Anion interaction with ferrocene-functionalised cyclic and open-chain polyaza and aza-oxa cycloalkanes

Article abstract of DOI:10.1039/b000079p

A family of ferrocene-functionalised receptors of different topologies have been used as receptors for anions. The compounds have been designed to contain both amine nitrogen and ether oxygen atoms and comprises from monoaza to pentaaza derivatives both open-chain (L1, L2, L3) or cyclic (L4, L5) and having from one to five ferrocenyl groups. Solution studies directed to determine the protonation constants of L1, L2 and L3 have been carried out in water (0.1 mol dm3 KNO3, 25 °C) and those of L4 and L5 in 1,4-dioxane-water (70:30 v/v, 0.1 mol dm^-3 KNO₃, 25 °C). The protonation behaviour of the receptors can be explained taking into account electrostatic considerations. Speciation studies in the presence of phosphate have been carried out in water for L', L2 and L3 and in dioxane-water for L4 and L5. Speciation studies have also been performed in the presence of ATP with L1, L2 and L3 in water. Selectivity of a mixture of receptors against a certain anion is discussed in terms of ternary diagrams. The shift of the redox potential of the ferrocenyl groups as a function of the pH has been studied. The difference between the oxidation potentials at basic and acidic pH has been determined experimentally and is compared with that theoretically predicted using an electrostatic model previously reported. The electrochemical shift in the presence of ATP and phosphate has been measured in water for L1, L2 and L3 and in the presence of phosphate and sulfate in 1,4-dioxane-water for L4 and L5 as a function of the pH. The electrochemical response found against those anions is quite poor with maximum cathodic shifts off. 30tO mV. The electrochemical response induced by HSO4 and H2PO4- has also been studied in acetonitrile solutions where a large cathodic shift for H₂PO^4- up to ca. 200 mV was found. The Royal Society of Chemistry 2000.

Full text of DOI:10.1039/b000079p

View Online / Journal Homepage / Table of Contents for this issue
Anion interaction with ferrocene-functionalised cyclic and
open-chain polyaza and aza-oxa cycloalkanes
Paul D. Beer,*a James Cadman,a José Manuel Lloris,b Ramón Martínez-Máñez,*b Juan Soto,b
Teresa Pardob and Ma Dolores Marcosc
a Inorganic Chemistry Laboratory, University of Oxford, South Parks Road, Oxford,
UK OX1 3QR. E-mail: paul.beer@inorganic.chemistry.oxford.ac.uk
b Departamento de Química, Universidad Politécnica de Valencia, Camino de Vera s/n,
46071 Valencia, Spain. E-mail: rmaez@qim.upv.es
c Institut de Ciència dels Materials de la Universitat de València (ICMUV), PO Box 2085,
46071 Valencia, Spain
Received 6th January 2000, Accepted 24th March 2000
A family of ferrocene-functionalised receptors of different topologies have been used as receptors for anions.
The compounds have been designed to contain both amine nitrogen and ether oxygen atoms and comprises
from monoaza to pentaaza derivatives both open-chain (L1, L2, L3) or cyclic (L4, L5) and having from one to five
ferrocenyl groups. Solution studies directed to determine the protonation constants of L1, L2 and L3 have been
carried out in water (0.1 mol dmϪ3 KNO3, 25 ЊC) and those of L4 and L5 in 1,4-dioxane–water (70:30 v/v, 0.1 mol
dmϪ3 KNO3, 25 ЊC). The protonation behaviour of the receptors can be explained taking into account electrostatic
considerations. Speciation studies in the presence of phosphate have been carried out in water for L1, L2 and L3 and
in dioxane–water for L4 and L5. Speciation studies have also been performed in the presence of ATP with L1, L2 and
L3 in water. Selectivity of a mixture of receptors against a certain anion is discussed in terms of ternary diagrams.
The shift of the redox potential of the ferrocenyl groups as a function of the pH has been studied. The difference
between the oxidation potentials at basic and acidic pH has been determined experimentally and is compared with
that theoretically predicted using an electrostatic model previously reported. The electrochemical shift in the presence
of ATP and phosphate has been measured in water for L1, L2 and L3 and in the presence of phosphate and sulfate in
1,4-dioxane–water for L4 and L5 as a function of the pH. The electrochemical response found against those anions
is quite poor with maximum cathodic shifts of ca. 30–40 mV. The electrochemical response induced by HSO4Ϫ and
H2PO4Ϫ has also been studied in acetonitrile solutions where a large cathodic shift for H2PO4Ϫ up to ca. 200 mV was
found.
phosphate and sulfate. The receptors have been designed to
Introduction
contain both amine nitrogen and ether oxygen atoms and com-
Designing of molecules containing appropriate signalling
prise from monoaza to pentaaza derivatives both open-chain or
subunits has been a synthetic strategy for the development of
new sensing receptors of application in sensor technologies. In
these systems, co-ordination of the target guest to the binding
cyclic, having from one to five ferrocenyl groups.
sites followed by a further interaction of the complex with the
covalently attached signalling subunits allows transformation
of chemical information into a measurable signal. The possibil-
ity of modulation of the co-ordination sites, the nature and
number of signalling units (bearing in mind that external fac-
tors such as for example the solvent may also influence) makes
the number of potential receptors very large and the search for
selective receptors against target substrates a challenge. Among
different signalling subunits described in the literature there is
an increasing interest for redox-active groups.1–7 In these sys-
tems the signalling effect is based on the fact that co-ordination
can induce a shift of the redox potential of the electroactive
subunits. A number of works have allowed the synthesis of
suitable and selective receptors that electrochemically can
recognise cationic,8–16 anionic17–21 and neutral22,23 molecules.
New advances in this area would probably involve the use of
different not well studied redox-active units and the search for
potential applications as electrochemical sensors based on
amperometric measurements. With the aim of advance in the
knowledge of redox-active ferrocene-functionalised systems
and taking into account that relatively little effort has been
devoted to the study of their anion sensing ability in water,24–39
we report here a solution and electrochemical study on a family
of new receptors in the presence of some anions such as ATP,
Experimental
Physical measurements
Potentiometric titrations were carried out in water (0.1 mol
dmϪ3 KNO3) for L1, L2 and L3 and in 1,4-dioxane–water (70:30
v/v, 0.1 mol dmϪ3 KNO3) for L4 and L5 using a reaction vessel
water-thermostatted at 25.0 0.1 ЊC under a nitrogen atmos-
phere. Experimental potentiometric details have been published
previously.36 The concentration of the metal ions was deter-
mined using standard methods. The computer program
SUPERQUAD40 was used to calculate the protonation and
stability constants. The titration curves for each system (ca. 250
experimental points corresponding to at least three titration
curves, pH = Ϫlog [H] range investigated 2.5–10, concentration
of the ligand and metal ion being ca. 1.2 × 10Ϫ3 mol dmϪ3) were
treated either as a single set or as separated entities without
significant variations in the values of the stability constants.
Finally the data sets were merged and treated simultaneously to
give the stability constants. Electrochemical data were obtained
in water, 1,4-dioxane–water (70:30 v/v) and in dry acetonitrile,
with a programmable function generator Tacussel IMT-1,
connected to a Tacussel PJT 120–1 potentiostat. The working
electrode was platinum with a saturated calomel reference
electrode separated from the test solution by a salt bridge
DOI: 10.1039/b000079p
J. Chem. Soc., Dalton Trans., 2000, 1805–1812
This journal is © The Royal Society of Chemistry 2000
1805

View Online
amine) 1 (2.05 mmol), also in 30 mL of acetonitrile, and stirred
overnight under a nitrogen atmosphere and darkness. The
solution was evaporated and then methanol and small portions
of sodium tetrahydroborate (to a fivefold excess) were added
and the mixture was stirred for 3 hours. After removal of
methanol on a rotary evaporator, 40 mL of dichloromethane
were added and the off-white inorganic solid filtered off. The
dichloromethane was then evaporated and the organic oil
taken up in 50 mL of acetone and refluxed for 2.5 hours. After
cooling, any further solid was filtered off to give the crude
product as a viscous orange oil. Purification was achieved with
a silica gel column using methanol–aqueous ammonia (99:1
v/v) as eluent. The main yellow fraction was collected and
1
concentrated to an orange oil. Yield 0.51 g, 70%. H NMR
(CDCl3, TMS): δ 1.44 (s, 18 H, tBu), 2.82 (t, 4 H, CH2N), 3.27
(m, 4 H, CH2NHCO), 3.50 (s, 4 H, Cp CH2N), 3.58 (s, 16 H,
OCH2), 4.07 (d, 4 H, Cp H), 4.15 (d, 4 H, Cp H) and 5.21 (br, 2
H, NHCO).
1,1Ј-Bis(10-amino-5,8-dioxa-2-azadecyl)ferrocene L2. The
tert-butoxycarbonyl of compound 2 was removed by adding 5
mL of trifluoroacetic acid to 2 (0.49 g) dissolved in 10 mL of
dichloromethane at room temperature. After stirring for thirty
minutes, the solvent was removed and the salt subsequently
neutralised with 1 mol dmϪ3 KOH. The water was removed and
then the yellowish solid triturated with hot dichloromethane
to afford the free amine in 70% yield. The compound was
converted into the hydrochloride salt in quantitative yield by
bubbling HCl(g) through a dry diethyl ether solution of it.
Found: C, 42.99; H, 7.10; N, 7.83. C24H46Cl4FeN4O4ؒH2O
1
requires C, 43.00; H, 7.22; N, 8.36%). NMR (CD3OD): H,
δ 3.06 (t, 4 H, CH2N), 3.10 (t, 4 H, NCH2), 3.63 (s, 8 H, OCH2-
CH2O), 3.66 (m, 8 H, NCH2CH2O), 4.09 (s, 4 H, CpCH2N),
4.32 (d, 4 H, Cp H) and 4.48 (d, 4 H, Cp H); 13C, δ 40.60
(NCH2), 47.36 (NCH2), 48.17 (NCH2), 66.91 (CH2O), 67.82
(CH2O), 71.32 (OCH2CH2O), 71.35 (OCH2CH2O), 71.86 (Cp),
72.81 (Cp) and 78.34 (Cipso of Cp). Electrospray MS: m/z 507,
[M Ϫ 3HCl]ϩ; 529, [M Ϫ 3HCl ϩ Na]ϩ; and 543 [M Ϫ 4HCl]ϩ.
1,1Ј-Bis(5-morpholino-2-azapentyl)ferrocene L3. Ferrocene-
1,1Ј-dicarbaldehyde (0.24 g, 0.99 mmol) in 25 mL of dry
acetonitrile was added to 0.28 g of 4-morpholino-1-azabutane
(1.94 mmol) in 25 mL of dry acetonitrile with 4 Å molecular
sieves, and then stirred overnight under nitrogen in the dark.
The resulting dark orange solution was filtered and washed
with a small amount of chloroform. The solution was evapor-
ated to an orange oil which was then reduced in methanol with
a fivefold excess of sodium tetrahydroborate for three hours.
The work-up procedure was as for L2 using 95:5 v/v methanol–
ammonia as eluent. Yield 0.31 g, 64%. Found: C, 45.08; H, 7.73;
N, 7.86. C26H46Cl4FeN4O2ؒ3H2O requires C, 44.72; H, 7.50; N,
8.02%). NMR (CDCl3): 1H, δ 1.42 (bs, 2 H, NH), 1.64 (tt, 4 H,
CCH2C), 2.35 (t, 4 H, NCH2), 2.39 (bs, 8 H, NCH2(morph)),
2.63 (t, 4 H, NHCH2), 3.47 (s, 4 H, CpCH2N), 3.67 (bs, 8 H,
OCH2), 4.03 (d, 4 H, C pH) and 4.09 (d, 4 H, C pH); 13C, δ 26.75
(CCH2C), 47.99 (NCH2), 48.95 (NCH2), 53.75 (NCH2CH2O),
57.23 (CH2(morph)), 66.93 (CH2O), 68.25 (Cp), 68.82 (Cp),
86.96 (Cipso of Cp). FABMS: m/z 535, [M Ϫ 4HCl]ϩ; and 499,
[M Ϫ 3HCl]ϩ.
containing the solvent/supporting electrolyte. The auxiliary
electrode was platinum wire.
Syntheses
N-tert-Butoxycarbonyl-2,2Ј-ethylenedioxybis(ethylamine) 1.
The following procedure was similar to that described in refer-
ence 41. Di(tert-butoxycarbonyl) (2.45 g, 0.011 mol) was dis-
solved in 30 mL of 1,4-dioxane. This was added to 12.50 g of
2,2Ј-(ethylenedioxy)bis(ethylamine) (0.085 mol) also dissolved
in 30 mL of 1,4-dioxane, over 2.5 hours. The mixture was
stirred for 24 hours after which the solvent was removed on a
rotary evaporator giving a white semi-solid. Water (50 mL) was
added and the white precipitate of the bis-substituted by-
product subsequently filtered off. The aqueous filtrate was
extracted with 3 × 50 mL of dichloromethane. This organic
layer was backwashed with 2 × 10 mL of water to remove any
excess of diamine, then dried with MgSO4, filtered and reduced
in vacuum resulting in a clear viscous oil. Yield 1.97 g, 70%
(Found: C, 52.24; H, 10.23; N, 10.51. C11H24N2O4ؒ0.5H2O
requires C, 51.34; H, 9.79; N, 10.89%). NMR (CDCl3, TMS):
1,4,7,10,13-Penta(ferrocenylmethyl)-1,4,7,10,13-pentaaza-
cyclopentadecane, L4. 1,4,7,10,13-Pentaazacyclopentadecane
(0.4 g, 1.86 mmol) and 4.29 g (11.16 mmol) of (ferrocenyl-
methyl)trimethylammonium iodide42 were heated to reflux in
acetonitrile (300 mL) for 4 days in the presence of sodium car-
bonate (6 g). The warm reaction mixture was filtered and the
yellow solution evaporated to dryness. The resulting solid was
dissolved in dichloromethane and chromatographed using
dichloromethane–methanol (99:1) as eluent. Further recrystal-
lisation in dichloromethane–hexane gave L4 as an orange solid
(700 mg, 30%) (Found: C, 62.24; H, 6.36; N, 5.48. C65H75-
t
1H, δ 1.12 (br, 2 H, NH), 1.19 (s, 9 H, Bu), 2.63 (t, 2 H,
CH2NH2), 3.07 (m, 2 H, CH2NHCO), 3.28 (t, 2 H, OCH2), 3.31
(t, 2 H, OCH2), 3.38 (s, 4 H, OCH2CH2O) and 5.28 (br, 1 H,
NHCO). FABMS: m/z 249 [M ϩ H]ϩ.
N-(1,1Ј-Bis(methyl)ferrocene)-NЈ-tert-butoxycarbonyl-2,2Ј-
ethylenedioxybis(ethylamine) 2. Ferrocene-1,1Ј-dicarbaldehyde
(0.25 g, 1.03 mmol) in 30 mL of dry acetonitrile was added to
0.51 g of N-tert-butoxycarbonyl-2,2Ј-ethylenedioxybis(ethyl-
1806
J. Chem. Soc., Dalton Trans., 2000, 1805–1812

View Online
Table 1 Stepwise protonation constants for L1, L2, L3, L4 and L5. Water (25 ЊC, 0.1 mol dmϪ3 potassium perchlorate) was used for L1, L2 and L3,
dioxane–water (70:30 v/v, 25 ЊC, 0.1 mol dmϪ3 potassium nitrate) for L4 and L5
L1
9.57(1)a
17.81(2)
—
—
—
8.24
—
—
L2
L3
L4
L5
L ϩ Hϩ
[HL]ϩ
9.97(1)
10.18
19.61
26.11
31.93
—
9.43
6.50
5.82
9.38(2)
8.40(2)
L ϩ 2Hϩ
[H2L]2ϩ
[H3L]3ϩ
[H4L]4ϩ
[H5L]5ϩ
19.85(1)
28.79(2)
37.68(2)
—
9.88
8.94
16.43(2)
21.34(3)
24.42(3)
26.76(4)
7.05
4.91
3.08
15.09(3)
6.69
L ϩ 3Hϩ
L ϩ 4Hϩ
L ϩ 5Hϩ
[HL]ϩ ϩ Hϩ
[H2L]2ϩ ϩ Hϩ
[H3L]3ϩ ϩ Hϩ
[H4L]4ϩ ϩ Hϩ
[H2L]2ϩ
[H3L]3ϩ
[H4L]4ϩ
[H5L]5ϩ
8.89
—
—
2.34
a Values in parentheses are the standard deviations in the last significant digit.
Fe5N5ؒ3H2O requires C, 62.00; H, 6.43; N, 5.56%). NMR
1
(CDCl3): H, δ 4.10, 4.09 (two br resonances, 45 H, C5H5 and
C5H4), 3.36 (br, 10 H, CH2) and 2.43 (br, 20 H, CH2); 13C-{1H},
δ 83.40 (Cipso, C5H4), 70.20, 67.80 (C5H4), 68.41 (C5H5), 54.13
(CH2) and 51.12 (CH2). Mass spectrum (FAB): m/z 1205 (Mϩ).
Results and discussion
The synthesis of receptors L1 and L5 has been published else-
where (see reference 43). The synthesis of L2 has been carried
out by reaction of ferrocene-1,1Ј-dicarbaldehyde with the
diamine 2,2Ј-ethylenedioxybis(ethylamine) that was previously
blocked in one amine position with a butoxycarbonyl (BOC)
group followed by further reduction of the imine formed and
elimination of the BOC group with trifluoroacetic acid. L3 was
synthesized by reaction of the dicarbaldehyde with amino-
propylmorpholine to produce a bis-imine that was reduced
with NaBH4 by using standard procedures. The synthesis of
L4 was carried out by reaction of 1,4,7,10,13-pentaazacyclo-
pentadecane with an excess of (ferrocenylmethyl)trimethyl-
ammonium iodide in acetonitrile in the presence of sodium
carbonate. 1H, 13C NMR, mass spectra and elemental
analysis of L2, L3 and L4 were consistent with the proposed
formulations.
Fig. 1 Distribution diagram of the L4–Hϩ system.
second protonation occur near an ammonium group. In L2
and L3 the free rotation of the cyclopentadienyl rings minimises
the charge repulsion effect between ammonium groups. Addi-
tionally the small difference between the first two protonation
constants suggests that the second proton attacks the arm in L2
or L3 that is unprotonated. The difference between the third and
fourth protonations of L2 and L3 is in accordance with the fact
that larger spacers between nitrogens produce larger basicity.
The difference between the second and third protonation con-
stants for L2 is 0.94 logarithm units, whereas the difference
between the logarithm of the second and third protonation
steps in L3 is 2.93. The third protonation in L3 would take place
at a nitrogen separated by one propane chain from one already
protonated nitrogen, whereas in L2 the separation between the
nitrogen to be protonated and one ammonium group is larger
due to the presence of the (CH2)2O(CH2)2O(CH2)2 spacer. A
similar effect is found when comparing the difference between
the third and the fourth protonation basicity constants of L2
and L3.
L1 to L5 are a family of redox-functionalised polyaza or
aza-oxa receptors showing a variety of topologies from cyclic
to open-chain and having from one to five ferrocenylmethyl
groups. It is well known that polyammonium receptors can
interact with anions via electrostatic effects or by formation of
hydrogen bonding networks and therefore L1 to L5 are candi-
dates to form complexes with anionic guests. Additionally, the
presence of redox-active groups near the co-ordination sites
could allow receptors L1 to L5 electrochemically to sense the
presence of anions. Potentiometric and electrochemical studies
have been carried out in order to characterise the solution
behaviour against selected guests.
Protonation studies
Solution studies directed to the determination of protonation
constants and stability constants for the formation of com-
plexes of L1, L2, L3, L4 and L5 with anions have been carried
out. Data for L1, L2 and L3 have been obtained in water (25 ЊC,
0.1 mol dmϪ3 KNO3), whereas data for L4 and L5 have been
determined in 1,4-dioxane–water (70:30 v/v, 25 ЊC, 0.1 mol
dmϪ3 KNO3) due to their low solubility in water or other
solvent mixtures. Protonation constants of receptors L1 to L5
are shown in Table 1.
The protonation behaviour of the receptors is as expected
and can satisfactorily be explained by taking into account
electrostatic considerations. Among the three receptors meas-
ured in water, L2 shows the more basic behaviour. The first
protonation constant is similar for all three receptors, whereas
the second protonation is less basic for L1 consistent with the
cyclic nature of the ferrocenophane moiety that makes the
Despite the use of a different solvent, in comparison with L2
and L3, its cyclic nature makes L4 more basic than the open-
chain polyamines. For instance the difference between the
logarithms of the first and fourth protonation constant of L2
and L3 is 1.08 and 4.36, whereas the same difference for L4
is 6.30. Fig. 1 shows the distribution diagram of the L4–Hϩ
system. More detailed protonation studies on polyazaalkanes
1
are usually developed by monitoring H or 13C NMR spectra
as a function of the pH.
From a different viewpoint stepwise protonation constants
can be calculated using eqn. (1).44,45 This equation works quite
log βi =
e2NA
1
1
i
k Ϫ 1
͚
i
k Ϫ 1
͚
i log K1 Ϫ
ͫ
͚
ϩ B ͚
ͬ
(1)
2
k = 1 l = 1
2.3RTπε0ε
rkl
rkl
k = 1 l = 1
J. Chem. Soc., Dalton Trans., 2000, 1805–1812
1807

View Online
well for open-chain polyamines and allows one to predict the
overall protonation constants providing that the first proton-
ation is known (K1). ε represents the macroscopic relative
permittivity of the medium and ε0 the permittivity constant
in vacuo. The distances rkl between ammonium groups can be
obtained by using adequate modelling programs (B = 30.6 ÅϪ2
in water).46 The equation assumes that the decrease of the suc-
cessive protonation constants is mainly due to electrostatic
repulsion between ammonium groups. Following this approach
the logarithms of the calculated and experimental overall pro-
tonation constants for receptors L1 to L3 are shown in Table 2.
The protonation path assumed is that minimising electrostatic
repulsion in each protonation step. There is in general a good
agreement between the experimental and calculated proton-
ation constants except for the fourth protonation of L3.
thermodynamic stability constants found from potentiometric
titration experiments in the presence of phosphate for receptors
L1, L2, L3, L4 and L5. For L1 three different complexes have been
2Ϫ
found that may involve the interaction of [HL1]ϩ ϩ HPO4
,
2Ϫ
Ϫ
[H2L1]2ϩ ϩ HPO4 and [H2L1]2ϩ ϩ H2PO4 . Assuming these
interactions between species, the logarithms of the stability
constants lie between 2.76 and 3.35 (see Table 3). A similar
interpretation appears to be operative for the L5–Hϩ–phosphate
system, although the logarithms of the formation stability
constants are larger for L5 than for L1.
Logarithms of the protonation constants of L2 are in the
range 9.97 to 8.89. Bearing in mind that the logarithm of the
first protonation of phosphate is 12.03 and that the second
protonation does not take place until pH 7.03, the existing
species in solution are tentatively assigned to the interaction
2Ϫ
of HPO4 with the species [H2L2]2ϩ, [H3L2]3ϩ and [H4L2]4ϩ.
Anion binding studies
Assuming these interactions, the logarithms of the stability
2Ϫ
constants for the equilibria [H2L2]2ϩ ϩ HPO4
[H3L2-
Speciation studies have been carried out in water (0.1 mol dmϪ3,
KNO3, 25 ЊC) for L1, L2 and L3 and in 1,4-dioxane–water
(70:30 v/v, 0.1 mol dmϪ3 KNO3, 25 ЊC) for L1 and L5. Tables 3
and 4 report the stability constants found for the L–Hϩ–A sys-
tems (see Table 3 for phosphate, Table 4 for ATP). Ferrocene-
functionalised polyamines have proved to be good receptors
for the electrochemical recognition of target anions4 since (i)
polyamines give in aqueous solution highly charged protonated
species able to interact with anions via electrostatic forces and
by formation of hydrogen bonding interactions and (ii) ferro-
cenium cations formed upon oxidation during electrochemical
experiments are also capable of producing favourable electro-
static interactions with anions (see below).20 Table 3 shows the
2Ϫ
PO4], [H3L2]3ϩ ϩ HPO4
HPO4
[H4L2PO4]ϩ and [H4L2]4ϩ
ϩ
2Ϫ
[H5L2PO4]2ϩ are 2.43, 2.73 and 2.56, respectively,
values that are in the range of those found for the interaction of
L1 with phosphate. Additionally the complex [H6L2PO4]3ϩ has
also been found to exist and probably involves [H4L2]4ϩ and
Ϫ
H2PO4 . The last protonation constant of L3 (log K = 5.82 for
[H3L3]3ϩ ϩ Hϩ
[H4L3]4ϩ) is even smaller than the second
protonation of phosphate (log K = 7.03 for HPO42Ϫ ϩ
Table 4 Logarithms of the stability constants for the interaction of L1,
L2 and L3 with ATP in water (25 ЊC, 0.1 mol dmϪ3 KNO3)
L1
L2
L3
Table 2 Calculated and experimental overall protonation constants
L ϩ 2Hϩ ϩ ATP4Ϫ
L ϩ 3Hϩ ϩ ATP4Ϫ
L ϩ 4Hϩ ϩ ATP4Ϫ
L ϩ 5Hϩ ϩ ATP4Ϫ
L ϩ 6Hϩ ϩ ATP4Ϫ
[H2L]2ϩϩ ATP4Ϫ
[H3L]3ϩϩ ATP4Ϫ
[H4L]4ϩϩ ATP4Ϫ
[H2L]2ϩϩ HATP3Ϫ
[H2L]2ϩϩ H2ATP2Ϫ
[H3L]3ϩϩ HATP3Ϫ
[H3L]3ϩϩH2ATP2Ϫ
[H4L]4ϩϩ HATP3Ϫ
[H4L]4ϩϩH2ATP2Ϫ
[H2LATP]2Ϫ 20.81(2)a 23.11(5) 22.63(2)
for the L1, L2 and L3 receptors
[H3LATP]Ϫ 27.91(2)
32.22(5) 29.98(2)
40.72(7) 36.32(2)
47.26(9) 42.19(3)
[H4LATP]
[H5LATP]ϩ
[H6LATP]2ϩ
[H2LATP]2Ϫ
[H3LATP]Ϫ
[H4LATP]
32.47(2)
—
—
3.00
—
—
Receptor
log βi(calc.)a
log βi(exp.)b
—
46.28(4)
3.02
3.87
4.39
3.59
5.92
3.43
5.29
3.48
3.56
L1
L2
—
17.92
—
19.68
28.88
37.14
—
19.48
26.72
34.27
9.57
17.81
9.97
19.85
28.79
37.68
10.18
19.61
26.11
31.93
3.26
3.43
3.04
5.59
10.08
5.15
7.68
2.80
—
[H3LATP]Ϫ
[H4LATP]
3.32
3.87
—
—
—
[H4LATP]
L3
[H5LATP]ϩ
[H5LATP]ϩ
[H6LATP]2ϩ
—
a Values in parentheses are the standard deviations in the last significant
digit.
a From eqn. (1). b From potentiometry.
Table 3 Logarithms of the stability constants for the interaction of L1, L2, L3, L4 and L5 with phosphate in water (25 ЊC, 0.1 mol dmϪ3 KNO3) for
L1, L2, L3 and in 1,4-dioxane–water (70:30 v/v, 25 ЊC, 0.1 mol dmϪ3 KNO3) for L4, L5
L1
L2
L3
L4
L5
3Ϫ
L ϩ 2Hϩ ϩ PO43Ϫ
[H2LPO4]Ϫ
[H3LPO4]
24.36(2)a
33.02(2)
40.22(3)
—
—
6.55
—
—
—
2.76
3.18
3.35
—
—
—
—
—
40.31(4)
46.48(5)
51.23(5)
—
23.99(2)
32.01(2)
38.91(1)
—
—
8.9
—
—
—
3.85
5.18
3.52
—
—
—
L ϩ 3Hϩ ϩ PO43Ϫ
L ϩ 4Hϩ ϩ PO43Ϫ
L ϩ 5Hϩ ϩ PO43Ϫ
L ϩ 6Hϩ ϩ PO4
34.31(9)
43.55(7)
52.27(9)
60.21(7)
—
5.52
5.87
—
34.51(4)
42.38(4)
48.93(4)
55.70(4)
—
8.4
10.45
—
[H4LPO4]ϩ
[H5LPO4]2ϩ
[H6LPO4]3ϩ
[H2LPO4]Ϫ
[H3LPO4]
3Ϫ
[H2L]2ϩ ϩ PO43Ϫ
[H3L]3ϩ ϩ PO43Ϫ
[H4L]4ϩ ϩ PO43Ϫ
[H5L]5ϩ ϩ PO4
—
[H4LPO4]ϩ
[H5LPO4]2ϩ
[H2LPO4]Ϫ
[H3LPO4]
15.89
19.72
—
2Ϫ
[HL]ϩ ϩ HPO4
—
—
2Ϫ
[H2L]2ϩ ϩ HPO4
2.43
4.64
2.73
4.42
2.56
3.47
—
2.87
3.71
4.24
3.76
4.97
4.71
—
—
Ϫ
[H2L]2ϩ ϩ H2PO4
[H4LPO4]ϩ
[H4LPO4]ϩ
[H5LPO4]2ϩ
[H5LPO4]2ϩ
[H6LPO4]3ϩ
[H6LPO4]3ϩ
3.84
7.23
5.10
10.32
6.77
12.73
2Ϫ
[H3L]3ϩ ϩ HPO4
Ϫ
[H3L]3ϩ ϩ H2PO4
—
—
—
—
2Ϫ
[H4L]4ϩ ϩ HPO4
Ϫ
[H4L]4ϩ ϩ H2PO4
—
—
2Ϫ
[H5L]5ϩ ϩ HPO4
a Values in parentheses are the standard deviations in the last significant digit.
1808
J. Chem. Soc., Dalton Trans., 2000, 1805–1812

View Online
Fig. 2 Distribution diagram of the L3–Hϩ–ATP system.
Ϫ
Hϩ
H2PO4 ). That makes the determination of the nature
of the phosphate complexes rather difficult taking into account
only stability constant values.
For the L4 receptor, even the third protonation of phosphate
is larger than the last protonation of the receptor making it
again difficult to know the nature of the complexes. Neverthe-
less the stability constants found in 1,4-dioxane–water for the
L4–Hϩ–phosphate system seem to be higher than those found
for L2 and L3 in water. In the case of L3-phosphate the inter-
3Ϫ
action between PO4 and the protonated species, log K = 8.40
3Ϫ
for [H3L3]3ϩ ϩ PO4
[H3L3PO4] and log K = 10.45 for
3Ϫ
[H4L3]4ϩ ϩ PO4
[H4L3PO4]ϩ. In the case of the L4–
phosphate system the stability constants for the analogous equi-
libria are log K = 15.89 and 19.72. The different affinity can be
explained by taking into account the reduction of the permit-
tivity in dioxane–water when compared with water. The different
topology of the ligands (cyclic versus acyclic) and the number
of nitrogen atoms also can influence the observed behaviour.
In Table 4 the stability constants of the acyclic L1, L2 and L3
receptors with ATP are reported. Fig. 2 plots the distribution
diagram for the L3–Hϩ–ATP system. Receptor L1 and receptor
L2 are fully protonated at pH lower than ca. 8, whereas the first
protonation of the ATP in water is at ca. 6.7 and therefore the
complexes expected in solution would involve the interaction of
ATP4Ϫ with the protonated species HjLjϩ. Additionally proton-
ated species of the type [H5L2ATP]ϩ have also been found and
probably are related with the interaction of [H4L2]4ϩ with
HATP3Ϫ. Spermine [H2N(CH2)3NH(CH2)4NH(CH2)3NH2] is
an open-chain tetraamine whose interaction with ATP has also
been reported in water.47 In this case the logarithm of the stabil-
ity constant between the tetraprotonated spermine and the
ATP4Ϫ was found to be 3.97 which is a value close to that found
for the interaction of [H4L2]4ϩ and ATP4Ϫ (log K = 3.04 for
Fig. 3 Formation stability constants (log K) as a function of the
[H4L2]4ϩ ϩ ATP4Ϫ
[H4L2ATP]).
number of protons (n) for the interaction of (a) L1, L2, L3, L4 and L5
with phosphate and (b) L1, L 2 and L3 with ATP.
In order to summarise the interaction of the ligands L1, L2,
L3, L4 and L5 with phosphate and L1, L2 and L3 with ATP, we
have plotted in Fig. 3(a) and 3(b) the formation stability con-
stants of the different HjLPO4
constants with phosphate are found for the receptor L4 which
is the one containing the largest number of nitrogen atoms.
Receptors L1 and L5 (both containing two amino groups) dis-
play a similar co-ordination behaviour, the stability constants
of L5, obtained in dioxane–water, being slightly larger than
those of L1 in water. Of interest is the comparison between the
stability constants of L2 and L3 with phosphate. Both receptors
contain four amino groups and, despite the larger basicity of L2
and the presence of oxygen atoms that can give hydrogen bond-
ing interactions with protonated phosphate species, the higher
formation stability constants are found for L3. A similar
behaviour is found in the interaction of L2 and L3 with ATP;
the larger stability constants are obtained with L3. Molecular
modelling of the interaction of L2 and L3 with phosphate and
ATP showed that the receptor L3 is able to embrace the sub-
j Ϫ 3
and HjLATPj Ϫ 4 species as a
function of the number of protons in the complex. To calculate
these constants we have proceeded as follows. From the distri-
bution diagrams of the L–Hϩ–phosphate or L–Hϩ–ATP systems
is determined the pH at which a certain complex HjLPO4j Ϫ 3 or
HjLATPj Ϫ 4 is formed. Then, it is assumed that the interaction
to give the corresponding complex is established between the
L–Hϩ and the anion–Hϩ species that existed in the highest
percentage at this pH. From these plots some information
regarding host design and selectivity could be acquired. The
formation stability constants against phosphate are higher for
L4 and L5 than for L1, L2 and L3. This is probably due to the
lower relative permittivity of dioxane–water mixtures when
compared with that of pure water. Larger formation stability
J. Chem. Soc., Dalton Trans., 2000, 1805–1812
1809

View Online
Table 5 Experimental and calculated electrochemical shift from acidic
to basic pH for receptors L1, L2, L3, L4 and L5
a
b
Ligand
(∆E1/2
)
(∆E1/2)
exp.
calc.
L1
L2
L3
L4
L5
275
330
380
85
290
310
360
80
120
140
a From eqn. (2). b From electrochemical data.
be attributed to the higher basicity of L2 when compared with
L3, whereas when both receptors are protonated the stronger
interaction of L3 might be due to the closer proximity between
polyammonium groups. Ternary diagrams for the L2–Hϩ–ATP,
phosphate system allow one to determine pH ranges of select-
ive complexation of L2 in an equimolar mixture of ATP and
phosphate. They show the preferential co-ordination of L2
towards ATP at basic pH and a selective complexation of phos-
phate at acidic pH. Similar results are obtained for the ternary
L3–Hϩ–ATP, phosphate system.
Fig. 4 Distribution diagram of the ternary system L2, L3–Hϩ-
phosphate. The sum of percentages of complexed species are plotted
versus pH. ᮀ, Phosphate; ᭛, L2–phosphate; ᭝, L3–phosphate.
In conclusion, larger formation stability constants with
phosphate or ATP are obtained with receptors containing more
amino groups. By comparison between receptors having the
same number of basic centres (for instance L2 and L3), larger
stability constants are found for the receptor having closer
amino groups. An enhancement of the formation stability
constants is found in dioxane–water mixtures when compared
with those obtained in water.
Electrochemical behaviour
The shift of the redox potential of the ferrocenyl groups as a
function of the pH in the presence of phosphate and ATP has
been monitored in water for L1, L2 and L3 and in the presence
of phosphate and sulfate in 1,4-dioxane–water (70:30 v/v) for
L4 and L5. One of the most promising applications of redox-
functionalised molecules is the incorporation in amperometric
sensing devices of certain receptors showing a large and selec-
tive electrochemical shift against target substrates.
As already observed in related systems there is a steady
anodic shift of the redox potential of the ferrocenyl groups
from basic to acidic pH. The difference between the oxidation
potential at basic and acidic pH has been measured for all the
receptors and is shown in Table 5. From an electrochemical
viewpoint polyammonium species are more difficult to be
oxidised than the unprotonated receptor because there is extra
electrical work related with the approach to a positively
charged electrode of positively charged species. It has recently
been reported50 that in systems where there is no electronic
communication between protonated groups and the ferrocene
moiety the electrochemical shift due to the protonation of
amines can theoretically be calculated by taking into account
the free energy associated with the electrostatic repulsion
between the ferrocenium cation and the ammonium groups.
Following this approach eqn. (2) can be deduced with redox
Fig. 5 Distribution diagram of the ternary system L2, L3–Hϩ–ATP.
The sum of percentages of complexed species are plotted versus pH.
ᮀ, ATP; ᭛, L2–ATP; ᭝, L3–ATP.
strates giving interaction with the four basic centres. In contrast
L2 is only able to interact with the two nearest ammonium
groups to the ferrocenyl unit, the others being too far from the
anion to give a noticeable contact. This is in line with the fact
that the largest stability constant found for the interaction of L2
and phosphate is close to the largest stability constant found for
the interaction of L1 or L5 (containing two ammonium groups)
with phosphate (see Fig. 3). Therefore, the presence of closer
amino groups in L3 separated by a propyl instead of a
(CH2)2O(CH2)2O(CH2)2 spacer as in L2 produces a stronger
interaction between the anion and the receptor. In order to
determine the selectivity in a mixture of L2 and L3 in the pres-
ence of phosphate, we have plotted the ternary diagram for L2,
L3–Hϩ–phosphate (Fig. 4) by means of a calculation of the
overall percentages of each receptor bound to the anion as a
function of the pH.48,49 It reveals that there is a selective com-
plexation of phosphate with receptor L3 especially at acidic pH.
To study the selectivity in a mixture of the receptors L2 and L3
against ATP we have also plotted diagrams of the ternary sys-
tems L2, L3–Hϩ–ATP containing ATP and each receptor (Fig.
5). This reveals the selective complexation of ATP with L3 at
acidic pH and a selective complexation of ATP by L2 at basic
pH. Preferential ATP interaction by receptor L2 at basic pH can
1
14398
1
2 × 106
1
∆E =
ͫ
͚ ͚
ϩ
ͩ
ϩ 1051
ͪ
͚ ͚
ͬ
(2)
2
j
i
j
i
jn
ε
rji
ε2
rji
groups ( j) and number of electrons (n). This equation allows
prediction of the maximum electrochemical shift in ferrocene-
functionalised receptors due to protonation processes. The only
parameter to be determined is the distance between the
ferrocene (iron atom) and all the ammonium groups, and the
relative permittivity of the medium (ε can approximately be
calculated for a mixture of solvents as Σxiεi, where xi = molar
fraction of the solvent i and εi = permittivity of solvent i). In
case a crystal structure is not available, determination of the rji
1810
J. Chem. Soc., Dalton Trans., 2000, 1805–1812

View Online
Table 7 Voltammetric data for the free L4, L5 receptors and for L4, L5
Table 6 Cathodic electrochemical shifts of L1, L2 and L3 in the
presence of ATP and phosphate in water (25 ЊC, 0.1 mol dmϪ3 KNO3)
and cathodic electrochemical shifts of L4 and L5 in the presence of
phosphate and sulfate in dioxane–water (70:30 v/v)
Ϫ
Ϫ
in the presence of HSO4 and H2PO4 in dry acetonitrile, in 0.1 mol
dmϪ3 tetrabutylammonium perchlorate
System
Epa/V
Epc/V
E1/2/V
Ligand
Phosphate (pH)
ATP (pH)
Sulfate(pH)
L4
500
500
300
480
480
395
400
170
Ϫ200
370
130
0
450
Ϫ
L1
L2
L3
L4
L5
30a (7–8)
34 (7–8)
25 (8–9)
15 (6–7)
10 (6–7)
40 (7–8)
30 (7–8)
30 (7–8)
L4 ϩ 10 equivalents HSO4
L4 ϩ 10 equivalents H2PO4
Ϫ
L5
425
Ϫ
15 (6–7)
<5
L5 ϩ 10 equivalents HSO4
L5 ϩ 10 equivalents H2PO4
Ϫ
a In mV.
Epa and Epc are the oxidation and reduction wave potential; E1/2
(Epa ϩ Epc)/2; ∆Eac = (Epa Ϫ Epc).
=
Ϫ
Ϫ
induced by HSO4 and H2PO4 in the receptors L4 and L5 in
acetonitrile solutions (due to its insolubility ATP could not be
studied). Fig. 6 shows the changes in the voltammetric wave of
Ϫ
receptor L4 after progressive addition of H2PO4 . L4 shows a
typical reversible oxidation wave at ca. 500 mV. Addition of 4
equivalents of H2PO4Ϫ produces the appearance of a new oxid-
ation peak at ca. 300 mV. Successive addition of this anion
results in the extinction of the former oxidation wave of the
free receptor and the appearance of a cathodically shifted irre-
versible wave (electrochemical–chemical (EC) process). The
observed behaviour can be explained taking into account that
both the free receptor and its oxidised form can bind the
H2PO4Ϫ anion. The cathodic shift of the anodic peak (Epa) from
500 to 300 mV is due to the formation of the complex [L4H2-
PO4]Ϫ. This complex has a negative charge and its anionic
nature makes it easier to oxidise when compared to the free
neutral receptor. The new cathodic peak (Epc) at Ϫ200 mV after
Fig. 6 Cyclic voltammogram (100 mV sϪ1) of L4 in the presence of
successive amounts of H2PO4Ϫ in acetonitrile, 0.1 mol dmϪ3 tetrabutyl-
ammonium perchlorate.
Ϫ
addition of H2PO4 (600 mV shifted from the former Epc peak
of the free receptor at 400 mV) is assigned to the extra strong
interaction between the oxidised ferrocenium groups and the
phosphate anion. A similar behaviour is observed when increas-
ing quantities of H2PO4Ϫ are added to acetonitrile solutions of
L5 (see Table 7).
distances by a modelling program is usually sufficient. The
results compare well with those experimentally obtained (see
Table 5).
Electrochemical studies of L1, L2 and L3 as a function of the
pH in water in the presence of phosphate and ATP showed
maximum electrochemical shifts of ca. 30–40 mV at neutral
pH (see Table 6). We have also recently reported51 poor
electrochemical shifts of ca. 30–40 mV at neutral pH against
phosphate using related ferrocene-functionalised tetraamines.
However, whereas in the present work maximum cathodic shifts
in the presence of ATP were found to be less than 30 mV,
for other ferrocene-functionalised tetraamines ATP has been
reported cathodically to shift the oxidation potential of the
ferrocenyl groups up to ca. 100 mV.51
We have also studied the changes in the oxidation wave
Ϫ
profile of L4 in the presence of the HSO4 anion. Increasing
Ϫ
amounts of HSO4 (up to 12 equivalents) do not produce any
significant shift in the anodic peak of the ferrocenyl groups
although, as for phosphate, an EC profile was observed after
addition of four equivalents. The Epc peak in the presence of
sulfate appears at 170 mV (230 mV shifted from that of the free
receptor) whereas that for phosphate was found near Ϫ200 mV.
The results suggest that there is a selective sensing response for
phosphate over sulfate. This contrasts with what was observed
for the analogous electroactive-functionalised receptor 1,15-
diferrocenyl-2,5,8,11,14-pentaazapentadecane (a pentaaza
open-chain polyamine functionalised with two ferrocenyl
groups) for which both phosphate and sulfate produced similar
changes in the voltammogram of the ferrocenyl groups.52 L5
shows a similar behaviour in the presence of sulfate to that
found for L4 (see Table 7).
The electrochemical behaviour of L4 and L5 as a function of
the pH in 1,4-dioxane–water (70:30 v/v) mixtures has been
studied in the presence of phosphate and sulfate (ATP is not
soluble in this mixture). Results are shown in Table 6. For these
receptors neither the presence of sulfate or phosphate modifies
significantly the oxidation potential of the ferrocenyl groups
(the largest electrochemical shift is less than ca. 20 mV). Despite
the cyclic nature of L4 the electrochemical response against
anions is very poor. Such a poor response has also been
observed for the N-ferrocene-functionalised derivative of the
polyazacycloalkane 1,4,7,10,13,16-hexaazacyclooctadecane.45
From a practical point of view ferrocene-functionalised
receptors might be used in solution (probably in water) or
can be incorporated into suitable amperometric devices. In the
former case the use of molecules showing a large electro-
chemical response in an aqueous environment would be
required. In contrast, the response of redox-active groups
incorporated into a solid electrode would be quite different to
that observed in water due to the lower relative permittivity of
the molecule in this medium. In order to study how the change
of the relative permittivity affects the electrochemical response
against anions, we have also studied the electrochemical shift
Acknowledgements
We should like to thank the Dirección General de Investigación
Científica y Técnica (proyecto PB98-1430-C02-02 and 1FD97-
0508-C03-01 and AMB99-0504.C02-01) for support. We would
also like to thank EPSRC for a studentship to J. C. and for use
of the mass spectrometry service at the University of Wales,
Swansea.
References
1 P. L. Boulas, M. Gómez-Kaifer and L. Echegoyen, Angew. Chem.,
Int. Ed., 1998, 37, 216.
2 T. Jorgensen, T. K. Hansen and J. Becher, Chem. Soc. Rev., 1994, 41.
J. Chem. Soc., Dalton Trans., 2000, 1805–1812
1811

View Online
31 C. Dusemund, K. R. A. S. Sandanayake and S. Shinkai, J. Chem.
3 P. D. Beer, P. A. Gale and Z. Chen, J. Chem. Soc., Dalton Trans.,
1999, 1897.
Soc., Chem. Commun., 1995, 333.
4 P. D. Beer, Acc. Chem. Res., 1998, 31, 71.
5 P. D. Beer and D. K. Smith, Prog. Inorg. Chem., 1997, 46, 1.
6 P. D. Beer, Chem. Soc. Rev., 1989, 409.
7 R. Martínez-Máñez, J. Soto, J. M. Lloris and T. Pardo, Trends Inorg.
Chem., 1998, 5, 183.
8 A. Chesney, M. R. Bryce, A. S. Batsanov, J. A. K. Howard and
L. M. Goldenberg, Chem. Commun., 1998, 677.
9 Z. Chen, A. J. Pilgrim and P. B. Beer, J. Electroanal. Chem.
Interfacial Electrochem., 1998, 444, 209.
10 H. Plenio and C. Aberle, Organometallics, 1997, 16, 5950.
11 P. D. Beer and K. Y. Wild, Polyhedron, 1996, 15, 775.
12 G. C. Dol, P. C. J. Kamer, F. Hartl and P. W. N. M. Van Leeuwen,
J. Chem. Soc., Dalton Trans., 1998, 2083.
32 P. D. Beer, Z. Chen, M. G. B. Drew, J. Kingston, M. Ogden and
P. Spencer, J. Chem. Soc., Chem. Commun., 1993, 1046.
33 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.
34 M. J. L. Tendero, A. Benito, J. Cano, J. M. Lloris, R. Martínez-
Máñez, J. Soto, A. J. Edwards, P. R. Raithby and M. A. Rennie,
J. Chem. Soc., Chem. Commun., 1995, 1634.
35 M. J. L. Tendero, A. Benito, R. Martínez-Máñez, J. Soto, J. Payà,
A. J. Edwards, P. R. Raithby and M. A. Rennie, J. Chem. Soc.,
Dalton Trans., 1996, 343.
36 M. J. L. Tendero, A. Benito, R. Martínez-Máñez, J. Soto, E. García-
España, J. A. Ramírez, M. I. Burguete and S. V. Luis, J. Chem. Soc.,
Dalton Trans., 1996, 2923.
13 J. D. Carr, S. J. Coles, W. W. Hassan, M. B. Hursthouse, K. M. A.
Malik and J. H. R. Tucker, J. Chem. Soc., Dalton Trans., 1999, 57.
14 H. Plenio and C. Aberle, Chem. Commun., 1998, 2697.
15 P. D. Beer and D. K. Smith, J. Chem. Soc., Dalton Trans., 1998,
417.
16 M. C. Grossel, D. G. Hamilton, J. I. Fuller and E. Millan-Barios,
J. Chem. Soc., Dalton Trans., 1997, 3471.
17 P. A. Gale, Z. Chen, M. G. B. Drew, J. A. Heath and P. D. Beer,
Polyhedron, 1998, 405.
18 M. Scherer, J. L. Sessler, A. Gebaver and V. Lynch, Chem.
Commun., 1998, 85.
37 M. E. Padilla-Tosta, R. Martínez-Máñez, T. Pardo, J. Soto and
M. J. L. Tendero, Chem. Commun., 1997, 887.
38 J. M. Lloris, R. Martínez-Máñez, T. Pardo, J. Soto and M. E.
Padilla-Tosta, Chem. Commun., 1998, 837.
39 J. M. Lloris, R. Martínez-Máñez, T. Pardo, J. Soto and M. E.
Padilla-Tosta, J. Chem. Soc., Dalton Trans., 1998, 2635.
40 P. Gans, A. Sabatini and A. Vacca, J. Chem. Soc., Dalton Trans.,
1985, 1195.
41 A. P Krapcho and C. S. Kuell, Synth. Commun., 1990, 20, 2559.
42 D. Lednicer and C. R. Hauser, Org. Synth., 1978, 40, 31.
43 J. M. Lloris, R. Martínez-Máñez, M. E. Padilla-Tosta, T. Pardo,
J. Soto, P. D. Beer, J. Cadman and D. K. Smith, J. Chem. Soc.,
Dalton Trans., 1999, 2359.
19 P. D. Beer, M. G. B. Drew and R. Jagessar, J. Chem. Soc., Dalton
Trans., 1997, 881.
20 P. D. Beer, A. R. Graydon, A. O. M. Johnson and D. K. Smith,
Inorg. Chem., 1997, 36, 2112.
21 Z. Chen, A. R. Graydon and P. D. Beer, J. Chem. Soc., Faraday
Trans., 1996, 92, 97.
44 J. M. Lloris, R. Martínez-Máñez, E. Perales and J. Soto, J. Chem.
Res. (S), 1998, 432.
45 J. M. Lloris, R. Martínez-Máñez, M. Padilla-Tosta, T. Pardo and
M. J. L. Tendero, J. Chem. Soc., Dalton Trans., 1998, 3657.
46 PCMODEL, Molecular Modelling for Personal Computers and
Workstation, Serene Software, Bloomington, IN, 1988.
47 B. Dietrich, L. Fyles and M. Lehn, Helv. Chim. Acta, 1979, 2763.
48 J. M. Lloris, R. Martínez-Máñez, M. E. Padilla-Tosta, T. Pardo,
J. Soto, E. García-España, J. A. Ramírez, M. I. Burguete, S. V. Luis
and E. Sinn, J. Chem. Soc., Dalton Trans., 1999, 1779.
49 A. Andrés, J. Aragó, A. Bencini, A. Bianchi, A. Domenech, V. Fusi,
E. Gracía-España, P. Paoletti and J. A. Ramírez, Inorg. Chem., 1993,
32, 3418.
22 C. Li, J. C. Medina, G. E. M. Maguire, E. Abel, J. L. Atwood and
G. W. Gokel, J. Am. Chem. Soc., 1997, 119, 1609.
23 J. D. Carr, L. Lambert, D. E. Hibbs, M. B. Hursthouse, K. M. A.
Malik and J. H. R. Tucker, Chem. Commun., 1997, 1649.
24 H. Plenio, D. Burth and R. Vogler, Chem. Ber., 1997, 130, 1405.
25 P. D. Beer, M. G. B. Drew and D. K. Smith, J. Organomet. Chem.,
1997, 543, 259.
26 P. D. Beer, Z. Chen, M. G. B. Drew, A. O. M. Johnson, D. K. Smith
and P. Spencer, Inorg. Chim. Acta, 1996, 246, 143.
27 H. Yamamoto, A. Ori, K. Ueda, C. Dusemund and S. Shinkai,
Chem. Commun., 1996, 407.
50 A. Benito, R. Martínez-Máñez, J. Soto and M. J. L. Tendero,
J. Chem. Soc., Faraday Trans., 1997, 93, 2175.
28 R. C. Mucic, M. K. Herrlein, C. A. Mirkin and R. L. Letsinger,
Chem. Commun., 1996, 555.
29 H. Plenio and R. Diodone, J. Organomet. Chem., 1995, 495, 73.
30 A. Ori and S. Shinkai, J. Chem. Soc., Chem. Commun., 1995,
1771.
51 P. D. Beer, J. Cadman, J. M. Lloris, R. Martínez-Máñez, M. E.
Padilla-Tosta, T. Pardo, D. K. Smith and J. Soto, J. Chem. Soc.,
Dalton Trans., 1999, 127.
52 J. M. Lloris, R. Martínez-Máñez, M. E. Padilla-Tosta, T. Pardo and
J. Soto, Helv. Chim. Acta, 1999, 82, 1445.
1812
J. Chem. Soc., Dalton Trans., 2000, 1805–1812