Modulation of the metal recognition properties of a new type of azaferrocenophane-based chemosensors: Co-ordination chemistry towards Mg 2+, Ca2+, Zn2+ and Ni2+

Article abstract of DOI:10.1039/b400953c

The structure-redox chemistry relationship of a new type of azaferrocenophane-based chemosensors, 3 and 4, in the presence of protons and several kinds of metal ions, has been studied. Electrochemical studies, carried out in CH₂Cl₂, in the presence of increasing amounts of Mg², Ca², Zn¹ and Ni² showed that the wave corresponding to the Fc/ Fc⁺ couple of the uncomplexed ligands is gradually replaced by a new reversible wave at more positive potentials corresponding to the Fc/Fc⁺ couple of the complexed ligands. The maximum shift of the ferrocene oxidation wave was found for 4b in the presence of Mg²⁺, whereas for 3f a selective sensing response for Mg² in the presence of hydrated Ca²⁺ cations was observed, with a concomitant highly visual output response consisting of a deep purple colour.

Full text of DOI:10.1039/b400953c

View Article Online / Journal Homepage / Table of Contents for this issue
Modulation of the metal recognition properties of a new type of
azaferrocenophane-based chemosensors: co-ordination chemistry
towards Mg2؉, Ca2؉, Zn2؉ and Ni2؉
Alberto Tárraga,* Pedro Molina,* Juan L. López and María D. Velasco
Departamento de Química Orgánica, Facultad de Química, Universidad de Murcia,
Campus de Espinardo, E-30100 Murcia, Spain. E-mail: pmolina@um.es;
Fax: ϩ34 968 364 149; Tel: ϩ34 968 367 496
Received 20th January 2004, Accepted 26th February 2004
First published as an Advance Article on the web 22nd March 2004
The structure–redox chemistry relationship of a new type of azaferrocenophane-based chemosensors, 3 and 4, in the
presence of protons and several kinds of metal ions, has been studied. Electrochemical studies, carried out in CH2Cl2,
in the presence of increasing amounts of Mg2ϩ, Ca2ϩ, Zn2ϩ and Ni2ϩ showed that the wave corresponding to the Fc/
Fcϩ couple of the uncomplexed ligands is gradually replaced by a new reversible wave at more positive potentials
corresponding to the Fc/Fcϩ couple of the complexed ligands. The maximum shift of the ferrocene oxidation wave
was found for 4b in the presence of Mg2ϩ, whereas for 3f a selective sensing response for Mg2ϩ in the presence of
hydrated Ca2ϩ cations was observed, with a concomitant highly visual output response consisting of a deep purple
colour.
As a continuation of our studies on the synthesis of aza-
Introduction
heteroaryl-substituted ferrocene derivatives displaying redox-
One of the most attractive ways of achieving sensor design is to
switchable character and electrochemical sensing behaviour30–32
functionalise a receptor capable of both selective substrate
binding with a metal centre and reporting on the recognition
(Scheme 1) we present here the synthesis and binding properties
of
a new range of 1,1Ј-[(3,4-dihydro-2,4-quinolinediyl)-
event through a variety of physical responses. In this context,
the design of redox-active receptors in which a change in
electrochemical behaviour can be used to monitor complex-
ation of guest species is an increasingly important area of
molecular recognition.1–10 Thus, from a synthetic standpoint,
ferrocene is a very convenient building block for redox-active
ligands as it can be relatively easily functionalized and
incorporated in many structures. These facts, coupled with its
electrochemical and UV-vis spectroscopic properties, which can
be perturbed by the proximity of bound guests, demonstrate
that ferrocene is a particularly attractive functional antenna in
this research area.
(2-hydroxy-1,2-ethanediyl)]ferrocenes, substituted at the
1 position, with aryl, heteroaryl or alkynyl groups. Additionally,
the correlation between electrochemical and optical properties
of these ferrocene derivatives has also been studied.
Most of the receptors containing a neutral ferrocenyl moiety
are substituted with macrocyclic ligands which exhibit interest-
ing electrochemical ion recognition effects because the com-
plexing ability of the receptors can be switched on and off by
varying the applied electrochemical potential. In general, cation
binding at an adjacent receptor site of a ferrocene-based host,
includes a positive shift in the redox potential of the ferrocene/
ferrocenium (Fc/Fcϩ) couple either by through-bond and/or
through-space electrostatic interactions, interference or con-
formational change.11 In addition, it has been suggested that an
added degree of recognition is conferred by appearance of a
new set of redox waves (two-wave behaviour) associated with
the oxidation of the ferrocenyl subunits in the host–guest
complex, compared with a single gradual shift in the potential
of the original ferrocene redox couple.5
Scheme 1
Results and discussion
Synthesis
The 1,5-difunctionalized [5]ferrocenophane derivatives 3 were
prepared from [5]ferrocenophane 1, readily available in 39%
overall yield from 1,1Ј-diacetylferrocene by sequential treat-
ment with o-azidobenzaldehyde and tri-n-butylphosphane,33
and the corresponding organolithium derivative 2, in THF
at Ϫ30 ЊC. Purification by column chromatography afforded
compounds 3 in 55–80% yield as orange solids (Scheme 2).
Electrochemical studies
Most of the reported ferrocene-based redox-responsive
ionophores feature crown ethers or polyaza-macrocycles as
receptor units, due to the great possibility of modulation of
host co-ordination environments that they can offer by tailor-
ing the cavity size. However, no selectivity towards magnesium
or calcium ions has been reported for these systems.12–24 In this
context, a number of redox-active ferrocene based receptors
such as imine-,25 oxazoline-,26–27 imidazoline-,28 and pyridine-
functionalized29 ferrocene ligands have been described, which,
in organic solvents, selectively respond to, but do not clearly
differentiate between, Mg2ϩ and Ca2ϩ ions with no interference
from a large excess of Liϩ, Naϩ and Kϩ.
Firstly, the electrochemical behaviour of ferrocene ligands 3 on
their own, as well as in the presence of variable concentrations
of HBF4, was investigated. Thus, upon protonation by addition
of stoichiometric amounts of HBF4 in CH2Cl2 to a solution of
ligands in CH2Cl2 all the alcohols 3a–3f exhibited similar
behaviour displaying in their corresponding voltammograms
(CVs) the characteristic two-wave behaviour associated with
the high difference between the half-wave potentials of the free
ligand and the protonated species redox couples (∆E1/2
=
E1/2 Ϫ E1/2 free). The magnitude of the electrochemical
protonated
shift (∆E1/2) on protonation provides important thermo-
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 2 0 0 4
D a l t o n T r a n s . , 2 0 0 4 , 1 1 5 9 – 1 1 6 5
1159

View Article Online
Table 1 Electrochemical proton-dependence of 3 and 4
Ligand L
E1/2L/Va
E1/2LHϩ/Va
∆E1/2/mV
BEF
RCE
3a
3b
3c
3d
0.120
0.160
0.180
0.070
0.190
0.190
0.190
0.100
0.110
0.510c
0.740c
0.510
0.550
0.570
0.110
0.610
0.610
0.580
0.490
0.510
0.540
1.080
390
390
390
40b
420
420
390
390
400
30b
340
2.6 × 10Ϫ7
2.6 × 10Ϫ7
2.6 × 10Ϫ7
2.6 × 10Ϫ1
8.0 × 10Ϫ8
8.0 × 10Ϫ8
2.6 × 10Ϫ7
2.6 × 10Ϫ7
1.7 × 10Ϫ7
3.1 × 10Ϫ1
1.8 × 10Ϫ6
3.9 × 106
3.9 × 106
3.9 × 106
4.7
1.3 × 107
1.3 × 107
3.9 × 106
3.9 × 106
5.8 × 106
3.2
3e
3f
4a
4b
4c
5.6 × 105
a E1/2 are given in CH2Cl2 versus Fcϩ/Fc as internal standard. b E1/2 for the monosubstituted ferrocene unit. c E1/2 are given in CH2Cl2 versus DMFcϩ/
DMFc as internal standard (DMFc = decamethylferrocene). When the experiment was carried out in CH3CN/CH2Cl2 (3:2), the values obtained
were: E1/2 (monosubstituted Fc) = 0.540 V and E1/2 (disubstituted Fc) = 0.760 V, vs DMFc.
ferrocenyl unit induces some thermodynamic instability to the
protonated species with reference to those in which the sub-
stituent R has a small steric volume.
When the same voltamperometric study was performed with
compounds 3c–3f, in which the presence of an acetylenic
moiety moves away the bulky group R from the C᎐N unit,
᎐
it was observed that the ∆E1/2 = E1/2 protonated ϪE1/2 free for 3d (420
mV) is larger than those observed for 3c and 3f, although the
steric hindrance of the monosubstituted ferrocenyl unit present
in 3d is larger than that of the substituents present in 3c–3f.
Therefore, the value of the redox shifts observed would depend
not only on the steric effect of the substituent at the 1 position
of the ferrocenophane bridge but also on some other electronic
factors.
Scheme 2
Thus, to gain a better understanding into the factors
which determine the extent of the redox shifts in these type of
ferrocene derivatives upon protonation, and with the aim of
dynamic information. The shift in redox potential on proton-
ation is related to the ratio of protonation constant for the
equation: ∆E1/2 = (RT/nF )ln(Kred/Kox), where Kred is the proton
binding constant for neutral unoxidised redox responsive ligand
and Kox is the proton binding constant for its oxidised form.
The quantity Kred/Kox has been defined by Beer et al.7–9 as reac-
tion coupling efficiency (RCE), whereas the quantity Kox/Kred
has been defined as binding enhancement factor (BEF), and
represents a quantitative measure of the perturbation of the
redox centre induced by protonation to the receptor unit. The
results obtained are reported in Table 1. Although these
alcohols exhibit large binding constants and consequently large
RCE values, ranging from 3.9 × 106 to 1.3 × 107, they present
some differences which deserve some comment.
The CVs of 3a and 3b (R = Ph and Thy), and their corre-
sponding protonated derivatives, show a reversible one-electron
redox couple giving rise to a ∆E1/2 = 390 mV. This value is
almost the same as those obtained by using the already
described derivatives 4a (∆E1/2 = 390 mV) or 4b (∆E1/2 = 400
mV), respectively34 (Scheme 3). As a consequence, similar
RCE values are also observed in these cases. However, it is
noteworthy that when the same comparison is performed with
compound 4c35 the ∆E1/2 corresponding to the disubstituted
ferrocenyl moiety is smaller and the corresponding RCE value
decreases by a factor of ∼10. These observed differences can be
explained bearing in mind that in the latter system the steric
establishing a correlation between the distances C᎐N–mono-
᎐
substituted ferrocene and the positive shifts of these redox
potentials, we have investigated more closely the effect of the
protonation of the nitrogen atom in the C᎐N group on the
᎐
redox potentials of the structurally related homobimetallic
derivatives 4c (R ؍ Fc) and 3d. While the CV of 4c shows two
reversible one-electron oxidations for the two ferrocenyl groups
(∆E1/2 = 230 mV),35 compound 3d also exhibits two reversible
ferrocene couples but more closely spaced (∆E1/2 = 120 mV),
which indicates a less intermetallic interaction between the two
ferrocene moieties.36–38 Remarkably, upon protonation, the
anodic electrochemical shift for both compounds was larger
than for the free ligands: 540 mV, for 4c, and 500 mV, for 3d,
respectively (Fig. 1, Table 1).
In compounds 3d and 4c, bearing two ferrocene units, the
small electrochemical shifts of the monosubstituted ferrocene
(40 and 30 mV, respectively) and the large values obtained for
the disubstituted ferrocenyl moiety (420 and 340 mV) clearly
indicate that while the protonation process does not affect the
E1/2 of the monosubstituted ferrocenyl unit, it strongly affects
the E1/2 of the disubstituted ferrocene. Moreover, it is possible
to conclude that the E1/2 of the latter is influenced by interaction
of such a disubstituted unit with the electrogenerated mono-
substituted ferrocenium cation as well as with the C᎐N ؒ ؒ ؒ Hϩ,
᎐
interaction between the C᎐N moiety and the monosubstituted
᎐
after protonation. In addition, the difference in the RCE values
for both compounds could be explained on the basis of the
bigger steric hindrance existing in 4c in comparison to 3d.
The behaviour of the receptors 3e and 3f in the presence of
HBF4 was also investigated, the effect of this strong acid being
protonation of the C᎐N as well as the pyridine nitrogen atom.
᎐
In each case, the anodic redox potential shifts of the ferrocene
nucleus were 420 and 390 mV, respectively (Table 1). This
difference could probably be attributed to the higher electro-
static influence of the positive pyridinium cation, formed after
protonation of 3e, on the ferrocene redox potential, in com-
Scheme 3
D a l t o n T r a n s . , 2 0 0 4 , 1 1 5 9 – 1 1 6 5
1160

View Article Online
that a co-ordination process is taking place, the CVs obtained
after the addition of water should be the same as those corre-
sponding to the free ligand, because in such a case the excess
water in the media can displace the ligand from the metal
co-ordination sphere. However, when a protonation process
occurs, the addition of water promotes a slightly cathodic shift
of the redox wave which does not agree with that of the free
ligand. Although similar criteria have been used previously
to differentiate both types of processes, in some cases its
application is incorrect in the sense that the protonation process
is ruled out when the UV-vis spectra and the CVs obtained for
the ligand in organic solvents do not change after addition of
water to the media, suggesting that the protonation process is
only due to the added water26–27 and not to the presence of the
ionisable protons in the hydrated metal salts.
The results obtained by using the ligands 3b–d and 4b–c
revealed the following facts: (i) the appearance of a second wave
together with that corresponding to the free ligand on stepwise
addition of substoichiometric amounts of the appropriate ion,
which reflects the two-wave behaviour in all the cases studied
and additionally indicates large values of the corresponding
complexation constants; (ii) in all cases, a reduction wave was
observed as a consequence of the generation of a polarized
C᎐N ؒ ؒ ؒ Hϩ(Mnϩ) due to the interaction of the C᎐N group
᎐
᎐
with the proton or the metal ion; (iii) only protonation pro-
cesses were observed when [Ca(ClO4)2]ؒ4H2O was used; (iv) by
using [Zn(ClO4)2]ؒ6H2O and [Ni(ClO4)2]ؒ4H2O, the general
effect observed is protonation except for ligand 3b, bearing
a S atom, which gives rise to a co-ordination behaviour and
which shows selectivity between Zn2ϩ and Ni2ϩ to some extent
(Table 2); (v) the results obtained by using anhydrous salts
of Ca2ϩ and Mg2ϩ are quite similar; (vi) addition of Liϩ induces
no significant variation in the corresponding CVs of the free
ligands studied.
From the receptors considered, the potential shifts obtained
on co-ordination or protonation are very similar, which
indicates not only that the interaction of the binding site with
the proton or the metal cations is almost of the same intensity
but also that co-ordination or protonation is being performed
on the same donor atoms. Otherwise, lower potential shifts
promoted by the co-ordination processes, with respect to those
of protonation, should be expected.
On the other hand, ligand 3b (R = Thy), possessing a second
donor atom in the structure, gives rise to metal-ion-co-
ordination processes when either anhydrous or hydrated salts
of the cations studied are added. Only in the case of hydrated
Ca2ϩ salts, is a slow protonation response observed. These facts
indicate that the presence of a second donor atom in the mole-
cule should increase the binding strength of metal ions to this
receptor and, as a consequence, their corresponding complex-
ation constants. However, and in order to favour the effective
and simultaneous co-ordination of two different donor atoms
to one metal centre, it would be necessary for the cation to
be located far away from the redox centre and therefore the
electrostatic interactions of the cation–redox centre and those
Fig. 1 (a) Cyclic voltammogram of compound 3d before (black) and
after addition (red) of 1 equiv. of HBF4. Conditions: 0.5 mM of 3d and
0.1 M [nBu4NClO4] in CH2Cl2, using a Pt disk electrode and a scan rate
= 0.2 V sϪ1. (b) DPV of compound 3d Conditions: 0.5 mM of 3d and 0.1
M [nBu4NClO4] in CH2Cl2, using a Pt disk electrode, a Epulse = 10 mV
and a scan rate = 0.004 V sϪ1
.
parison with the same situation in 3f. The isolation as a solid of
the protonated form of 3e confirms this assumption.The CV is
identical to those obtained in the electrochemical cell and in the
1H-NMR spectrum the pyridine ring protons are downfield
shifted (δ = ϩ0.41 and ϩ0.20 ppm) appearing in the region
8.10–8.63 ppm characteristic of the pyridinium ring protons.
All these data indicate that the prepared new ferroceno-
phanes behave as excellent electrochemical sensors for protons
not only due to the basic character of their C᎐N bond but also
᎐
due to the existence of an efficient communication between
the binding site and the ferrocene nucleus as a result of a com-
bination of through-space electrostatic and through-conjugated
bond interactions. The RCE values shown in Table 1 indicate a
measure of the difficulty to protonate the oxidized form of the
ligand with reference to its reduced form.
It is worth noting that all the protonated ligands exhibited an
irreversible reduction wave at E1/2 = Ϫ0.43 V, which is absent in
the free ligand and is attributed to the reduction of the polar-
ized C᎐N, because of the stabilization of the LUMO, localized
᎐
on the imine group, which makes its reduction easier with
respect to the neutral form.
The effect of several metal ions (Liϩ, Mg2ϩ, Ca2ϩ, Ni2ϩ and
Zn2ϩ) on the receptor redox chemistry has also been investi-
gated, by using different types of anhydrous and hydrated
inorganic salts such as: [Ca(ClO4)2]ؒ4H2O, [Zn(ClO4)2]ؒ6H2O,
[Ni(ClO4)2]ؒ4H2O, LiClO4, Mg(ClO4)2, Zn(CF3SO3)2 and
Ca(CF3SO3)2. It is very well established that the metal ions
under study could exhibit two relevant properties in non-
aqueous solution: co-ordination of the metal ions by the ligand
and protonation of the ligand due to the use of hydrated metal
salts, bearing ionisable protons.39 It is therefore of great
importance to establish in each case whether co-ordination or
protonation effects are being observed and as a consequence it
is necessary to adopt some criteria that allow us to differentiate
experimentally between the co-ordination and protonation
processes. The criteria which we have used are based on the
different behaviour exhibited by the electrogenerated species
after the addition of water to that media. Therefore, in the case
through the C᎐N would be less effective. In other words: the
᎐
bigger the stability of the complex formed, the smaller the
efficiency of communication with the redox centre. Con-
sequently, the potential shifts of the ferrocene nucleus on
co-ordination are smaller than on protonation, because in the
latter case the proton should be preferentially bound to the
more basic C᎐N group. These data also suggest that although in
᎐
the quinolinophane framework, present in all the derivatives
studied, there are two different types of donor atoms (N and
O), the complexation processess should take place through the
C᎐N fragment.
᎐
It is worth noting that this data obtained from the voltam-
metric studies completely agreed with that obtained from the
X-ray crystallographic data of compound 4c35 which demon-
strates that the orientation of the –OH and C᎐N groups is not
᎐
D a l t o n T r a n s . , 2 0 0 4 , 1 1 5 9 – 1 1 6 5
1161

View Article Online
Table 2 Electrochemical cation-dependence of 3b–d and 4b–c
Ligand L
E1/2L/Va
E1/2 complex/Va
∆E1/2/mV
RCE
Effect
3bؒCa2؉
3bؒMg2؉
3bؒZn2؉ b
3bؒNi2؉ b
3cؒCa2؉
3cؒMg2؉
3cؒZn2؉ b
3dؒCa2؉
0.160
0.160
0.160
0.160
0.180
0.180
0.180
0.070
0.190
0.080
0.190
0.110
0.110
0.110
0.107
0.540e
0.760e
0.540e
0.760e
0.510
0.530
0.550
0.510
0.560
0.560
0.570
0.080
0.570
0.080
0.570
0.500
0.510
0.510
0.500
0.570
1.100
0.570
1.100
350
370
390
350
380
380c
390c
10c
380c
0c
8.2 × 105
1.8 × 106
3.9 × 106
7.0 × 105
8.2 × 106
2.7 × 106
2.7 × 106
1.5
Co-ordination
Co-ordination
Co-ordination
Co-ordination
Co-ordination
Co-ordination
Protonation
Co-ordination
3.0 × 106
––
3dؒMg2؉
Co-ordination
380c
390
400
400
390
30
2.7 × 106
3.9 × 106
5.8 × 106
5.8 × 106
3.9 × 106
3.2
4bؒCa2؉
4bؒMg2؉
4bؒZn2؉ b
4bؒNi2؉ b
4cؒMg2؉ d
Co-ordination
Co-ordination
Protonation
Protonation
Co-ordination
340
30
340
5.6 × 105
3.22
4cؒZn2؉
Protonation
5.6 × 105
a E1/2 are given in CH3CN/CH2Cl2 (3/2) versus Fcϩ/Fc as internal standard. b Hydrated M2ϩ
.
c In CH2Cl2. d In the presence of [Ni(ClO4)2]ؒ4H2O the
e
electrogenerated species are unstable in the timescale of the technique. E1/2 are given in CH3CN/CH2Cl2 (3/2) versus DMFcϩ/DMFc as internal
standard.
Table 3 Electrochemical cation-dependence of 3e and 3f
Ligand L
E1/2L/Va
E1/2 complex/Va
∆E1/2/mV
RCE
Effect
3eؒCa2ϩ b
3eؒCa2ϩ d
3eؒMg2ϩ
3eؒZn2ϩ b
3eؒZn2ϩ d
3fؒCa2ϩ b
3fؒCa2ϩ d
3fؒMg2ϩ
3fؒZn2ϩ b
3fؒZn2ϩ d
0.190
0.190
0.190
0.190
0.190
—
0.190
0.190
0.190
0,190
0.490
0.480
0.480
0.540
0.510
––
0.450
0.450
0.550
0.475
300
290
290
350
320
––
260
260
360
285
1.2 × 105
8.0 × 104
8.0 × 104
8.2 × 105
2.3 × 105
––
Co-ordinationc
Co-ordination
Co-ordination
Co-ordinationc
Co-ordination
No response
Co-ordination
Co-ordination
Protonationc
Co-ordination
2.5 × 104
2.5 × 104
1.2 × 106
6.6 × 104
a E1/2 are given in CH2Cl2 versus Fcϩ/Fc as internal standard. b [Ca(ClO4)2]ؒ4H2O and [Zn(ClO4)2]ؒ6H2O are the hydrated salts used. c This is a slow
.
process which needs the presence of a larger amount of M2ϩ to respond, in comparison with the case of using Mg2ϩ d Anhydrous Ca(CF3SO3)2 or
Zn(CF3SO3)2
appropriate for the simultaneous co-ordination of the metal,
process (Fig. 3). These results provide clear evidence of the
greater stability of the complex formed, overcoming the
hydration energy of the Ca2ϩ, in the case of using [Ca(ClO4)2]ؒ
and therefore the C᎐N is the centre mainly involved in the
᎐
binding processes.
At this point, and taking into account the above results, it
seemed interesting to us to test the complexation properties of
ligands 3e and 3f, with the characteristic feature of having a
pyridine unit linked to the ferrocenophane framework through
an acetylenic bond. In these cases, the nitrogen atom within
the heterocyclic ring is located far away from the nitrogen in the
C᎐N bond, allowing the influence of the presence of the pyr-
᎐
idine nitrogen atom on the redox potential shift of the ferrocene
nucleus and on the corresponding complexation constants to be
studied.
The cyclic voltammetric studies performed on addition of
Mg(ClO4)2, [Ca(ClO4)2]ؒ4H2O, [Zn(ClO4)2]ؒ6H2O, Zn(CF3SO3)2
and Ca(CF3SO3)2, are summarized in Table 3. Firstly, these data
indicate that the presence of the pyridine donor nitrogen atom
promotes a notable differentiation between the anodic potential
shifts of the ferrocene nucleus induced by the co-ordination
processes in comparison with those induced by protonation
(Fig. 2). Moreover, the magnitude of such electrochemical shifts
on protonation is larger than those induced by co-ordination,
which is in agreement with the above mentioned results.
For the molecule 3e, containing the 2-pyridyl unit, the
potential redox shift observed on addition of either anhydrous
or hydrated Ca2ϩ salts are the same. Nevertheless, in the latter
case, the response in CV depends greatly on the concentration
of added cation probably indicating a slower complexation
Fig. 2 Cyclic voltammogram of compound 3f, before (red) and after
addition of 0.5 equiv. of Mg(ClO)4 (green); after addition of 1 equiv. of
HBF4 (blue). Conditions: 0.5 mM of 3f and 0.1 M [nBu4NClO4] in
CH2Cl2, using a Pt disk electrode and a scan rate = 0.2 V sϪ1
.
D a l t o n T r a n s . , 2 0 0 4 , 1 1 5 9 – 1 1 6 5
1162

View Article Online
case of 3f, where the pyridine nitrogen atom is located at a bigger
distance from the C᎐N than in 3e, no variation of the CV was
᎐
observed upon addition of hydrated Ca2ϩ, and only a proton-
ation effect is observed when hydrated Zn2ϩ salts were added.
Spectrophotometric studies
Previous studies on ferrocene40–43 ligands have shown that a
characteristic band between 400 and 500 nm ascribed to
the lowest energy metal–ligand transition is perturbed by
complexation.44
Therefore, UV-vis spectroscopy was used for further com-
plexation experiments with 3f, because as has already been
mentioned this ligand exhibits a different electrochemical
response upon addition of Ca2ϩ and Mg2ϩ. Typically, upon
addition of Mg2ϩ, as the chelating metal ion, into a CH2Cl2
solution of 3f (c = 6.3 × 10Ϫ4 M), the MLCT band at λmax = 462
nm (ε = 1148.10 l molϪ1 cmϪ1), entirely disappeared and was
replaced by a new band at λmax = 525 nm (ε = 3475.16 l molϪ1
cmϪ1), accompanied by an increase in absorbance (Fig. 4).
Furthermore, its colour in the CH2Cl2 solution changed from
orange to purple, which can be used for “naked eye” detection
of Mg2ϩ. In the presence of Ca2ϩ cation the colour change was
less striking and only a modest darkening of the solution was
observed visually.
Fig. 3 DPV of 3e after addition of 3 equiv. of Ca(ClO4)2ؒ4(H2O)
(black); after addition of 3 equiv. of anhydrous Ca(CF3SO3)2 (blue);
after addition of 1 equiv. of HBF4 to the resulting solution (red).
Conditions: 0.5 mM of 3e and 0.1 M [nBu4NClO4] in CH2Cl2, using a Pt
disk electrode, a Epulse = 10 mV, a scan rate = 0.004 V sϪ1, and Fcϩ/Fc as
internal standard.
4H2O; and which was not observed in the previously studied
derivatives.
A similar behaviour to the above mentioned for Ca2ϩ is
observed when Mg(ClO4)2, or [Zn(ClO4)2]ؒ6H2O was added to
an electrochemical solution of 3e (Table 3) but in the latter case,
together with the usual positive shift in the redox potential,
a new binding response is observed, which is associated to a
protonation process. This fact has been experimentally proven
by the addition of 1 equiv. of HBF4 to the electrochemical
solution which resulted in the appearance of the redox wave,
corresponding to that assigned to the protonated species,
with concomitant disappearance of the redox wave due to co-
ordination. Futhermore, in this case the redox wave of the
ferrocene is shifted to the same potential value as that obtained
by a direct protonation process (∆E1/2 = 420 mV) (Table 1).
However, when anhydrous Zn(CF3SO3)2 is added only the
co-ordination process is observed (Table 3).
Fig. 4 UV-vis spectra of the titration of alcohol neutral 3f with
Mg(ClO4)2 in CH2Cl2 (c = 6.3 10Ϫ4 mol. dmϪ3). The initial spectrum is
that of starting alcohol 3f and the final spectrum corresponds to the
complexed form. Arrows indicate the absorptions that increased (up)
and decreased (down) during the experiment.
For comparison, we have also studied the electrochemical
response to the same set of metal cations of 3f, in order to
determine not only the effect of these cations on the ligand but
also the effect of the host–guest interaction on the oxidation
potential of the ferrocene moiety upon replacing a 2-pyridyl
by a 3-pyridyl unit. It is noteworthy that in contrast to the
ligand 3e, different behaviour was observed when hydrated or
anhydrous salts of Ca2ϩ were added to a solution of 3f (Table
3), because while no response was obtained with the former, a
co-ordination effect was observed with the latter. Furthermore,
upon addition of [Zn(ClO4)2]ؒ6H2O a protonation effect was
observed, whilst under the same conditions ligand 3e gives
rise to a co-ordination effect. However, addition of either
anhydrous Mg(ClO4)2 or Zn(CF3SO3)2 results in characteristic
co-ordination behaviour.
The association constant of the 3fؒMg2ϩ complex (Ka = 1.5 ×
1010
M
Ϫ2) could be evaluated by means of UV-vis spectro-
photometric titrimetry.45 An isosbestic point was found,
strongly indicating the presence of a unique complex in
solution. A Job plot experiment between 3f and Mg(ClO4)2 in
CH2Cl2 at 25 ЊC revealed a 1:2 (LM2) stoichiometry.
Conclusion
In summary, we have developed new azaferrocenophane deriv-
atives 3 and 4, which are efficient redox sensors for protons and
several kinds of metal cations. This recognition process is easily
monitored by CV, which displays a large positive shift in the
value of the Fc/Fcϩ redox couple (400 mV for 4b in the presence
of Mg2ϩ cations). To the best of our knowledge the ferrocene-
based ligand 3f is the first example of a dual electrochemical–
optical Mg2ϩ ion sensor in which the Fc/Fcϩ redox couple
is shifted (∆E1/2 = 260 mV) by cation recognition with con-
comitant naked eye detectable colour change from orange to
purple. Interestingly, the electrochemical response of this ligand
towards Ca2ϩ cation depends strongly on the nature of the salt
employed: no response in the presence of hydrated salts and a
two-wave behaviour when anhydrous salts are used.
Moreover, from the voltammetric behaviour of these deriv-
atives it is clear that the binding events are strongly affected by
the distance between the two binding sites as was previously
suggested for the case of 3b. This argument can also be
supported by the analysis of the evolution of the reduction
potential of the C᎐N unit after complexation, present in the
᎐
structure of these derivatives, because its magnitude should
be related to the distance C᎐N ؒ ؒ ؒ Mϩn: the bigger the
᎐
C᎐N ؒ ؒ ؒ Mϩn interaction, the easier the reduction of the C᎐N.
᎐
᎐
In this context, the reduction potentials of the C᎐N group
᎐
in 4c and 3f upon addition of Mg2ϩ, demonstrate that 3f
(Ep = Ϫ0.670 V) exhibits a more negative value than 4c (Ep =
Ϫ0.430 V), which is in agreement with a small C᎐N ؒ ؒ ؒ Mg2ϩ
᎐
interaction in 3f.
Experimental
All reactions were carried out under N2 using solvents which
were dried by routine procedures. All melting points were
On the other hand, the voltamperometric data also reveal
that the stability of the complex formed is dependent on the
relative position of the pyridine nitrogen atom, because in the
D a l t o n T r a n s . , 2 0 0 4 , 1 1 5 9 – 1 1 6 5
1163

View Article Online
determined on a Kofler hot-plate melting point apparatus
and are uncorrected. IR spectra were determined as Nujol
emulsions or films on a Nicolet Impact 400 spectrophotometer.
1H- and 13C-NMR spectra were recorded on a Bruker AC200,
300 or 400 MHz, and chemical shifts refer to signals of tetra-
methylsilane. The EI and FABϩ mass spectra were recorded on
a Fisons AUTOSPEC 500 VG spectrometer, using 3-nitro-
benzyl alcohol as matrix. The UV-vis spectra were run on a
Varian Cary 5000 UV-vis–NIR spectrophotometer. Micro-
analyses were performed on a Perkin–Elmer 240C instrument.
Electrochemical measurements were taken with a QUICELT-
RON potentiostat/galvanostat controlled by a personal com-
puter and driven by dedicated software. Electrochemical
experiments were conducted in a conventional three-electrode
cell under a N2 atmosphere at 25 ЊC. The working electrode was
a Pt disk (1 mm in diameter) polished before each recording.
The auxiliary electrode was a platinum wire. The reference
electrode was SCE. The experiments were in acetonitrile or
Cp), 4.73 (1H, m, Cp), 5.1 (1H, m, Cp), 6.61 (1H, d, J = 2.7 Hz),
6.74 (1H, t, J = 5.1 Hz), 7.06–7.08 (1H, m), 7.12 (1H, d, J = 7.2
Hz), 7.18 (1H, d, J = 7.8 Hz), 7.27 (1H, m), 7.38 (1H, d, J = 8.1
Hz); δC (CDCl3) 31.17 (CH2), 32.04 (CH), 43.03 (CH2), 64.97
(CH, Cp), 65.91 (q), 66.58 (CH, Cp), 67.37 (CH, Cp), 67.93
(CH, Cp), 69.43 (CH, Cp), 70.65 (CH, Cp), 71.25 (CH, Cp),
72.88 (CH, Cp), 86.57 (qipso Fc), 97.54 (qipso Fc), 122.90 (CH),
124.64 (CH), 125.56 (CH), 125.96 (CH), 125.96 (CH), 126.41
(CH), 126.46 (CH), 127.86 (CH), 132.10 (q), 145.40 (q), 158.50
(q), 168.50 (C᎐N); EIMS m/z 439 (Mϩ, 100%), 421 (21), 248
᎐
(55), 222 (10), 191 (22).
(3c): (70%) mp >300 ЊC (from CH2Cl2/Et2O) (Found: C,
74.85; H, 5.10; N, 3.22. C29H23FeNO requires C, 76.16; H, 5.07;
N, 3.06); νmax/cmϪ1: 3389, 2200, 1582, 1562, 1472, 1450, 1240,
1090, 936, 822 (Nujol); δH (CDCl3) 2.15 (1H, t, Jgem = 12.35 Hz,
3J = 12.35 Hz), 2.36 (1H, s), 2.47 (1H, dd, Jgem = 14.4 Hz,
3J = 5.71 Hz) 2.58 (1H, dd, Jgem = 14.8 Hz, 3J = 5.00 Hz), 3.40–
3.42 (1H, m), 3.49 (1H, d, Jgem = 13.32 Hz) 4.14 (1H, m, Cp),
4.25 (2H, m, Cp), 4.41 (1H, m, Cp), 4.48 (1H, m, Cp), 4.57 (1H,
m, Cp), 4.68 (1H, m, Cp), 5.06 (1H, m, Cp), 7.14–7.43 (9H, m);
δC (CDCl3) 31.19 (CH2), 32.2 (CH), 44.03 (CH2), 62,79 (q),
65.36 (CH, Cp), 65.52 (CH, Cp), 66.75 (CH, Cp), 67.43 (CH,
Cp), 69.11 (CH, Cp), 70.12 (CH, Cp), 70.70 (CH, Cp), 72.60
(CH, Cp), 80.40(qipso, Fc), 86.52 (q), 96.97 (qipso, Fc), 96.58 (q),
122.27 (CH), 125.23 (CH), 126.38 (CH), 128.36 (CH), 128.58 (2
× CH), 128.72 (q), 131.02 (2 CH), 131.93 (q), 145.06 (q), 167.55
dichloromethane solutions containing 0.1 M [NBun ]ClO4
4
(warning: caution) as supporting electrolyte and under these
experimental conditions, the ferrocenium/ferrocene couple was
observed at ϩ0.405 V vs. SCE in acetonitrile and ϩ0.535 V vs.
SCE in dichloromethane. Deoxygenation of the solutions was
achieved by bubbling nitrogen for at least 10 min and the work-
ing electrode was cleaned after each run. Cyclic voltammetry
(CV) curves were recorded at scan rates from 0.050 to 1 V sϪ1
and the scan potential was obtained in both positive and neg-
ative directions. The differential pulse voltammetry (DPV)
curves were recorded at a 0.004 V sϪ1 scan rate with a pulse
height of 10 mV and a step time of 50 ms.
(C᎐N); EIMS m/z 457 (Mϩ, 26%), 439 (100), 355 (28), 248 (41),
᎐
222 (7), 191 (8).
(3d): (62%) mp 205–208 ЊC (from CH2Cl2/Et2O) (Found: C,
70.31; H, 4.70; N, 2.39. C33H27Fe2NO requires C, 70.12; H, 4.81;
N, 2.48); νmax/cmϪ1: 3300, 2200, 2222, 1582, 1558, 1462, 1454,
1240, 1106, 866, 820 (Nujol); δH (CDCl3) 1.94 (1H, t, Jgem = 13.8
General procedure for the preparation of 1,1Ј-[(3,4-dihydro-2,4-
quinolinediyl)(2-alkynyl/aryl or heteroaryl-2-hydroxy-1,2-
ethanediyl)]ferrocene 3
3
3
Hz, J = 13.8 Hz), 2.26 (dd, Jgem = 13.5 Hz, J = 5.1 Hz), 2.32
3
(1H, dd, Jgem = 14.1 Hz, J = 4.5 Hz), 3.45–3.52 (2H, m), 4.12
(5H, s, Cp), 4.20 (3H, m, Cp), 4.31 (4H, m, Cp), 4.47 (1H, m,
Cp), 4.55 (1H, m, Cp), 4.67 (1H, m, Cp), 4.75 (1H, m, Cp), 4.88
(1H, m, Cp), 6.44 (1H, s, OH), 7.10–7.27 (4H, m); δC (CDCl3)
30.20 (CH), 30.38 (CH2), 43.17 (CH2), 62.77 (q), 65.27 (CH,
Cp), 65.57 (CH, Cp), 66.57 (CH, Cp), 67.21 (CH, Cp), 68.39 (2
× CH, Cp), 68.77 (CH, Cp), 69.42 (5 CH, Cp), 69.89 (CH, Cp),
77.45 (CH, Cp), 77.62 (CH, Cp), 70.66 (CH, Cp), 72.36 (CH,
Cp), 79.12 (qipso, Fc), 86.50 (qipso, Fc), 93.46 (q), 97.22 (q),
125.08 (2 × CH), 126.29 (CH), 127.20 (CH), 131.94 (q), 145.01
To a cooled (Ϫ30 ЊC) solution of 1,1Ј-[(3,4-dihydro-2,4-quin-
olinediyl)(2-oxo-1,2-ethanediyl)] ferrocene 1 (0.1 g, 0.28 mmol)
in dry Et2O (30 mL), the appropriate organolithium derivative
(1.5 equiv.) was added, under nitrogen. The stirred solution was
then allowed to reach room temperature (3 h) and then H2O (30
mL) was added. The reaction mixture was extracted with
CH2Cl2 (3 × 50 mL) and the combined organic layers were dried
over anhydrous Na2SO4, filtered off and evaporated under
reduced pressure. The crude product was then chromato-
graphed on a deactivated silica gel column using EtOAc/n-
hexane (1:1) as eluent to yield the final product.
(q), 167.34 (C᎐N); FABMS m/z 566 (Mϩ ϩ 1, 13%), 565
᎐
(Mϩ, 17%).
(3e): (55%) mp >320 ЊC (from CH2Cl2/Et2O) (Found: C,
73.55; H, 4.60; N, 6.23. C28H22FeN2O requires C, 73.37; H, 4.84;
N, 6.11); νmax/cmϪ1: 3349, 2205, 1582, 1558, 1464, 1270, 1094,
1072, 906, 822 (Nujol); δH (CDCl3) 1.91 (1H, t, Jgem = 13.8 Hz,
(3a): (80%) mp 223–226 ЊC (from EtOAc/n-hexane) (Found:
C, 74.58; H, 5.50; N, 3.18. C27H23FeNO requires C, 74.84; H,
5.35; N, 3.23). νmax/cmϪ1: 3390, 1580, 1565, 1440, 1120, 1095,
3
740 (Nujol); δH (CDCl3) 2.04 (1H, s, OH), 2.26 (1H, dd, Jgem
=
3J = 13.8 Hz), 2.23 (1H, dd, Jgem = 14.1 Hz, J = 4.8 Hz), 2.34
14.4 Hz, 3J = 11.7 Hz), 2.42 (1H, dd, Jgem = 14.4 Hz, 3J = 4.8 Hz),
(1H, dd, Jgem = 13.8 Hz, J = 5.4 Hz), 3.26–3.5 (2H, m), 4.13
3
3
2.53 (1H, dd, Jgem = 14.1 Hz, J = 5.1 Hz), 3.34–3.36 (1H, m),
(1H, m, Cp), 4.3 (2H, m, Cp), 4.45 (1H, m, Cp), 4.52 (1H, m,
Cp), 4.63 (1H, m, Cp), 4.73 (1H, m, Cp), 4.84 (1H, m, Cp), 6.35
(1H, s, OH), 7.1–7.31 (5H, m), 7.7 (1H, td, J = 7.8 Hz, J = 1.8
Hz), 8.43 (1H, d, J = 5.4 Hz); δC (CDCl3) 35.42 (CH), 35.56
(CH2), 47.71 (CH2), 67.89 (q), 70.48 (CH), 70.65 (CH), 71.89
(CH), 72.63 (CH), 74.30 (CH), 75.22 (CH), 75.82 (CH), 77.68
(CH), 85.17 (q), 85.39 (q), 91.68 (q), 101.58 (q), 128.28 (CH),
3.77 (1H, d, Jgem = 13.8 Hz), 4.10 (1H, m, Cp), 4.15 (1H, m, Cp),
4.20 (1H, m, Cp), 4.35 (1H, m, Cp), 4.50 (1H, m, Cp), 4.56 (1H,
m, Cp), 4.75 (1H, m, Cp), 5.10 (1H, m, Cp), 7.10–7.40 (4H, m),
7.40 (1H, d, J = 8.1 Hz); δC (CDCl3) 31.35 (CH2), 32.36 (CH),
42.62 (CH2), 64.84 (CH, Cp), 66.39 (CH, Cp), 67.33 (CH, Cp),
67.97 (CH, Cp), 69.38 (CH, Cp), 70.78 (CH, Cp), 71.37 (CH,
Cp), 71.73 (q), 72.88 (CH, Cp), 86.16 (qipso Fc), 98.08 (qipso Fc),
124.46 (2 × CH), 125.51 (CH), 126.31 (CH), 126.82 (CH),
127.83 (CH), 128.28 (2 × CH), 129.20 (q), 132.40 (q), 153.10
130.35 (2
× CH), 131.44 (CH), 132.02 (CH), 132.50
(CH), 136.96 (q), 141.69 (CH), 147.38 (q), 150.21 (q), 154.97
(CH), 172.52 (C᎐N); EIMS m/z 458 (Mϩ, 18%), 442 (13), 355
᎐
(q), 168.23 (C᎐N); EIMS m/z 433 (Mϩ, 100%), 415 (30), 248
(100), 312 (56), 248 (90), 191(42), 121 (12), 103 (51).
᎐
(56), 222 (11), 191 (21).
(3f ): (60%) mp >320 ЊC (from CH2Cl2/Et2O) (Found: C,
73.20; H, 4.65; N, 6.30. C28H22FeN2O requires C, 73.37; H, 4.84;
N, 6.11); νmax/cmϪ1: 3359, 2225, 1582, 1562, 1464, 1248, 1094,
1058, 970, 822 (Nujol); δH (CDCl3) 1.95 (1H, t, Jgem = 13.0 Hz,
3J = 13.0 Hz), 2.26 (1H, dd, Jgem = 14. Hz, 3J = 5 Hz), 2.38 (1H,
dd, Jgem = 13.8 Hz, 3J = 5.4 Hz), 3.28–3.52 (2H, m), 4.17 (1H, m,
Cp), 4.3 (2H, m, Cp), 4.48 (1H, m, Cp), 4.55 (1H, m, Cp), 4.66
(1H, m, Cp), 4.76 (1H, m, Cp), 4.87 (1H, m, Cp), 6.35 (1H, s,
(3b): (70%) mp 224–227 ЊC (from EtOAc) (Found: C, 68.59;
H, 4.70; N, 3.31. C25H21FeNOS requires C, 68.34; H, 4.82; N,
3.19); νmax/cmϪ1: 3380, 1585, 1561, 1450, 1123, 1040, 737
3
(Nujol; δH (CDCl3) 2.04 (1H, t, Jgem = 14.1 Hz, J = 14.1 Hz),
2.34 (1H, s, OH), 2.46–2.55 (2H, m), 3.4–3.49 (1H, m), 3.68
(1H, d, Jgem = 14.1 Hz), 4.03 (1H, m, Cp), 4.22 (1H, m, Cp), 4.24
(1H, m, Cp), 4.41 (1H, m, Cp), 4.5 (1H, m, Cp), 4.56 (1H, m,
D a l t o n T r a n s . , 2 0 0 4 , 1 1 5 9 – 1 1 6 5
1164

View Article Online
J. Soto, P. D. Beer, J. Cadman and D. K. Smith, J. Chem. Soc.,
Dalton Trans., 1999, 2359.
20 P. D. Beer, Z. Chen, M. G. B. Drew and A. J. Pilgrim, Inorg. Chim.
Acta, 1994, 225, 137.
21 P. D. Beer and P. V. Bernhardt, J. Chem. Soc., Dalton Trans., 2001,
1428.
22 J. M. Lloris, R. Martínez-Máñez, T. Pardo, J. Soto and
M. E. Padilla-Tosta, J. Chem. Soc., Dalton Trans., 1998, 1635.
23 P. D. Beer, J. Cadman, J. M. Lloris, R. Martínez-Máñez, J. Soto,
T. Pardo and D. Marcos, J. Chem. Soc., Dalton Trans., 2000, 1805.
24 C. Caltagirone, A. Bencini, F. Demartin, F. A. Devillanova,
A. Garau, F. Isaia, V. Lippolis, P. Mariani, U. Paple, L. Tei and
G. Verani, J. Chem. Soc., Dalton Trans., 2003, 901.
25 P. D. Beer and K. Y. Wild, Polyhedron, 1996, 15, 775.
26 A. Chesney, M. R. Bryce, A. S. Batsanov, J. A. K. Howard and
L. M. Goldenberg, J. Chem. Soc., Chem. Commun., 1998, 677.
27 O. B. Sutcliffe, A. Chesney and M. R. Bryce, J. Organomet. Chem.,
2001, 637–639, 134.
OH), 7.15–7.38 (5H, m), 7.7 (1H, d, J = 7.8 Hz), 8.43 (1H, m);
δC (CDCl3) 30.23 (CH), 30.42 (CH2), 42.71 (CH2), 62.84 (q),
65.29 (CH, Cp), 65.50 (CH, Cp), 66.70 (CH, Cp), 67.43 (CH,
Cp), 69.08 (CH, Cp), 70.01 (CH, Cp), 70.62 (CH, Cp), 72.51
(CH, Cp), 77.28 (qipso, Fc), 86.53 (qipso, Fc), 96.55 (q), 100.52
(q), 119.25 (q), 123.41 (CH), 125.15 (2 CH), 126.30 (CH),
127.31 (CH), 131.79 (q), 138.16 (CH), 145.04 (q), 148.60 (CH),
151.15 (CH), 167.34 (C᎐N); EIMS m/z 458 (Mϩ, 18%), 442 (13),
᎐
355 (100), 312 (56), 248 (90), 191(42), 121 (12), 103 (51).
Acknowledgements
We gratefully acknowledge the financial support of the DGI
(Ministerio de Ciencia y Tecnología, Spain) (Project BQU2001–
0014) and Fundación Séneca (CARM) (Project. No. PB/72/FS/
02). J.L.L. who is enrolled in the Ph.D programme of the UMU
(Universidad de Murcia), thanks the MCyT (Spain) for his
studentship.
28 O. B. Sutcliffe, M. R. Bryce and A. S. Batsanov, J. Organomet.
Chem., 2002, 656, 211.
29 V. Siemeling, B. Neumann, H-G. Stammler and A. Salmon,
Z. Anorg. Allg. Chem., 2002, 628, 2315.
30 A. Tárraga, P. Molina, D. Curiel and M. D. Velasco,
Organometallics, 2001, 20, 2145.
31 A. Tárraga, P. Molina, D. Curiel and M. D. Velasco, Tetrahedron,
2001, 57, 6765.
32 A. Tárraga, P. Molina, D. Curiel and M. D. Velasco, Tetrahedron
Lett., 2002, 43, 8453.
33 A. Tárraga, P. Molina and J. L. López, Tetrahedron Lett., 2000, 41,
2479.
34 A. Tárraga, P. Molina, J. L. López, M. D. Velasco, D. Bautista and
P. G. Jones, Organometallics, 2002, 21, 2055.
35 J. L. López, A. Tárraga, A. Espinosa, M. D. Velasco, P. Molina,
V. Lloveras, J. Vidal-Gancedo, C. Rovira, J. Veciana, D. J. Evans and
K. Wurts, Chem. Eur. J., 2004, 10, 1815.
36 D. P. Rillema, M. B. Callahan and K. B. Mack, Inorg. Chem., 1982,
2589.
37 D. Astruc, M. H. Desbois, M. Lacoste, N. Ardoin, L. Toupet and
F. Varrtet, in Mixed Valence Systems: Applications in Chemistry,
Physics, Biology, ed. K. Prassides, Kluwer, Amsterdam, 1991,
pp. 107–118.
References
1 G. W. Gokel, Chem Soc. Rev., 1992, 21, 39.
2 P. D. Beer, Adv. Inorg. Chem., 1992, 39, 79.
3 T. Jorgensen, T. K. Hansen and J. Becher, Chem. Soc. Rev., 1994, 23,
41.
4 L. Fabrizzi and A. Poggi, Chem. Soc. Rev., 1995, 197.
5 A. E. Kaifer, S. Mendoza, in Comprehensive Supramolecular
Chemistry, ed. G. W. Gokel, Pergamon, Oxford, 1996, vol. 1, p. 701.
6 P. L. Boulas, M. Gomez-Kaifer and L. Echegoyen, Angew. Chem.,
Int. Ed. Engl., 1998, 37, 216.
7 P. D. Beer and P. A. Gale, Adv. Phys. Org. Chem., 1998, 31, 1.
8 P. D. Beer, Acc. Chem. Res., 1998, 31, 71.
9 P. D. Beer, P. A. Gale and G. Z. Chen, Coord. Chem. Rev., 1999,
185–186, 3.
10 P. D. Beer and P. A. Gale, Angew. Chem., Int. Ed. Engl., 2001, 40,
485.
11 P. D. Beer, P. A. Gale and G. Z. Chen, J. Chem. Soc., Dalton Trans.,
1999, 1897.
12 M. P. Andrews, C. Blackburn, J. F. McAleer and V. D. Patel,
J. Chem. Soc., Chem. Commun., 1987, 1122.
38 L. O. Spreer, A. Li, D. B. MacQueen, C. B. Allan, J. Otvos,
M. Calvin, R. B. Frankel and G. C. Papaefthymiou, Inorg. Chem.,
1994, 33, 1753.
13 P. D. Beer, C. Blackburn, J. F. McAleer and V. D. Patel, Inorg. Chem.,
1990, 29, 378.
14 P. D. Beer, H. Sikanyika, C. Blackburn and J. F. McAleer, J. Chem.
Soc., Chem. Commun., 1989, 1831.
15 H. Plenio and R. Diodone, Inorg. Chem., 1995, 34, 3964.
16 J. C. Medina, T. T. Goodnow, M. T. Rojas, J. L. Atwood, B. C. Lynn,
A. E. Kaifer and G. W. Gokel, J. Am. Chem. Soc., 1992, 114,
10583.
17 M. C. Grossel, D. G. Hamilton, J. I. Fuller and E. Millan-Barios,
J. Chem. Soc., Dalton Trans., 1997, 3471.
39 P. D. Beer and D. K. Smith, J. Chem. Soc., Dalton Trans., 1998, 417.
40 S. R. Marder, J. W. Perry and B. G. Tiemann, Organometallics, 1991,
10, 1896.
41 B. J. Coe, C. J. Jones, J. A. McCleverty, D. Bloor and G. J. Cross,
J. Organomet. Chem., 1994, 464, 225C.
42 Lambert, W. Gaschler, M. Zabel, R. Matschiner and R. Wortmann,
J. Organomet. Chem., 1999, 592, 109.
43 T. J. Müller, A. Netz and M. Ansorge, Organometallics, 1999, 18,
5066.
44 J. D. Carr, S. J. Coles, M. B. Asan, M. B. Hurthouse, K. M. A. Malik
and J. H. R. Tucker, J. Chem. Soc., Dalton Trans., 1999, 57.
45 K. A. Connors, Binding Constants: The Measurements of Molecular
Complex Stability, John Wiley & Sons Ltd., New York, 1987, ch. 4.
18 M. C. Grossel, M. R. Goldspink, J. A. Hriljac and S. C. Weston,
Organometallics, 1991, 10, 851.
19 J. M. Lloris, R. Martínez-Máñez, M. E. Padilla-Tosta, T. Pardo,
D a l t o n T r a n s . , 2 0 0 4 , 1 1 5 9 – 1 1 6 5
1165