) P. Antunes, R. Delgado, M. G. B. Drew, V. Félix, H. Maecke,
Inorg. Chem. 2007, 46, 3144–3153.
[25] R. J. Deeth, J. Chem. Soc., Dalton Trans. 2001, 664–669.
[26] M. G. Basallote, J. Durán, M. J. Fernández-Trujillo, M. A.
Máñez, J. Chem. Soc., Dalton Trans. 1999, 3817–3823.
[27] M. G. Basallote, J. Durán, M. J. Fernández-Trujillo, M. A.
Máñez, Polyhedron 2001, 20, 75–82.
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[28] M. G. Basallote, J. Durán, M. J. Fernández-Trujillo, M. A.
Máñez, M. Quirós, J. M. Salas, Polyhedron 2001, 20, 297–305.
[29] M. G. Basallote, J. Durán, M. J. Fernández-Trujillo, M. A.
Máñez, J. Chem. Soc., Dalton Trans. 2002, 2074–2079.
[30] R. A. Read, D. W. Margerum, Inorg. Chem. 1981, 20, 3143–
3149.
E. García-España, M. J. Ballester, F. Lloret, J. M. Moratal, J.
Faus, A. Bianchi, J. Chem. Soc., Dalton Trans. 1988, 101–104.
M. Fontanelli, M. Micheloni, Proceedings of the I Spanish-Ital-
ian Congress on Thermodynamics of metal complexes, Diputa-
ción de Castellón, Castellón, Spain, 1990. Program for the
automatic control of the microburette and the acquisition of
the electromotive force readings.
[31] R. W. Hay, M. P. Pujari, R. Bembi, Transition Met. Chem.
1986, 11, 261–264.
[32] M. J. Fernández-Trujillo, B. Szpoganicz, M. A. Máñez, L. T.
Kist, M. G. Basallote, Polyhedron 1996, 15, 3511–3517.
[33] S. Siddiqui, R. E. Shepherd, Inorg. Chem. 1983, 22, 3726–3733.
[34] L. H. Chen, C. S. Chung, Inorg. Chem. 1988, 27, 1880–1883.
[35] A. Mendoza, J. Aguilar, M. G. Basallote, L. Gil, J. C.
Hernández, M. A. Máñez, E. García-España, L. Ruiz-Ramí-
rez, C. Soriano, B. Verdejo, Chem. Commun. 2003, 3032–3033.
[36] a) M. W. Hosseini, J.-M. Lehn, M. P. Mertes, Helv. Chim. Acta
1983, 66, 2454–2466; b) B. Dietrich, M. W. Hosseini, J.-M.
Lehn, R. B. Sessions, Helv. Chim. Acta 1985, 68, 289–299; c)
M. W. Hosseini, J.-M. Lehn, J. Chem. Soc., Chem. Commun.
1985, 1155–1157; d) M. W. Hosseini, J.-M. Lehn, L. Maggiora,
K. B. Mertes, M. P. Mertes, J. Am. Chem. Soc. 1987, 109, 537–
544; e) P. G. Yohannes, K. E. Plute, M. P. Mertes, K. B. Mertes,
Inorg. Chem. 1987, 26, 1751–1755; f) M. W. Hosseini, J.-M.
Lehn, J. Am. Chem. Soc. 1987, 109, 7047–7058; g) H. Jahan-
souz, Z. Jiang, R. H. Himes, M. P. Mertes, K. B. Mertes, J.
Am. Chem. Soc. 1985, 107, 8288–8289; h) R. C. Bethel, G.
Lowe, M. W. Hosseini, J.-M. Lehn, Bioorg. Chem. 1988, 16,
418–428; i) G. M. Blackburn, G. R. J. Thatcher, M. W. Hosse-
ini, J.-M. Lehn, Tetrahedron Lett. 1987, 28, 2779–2782; j) Z.
Jiang, P. Chalabi, K. B. Mertes, H. Jahansouz, R. H. Himes,
M. P. Mertes, Bioorg. Chem. 1989, 17, 313–329; k) M. W. Hos-
seini, A. J. Blacker, J.-M. Lehn, J. Am. Chem. Soc. 1990, 112,
3896–3904; l) A. V. Eliseev, H.-J. Schneider, J. Am. Chem. Soc.
1994, 116, 6081–6088.
[37] a) J. Aguilar, E. García-España, J. Guerrero, S. V. Luis, J. M.
Llinares, J. F. Miravet, J. A. Ramírez, C. Soriano, J. Chem.
Soc., Chem. Commun. 1995, 2237–2239; b) J. Aguilar, E.
García-España, J. A. Guerrero, S. V. Luis, J. M. Llinares, J. A.
Ramírez, C. Soriano, Inorg. Chim. Acta 1996, 246, 287–294; c)
J. Aguilar, B. Celda, V. Fusi, E. García-España, S. V. Luis,
M. C. Martínez, J. A. Ramírez, C. Soriano, R. Tejero, J. Chem.
Soc. Perkin Trans. 2 2000, 7, 1323–1328; d) J. Aguilar, A. B.
Descalzo, P. Díaz, V. Fusi, E. García-España, S. V. Luis, M.
Micheloni, J. A. Ramírez, P. Romani, C. Soriano, J. Chem. Soc.
Perkin Trans. 2 2000, 6, 1187–1192.
[49]
a) G. Gran, Analyst (London) 1952, 77, 661–671; b) F. I. Ros-
sotti, H. Rossotti, J. Chem. Educ. 1965, 42, 375.
P. Gans, A. Sabatini, A. Vacca, Talanta 1996, 43, 1739–1753.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery, T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tom-
asi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. H. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratch-
ian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Sal-
vador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Ba-
boul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Lia-
shenko, P.Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng,;A. Nanayakkara,
M.Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W.Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Revision
E.01, Gaussian, Inc., Pittsburgh PA, 2003.
A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
C. T. Lee, W. T. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–
789.
D. Andrae, U. Haussermann, M. Dolg, H. Stoll, H. Preuss,
Theor. Chim. Acta 1990, 77, 123–141.
V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995–2001.
M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput. Chem.
2003, 24, 669–681.
C. J. Cramer, D. G. Truhlar, Chem. Rev. 1999, 99, 2161–2200.
J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105,
2999–3093.
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
GLINT Software, Applied Photophysics Ltd., Leatherhead,
U. K., 1997.
[38] a) R. Hettich, H. J. Schneider, J. Am. Chem. Soc. 1997, 119,
5638–5647; b) S. Borah, M. S. Melvin, N. Lindquist, R. A.
Manderville, J. Am. Chem. Soc. 1998, 120, 4557–4562.
Received: June 8, 2008
Published Online: November 28, 2008
Eur. J. Inorg. Chem. 2009, 62–75
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
75