Synthesis, protonation and Cu2+ co-ordination studies on a new family of thiophenophane receptors

Article abstract of DOI:10.1039/a900700h

The synthesis of the new thiophenophanes 2,5,8,11-tetraaza[12](2,5)thiophenophane (L¹), 2,6,9,13-tetraaza-[14](2,5)thiophenophane (L²) and 2,5,8,11,14-pentaaza[15](2,5)thiophenophane (L³) is described. The stepwise protonation constants of the macrocycles are determined by pH-metric titration and their protonation patterns are analysed by means of ¹H and ¹³C NMR spectroscopy. L¹ and L² present first protonation on the thenylic nitrogens while L³ protonates first on the central nitrogens of the polyamine bridge. Molecular orbital calculations are used to locate the atomic orbitals contributing to the HOMO of the free amines. The interaction with Cu²⁺ shows for L¹ and L³ formation of binuclear complexes which readily hydrolyse to give very stable species. Electrochemical studies show for all three Cu²⁺-receptor systems an important stabilisation of the Cu⁺ oxidation state towards its disproportionation into Cu²⁺ and Cu⁰ oxidation states.

Full text of DOI:10.1039/a900700h

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, protonation and Cu2؉ co-ordination studies on a new
family of thiophenophane receptors†
Juan A. Aguilar,a Pilar Díaz,b Antonio Doménech,c Enrique García-España,b José M. Llinares,a
Santiago V. Luis,d José A. Ramírezb and Conxa Sorianoa
a Departamento de Química Orgánica, Facultad de Farmacia, Universidad de Valencia,
46100 Burjassot, Valencia, Spain
b Departamento de Química Inorgánica, Universidad de Valencia, C/ Dr. Moliner 50,
46100 Burjassot, Valencia, Spain
c Departamento de Química Analítica, Universidad de Valencia, C/ Dr. Moliner 50,
46100 Burjassot, Valencia, Spain
d Departamento de Química Inorgánica y Orgánica, Universidad Jaume I, Carretera Borriol s/n,
12080 Castellón, Spain
Received (in Cambridge) 26th January 1999, Accepted 12th April 1999
The synthesis of the new thiophenophanes 2,5,8,11-tetraaza[12](2,5)thiophenophane (L1), 2,6,9,13-tetraaza-
[14](2,5)thiophenophane (L2) and 2,5,8,11,14-pentaaza[15](2,5)thiophenophane (L3) is described. The stepwise
protonation constants of the macrocycles are determined by pH-metric titration and their protonation patterns are
analysed by means of 1H and 13C NMR spectroscopy. L1 and L2 present first protonation on the thenylic nitrogens
while L3 protonates first on the central nitrogens of the polyamine bridge. Molecular orbital calculations are used
to locate the atomic orbitals contributing to the HOMO of the free amines. The interaction with Cu2ϩ shows for
L1 and L3 formation of binuclear complexes which readily hydrolyse to give very stable species. Electrochemical
studies show for all three Cu2ϩ–receptor systems an important stabilisation of the Cuϩ oxidation state towards
its disproportionation into Cu2ϩ and Cu0 oxidation states.
In recent years we have been investigating the chemistry of a
series of cyclophanes characterised by the presence of single or
condensed aromatic spacers interrupting polyamine chains with
different numbers of nitrogen donors and a variety of hydro-
carbon chains between them. The acid–base properties and
co-ordination chemistry of these ligands were strongly affected
by the type of spacer introduced as well as by the length and
number of nitrogen atoms in the polyamine bridge.1–10
For instance, and contrary to expectations, NMR and micro-
calorimetric studies proved that the benzylic nitrogens were the
first sites bearing protonation in the tri- or tetraaza macrocycles
containing either 1,4-benzene or 1,4-durene‡ spacers. Moreover,
the tetraazaparacyclophane 2,5,8,11-tetraaza[12]paracyclo-
phane (L4) was, as far as we know, one of the first tetraaza-
macrocycles containing a continuous set of nitrogen donors
linked by ethylenic chains able to co-ordinate two metal ions7
and, even more, formation of tetranuclear Cu2ϩ complexes was
observed when substituting the 1,4-benzene spacer in L4 for a
1,4-naphthalene moiety (L10). In the latter case, each macro-
cycle co-ordinated two Cu2ϩ ions being the metal ions of the
two different subunits interconnected by µ-hydroxo bridges.8
A further step in the development of this research was the
introduction of spacers themselves containing additional
heteroatoms. Here we present the synthesis and acid–base
behaviour of the thiophenophane ligands 2,5,8,11-tetraaza-
[12](2,5)thiophenophane (L1), 2,6,9,13-tetraaza[14](2,5)thio-
phenophane (L2) and 2,5,8,11,14-pentaaza[15](2,5)thiopheno-
phane (L3). Additionally, in order to better understand the
protonation patterns of these compounds, we have performed
preliminary molecular orbital calculations to localise the fron-
tier orbitals in the free amines. Finally, the Cu2ϩ co-ordination
characteristics of the three receptors have been examined from
both thermodynamic and electrochemical points of view. Since
related receptors containing 1,4-benzene or 1,4-durene spacers
yielded significant stabilisation of the Cuϩ oxidation towards
disproportionation,9,10 special interest has been devoted to this
aspect.
Experimental
Synthesis of the ligands
2,5-Bis(bromomethyl)thiophene. 2,5-Bis(bromomethyl)thio-
phene was obtained by reacting 2,5-dimethylthiophene and
N-bromosuccinimide in CCl4. The reaction was induced with
visible light. The resulting mixture was filtered and the solution
evaporated to dryness. The solid residue was recrystallised from
1
hexane, yield 69%. H NMR δH (CDCl3), 4.66 (s, 4H), 6.93 (s,
2H).
2,5,8,11-Tetrakis(p-tolylsulfonyl)-2,5,8,11-tetraaza[12](2,5)-
thiophenophane.
1,4,7,10-Tetrakis(p-tolylsulfonyl)-1,4,7,10-
tetraazadecane (6.4 g, 7.4 mmol)1 and K2CO3 (24 mmol) were
suspended in refluxing CH3CN (300 cm3). To the mixture was
added dropwise a freshly prepared solution of 2,5-bis(bromo-
methyl)thiophene in CH3CN (100 cm3). After the addition was
completed, the suspension was refluxed for 24 h, then filtered
and the solution evaporated to dryness. The solid product was
recrystallised from dichloromethane–hexane, yield 64%, mp
1
1
† Figure S1 (Plots of H–1H and H–13C correlated NMR spectra of
triprotonated L3 at pH = 4.5 recorded in D2O at room temperature) and
Figure S2 (Plots of the 13C NMR chemical shifts of the aliphatic carbon
atoms as a function of pH. A) L1, B) L2 and C) L3) are available as
supplementary data. For direct electronic access see http://www.rsc.org/
suppdata/p2/1999/1159, otherwise available from BLDSC (SUPPL.
NO. 57539, pp. 4) or the RSC Library. See Instructions for Authors,
available via the RSC web page (http:www.rsc.org/authors).
1
195–198 ЊC; H NMR δH (CDCl3), 2.46 (s, 12H), 2.75 (s, 4H),
‡ Durene = 1,2,4,5-tetramethylbenzene.
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168
1159

View Article Online
2.97–3.05 (m, 4H), 3.06–4.01 (m, 4H), 4.26 (s, 4H), 6.85 (s, 2H),
7.33 (d, J = 8 Hz, 4H), 7.38 (d, J = 8 Hz, 4H), 7.70 (d, J = 8 Hz,
4H), 7.75 (d, J = 8 Hz, 4H); 13C NMR δC (CDCl3), 21.2(2), 47.0,
48.1, 49.0, 50.0, 127.0, 127.3, 128.1, 129.6, 129.7, 133.0, 135.0,
138.8, 143.5, 143.9; MS m/z (FAB) 873 (M ϩ 2H)ϩ.
The computer program SUPERQUAD,14 was used to
calculate the protonation and stability constants. The DISPO15
program was used to obtain the distribution diagrams. The
titration curves for each system (ca. 150 experimental points
corresponding to at least three measurements, pH range
investigated 2–10.5, concentration of ligands was 2 × 10Ϫ3
mol dmϪ3 and that of Cu2ϩ was in the range 1 × 10Ϫ3–4 × 10Ϫ3
mol dmϪ3) were treated either as a single set or as separ-
ated curves without significant variations in the values of the
stability constants. Finally, the sets of data were merged
together and treated simultaneously to give the final stability
constants.
2,5,8,11-Tetraaza[12](2,5)thiophenophane (L1). 2,5,8,11-
Tetrakis(p-tolylsulfonyl)-2,5,8,11-tetraaza[12](2,5)thiopheno-
phane (2 g, 2.3 mmol), NaH2PO4 (5 g, 35 mmol) and 4%
sodium amalgam (50 g) were suspended in a mixture of
MeOH–THF (5 cm3 :100 cm3) and refluxed for 72 h. The sus-
pension was filtered, and the organic solvents evaporated. To
the residue was added water (5 cm3) and this was then extracted
with CH2Cl2. After vacuum evaporation, the oily product was
chromatographed on silica (MeOH–NH3) to give the pure
product. The hydrobromide salt was finally obtained by adding
aqueous HBr to an ethanolic solution of the free amine, yield
46%, mp 215–216 ЊC; 1H NMR δH (D2O), 3.00 (s, 4H), 3.02 (t,
J = 7 Hz, 4H), 3.29 (t, J = 7 Hz, 4H), 4.53 (s, 4H), 7.31 (s, 2H);
13C NMR δC (D2O), 42.1, 43.5, 43.6, 45.9, 134.9, 135.2. Anal.
Calc. for C12H26Br4N4S: C, 25.1; H, 4.5; N, 9.8. Found: C, 25.1;
H, 4.6; N, 9.8%.
Spectroscopy
The 1H and 13C NMR spectra were recorded on Varian UNITY
300 and UNITY 400 spectrometers, operating at 299.95 and
1
399.95 MHz for H and at 75.43 and 100.58 MHz for 13C
respectively. The spectra of compounds L1–L3 were obtained
at room temperature in D2O or DMSO solutions. For the 13C
NMR spectra dioxane was used as a reference standard
1
(δC = 67.4 ppm) and for the H spectra the solvent signal. The
pH was calculated from the measured pD values using the
correlation, pH = pD Ϫ 0.4.16
2,6,9,13-Tetrakis(p-tolylsulfonyl)-2,6,9,13-tetraaza[14](2,5)-
thiophenophane. Yield, 90%, mp 191–194 ЊC; 1H NMR δH
(CDCl3), 1.63–1.65 (m, 4H), 2.40 (s, 6H), 2.42 (s, 6H), 2.84 (s,
4H), 3.01–3.07 (m, 8H), 4.25 (s, 4H), 6.79 (s, 2H), 7.24–7.34 (m,
8H), 7.61–7.70 (m, 8H); 13C NMR δC (CDCl3), 21.5(2), 29.4,
47.8, 47.9, 48.0, 49.7, 127.1, 127.2, 127.3, 129.7, 129.9, 135.0,
135.1, 140.8, 143.5, 143.8; MS m/z (FAB) 899 (M Ϫ H)ϩ.
Electrochemistry
Cyclic voltammograms were obtained using the three-electrode
arrangement already described.9 Hanging mercury drop elec-
trode and glassy carbon were used as the working electrodes. A
saturated calomel reference electrode (SCE) was used as a refer-
ence electrode. Differential pulse voltammetric experiments
were performed with a Metrohm E506 Polarecord stand. The
potential scan rate was 1–10 mV sϪ1. Electrochemical experi-
ments were performed in solutions of Cu(NO3)2ؒ3H2O or
Cu(ClO4)2ؒ2H2O (Merck) in doubly distilled water. Ligands
were prepared and purified as already described. NaClO4 0.15
mol dmϪ3 was used as supporting electrolyte. The pH was
adjusted to the required value by adding the appropriate
amounts of HClO4 and/or NaOH.
2,6,9,13-Tetraaza[14](2,5)thiophenophane (L2). Obtained as
1
its tetrahydrobromide salt. Yield, 46%. H NMR δH (D2O),
1.86–1.96 (m, 4H), 3.02 (t, J = 4 Hz, 4H), 3.07 (t, J = 6 Hz,
4H), 3.18 (s, 4H), 4.40 (s, 4H), 7.15 (s, 2H); 13C NMR δC (D2O),
22.8, 42.5, 43.7, 44.9, 45.2, 133.9, 134.5. Anal. Calc. for
C14H30Br4N4S: C, 27.7; H, 5.0; N, 9.2. Found: C, 28.1; H, 5.2;
N, 8.9%.
2,5,8,11,14-Pentakis(p-tolylsulfonyl)-2,5,8,11,14-pentaaza-
[15](2,5)thiophenophane. Yield 66%, mp 260–265 ЊC; 1H NMR
δH (CDCl3), 2.33 (s, 3H), 2.34 (s, 6H), 2.35 (s, 6H), 3.03–3.04 (m,
16H), 4.27 (s, 4H), 6.64 (s, 2H), 7.21 (d, J = 8 Hz, 4H), 7.26 (d,
J = 8 Hz, 6H), 7.51 (d, J = 8 Hz, 2H), 7.61 (d, J = 8 Hz, 4H),
7.64 (d, J = 8 Hz, 4H); 13C NMR δC (CDCl3), 21.6(3), 46.3, 48.4,
50.5(2), 51.4, 127.5, 127.6, 127.7, 128.4, 130.0(2), 133.7, 134.7,
135, 139.0, 144.0, 144.1; MS m/z (FAB) 1068 (M Ϫ H)ϩ.
Extended Hückel molecular orbital calculations
All calculations were performed by using a package of pro-
grams for molecular orbital analysis by Mealli,17 based on
CDNT (atom Cartesian co-ordinate calculations), ICON
(extended Hückel method with the weighted Hij formula) and
FMO (fragment molecular orbital) including the drawing
program CACAO (computer-aided composition of atomic
orbitals). The optimisation of the molecular conformations was
performed by using the program HYPERCHEM.18
2,5,8,11,14-Pentaaza[15](2,5)thiophenophane (L3). Obtained
1
as its pentahydrobromide salt. Yield 44%, mp 235–240 ЊC; H
NMR δH (D2O), 3.04 (t, J = 6 Hz, 4H), 3.12 (t, J = 6 Hz, 4H),
3.29 (s, 8H), 4.52 (s, 4H), 7.24 (s, 2H); 13C NMR δC (D2O), 42.5,
43.8, 44.9, 45.8, 46.0, 134.3, 134.7. Anal. Calc. for C14H32Br5-
N5S: C, 23.9; H, 4.6; N, 10.0. Found: C, 24.0; H, 4.67; N, 10.0%.
Results and discussion
Synthesis
The synthesis of the thiophenophanes L1–L3 was carried out by
a modification of the general Richman and Atkins procedure
employed for the preparation of azamacrocycles.19 This
method, which we had previously used for the synthesis of the
related azaparacyclophanes L4–L6,1 consists of the reaction of
the corresponding tosylated polyamine with 2,5-bis(bromo-
methyl)thiophene in refluxing CH3CN using K2CO3 as the
base. It is to be noted that high dilution conditions are not
required to obtain the cyclised products in high yields (64–
90%). 2,5-Bis(bromomethyl)thiophene was generated in situ
by reaction of N-bromosuccinimide and 2,5-dimethylthio-
phene, being induced by visible light. Detosylation of the cyclic
amine was best carried out with a Na/Hg amalgam in buffered
methanolic solution.20 When necessary purification was per-
formed by silica gel chromatography with MeOH–NH3 as the
eluent.1
Emf measurements
The potentiometric titrations were carried out in 0.15 mol dmϪ3
NaClO4 or 0.15 mol dmϪ3 NaCl at 298.1 0.1 K, by using the
experimental procedure (burette, potentiometer, cell, stirrer,
microcomputer, etc.) that has been fully described elsewhere.11
The acquisition of the emf data was performed with the
computer program PASAT.12 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 known amounts of HCl with CO2-free NaOH
solutions and determining the equivalent point by the Gran’s
method,13 which gives the standard potential, Eo, and the ionic
product of water (pKw = 13.73(1)). The concentrations of
Cu2ϩ solutions were determined gravimetrically by standard
methods.
1160
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168

View Article Online
Table 1 Logarithms of the stepwise protonation constants of the cyclophanes L1–L3 determined in 0.15 mol dmϪ3 NaCl at 298.1 K. The stepwise
protonation constants of the cyclophanes L4–L6 and those of the dibenzylated open-chain polyamines L7–L9 are also included for comparison
Reaction
L1a
L2a
L3a
L4b
L5b
L6b
L7c
L8c
L9c
L ϩ H↔HLd
9.37(1)e
8.35(2)
5.04(2)
2.69(4)
10.22(2)
8.90(3)
7.39(4)
4.20(6)
10.36(3)
9.18(2)
8.02(3)
4.16(4)
9.39
8.45
5.38
2.51
9.93
9.09
7.44
3.61
10.68
9.29
8.66
7.23
3.93
9.30
8.62
6.41
3.77
9.68
8.87
7.37
4.90
9.85
8.89
7.99
4.93
2.75
HL ϩ H↔H2L
H2L ϩ H↔H3L
H3L ϩ H↔H4L
H4L ϩ H↔H5L
a This work. b Taken from ref. 6, 0.15 mol dmϪ3 NaClO4, 298.1 K. c Taken from ref. 6, 0.15 mol dmϪ3 NaClO4, 298.1 K. d Charges have been omitted
for clarity. e Numbers in parentheses are standard deviations of the last significant figure.
basicity (log βH L = 25.5) than its related open-chain counter-
4
part L7 (log βH L = 28.1). However, L2 presents almost the same
4
overall basicity as L8 (log βH L = 30.7 and 30.8, respectively).
4
A similar behaviour was again observed for azaparacyclo-
phanes L4 and L5.
Significant differences are found for the next terms of both
series of cyclic receptors, L3 and L6 (see Scheme 1). While both
receptors present large basicities in their three first protonation
steps, L3 shows a much reduced constant for its fourth proton-
ation step. Moreover, for L3 just four out of the five possible
protonation constants were measurable in the pH range avail-
able for the potentiometric studies (pH 2.0–11.0). On the other
hand, although L3 and L9 present similar basicities in their
three first protonation steps, the constant for the fourth proton-
ation step is significantly lower for L3, which can probably be
attributed to a higher charge accumulation at this stage in the
Fig. 1 Distribution for the species existing in equilibria in the system
Hϩ–L3.
cyclic ligand.
In order to get further insight into the protonation patterns
Acid–base behaviour
1
of these compounds we have studied the variations of their H
1
In Table 1 are presented the logarithms of the basicity constants
of the thiophenophanes L1-L3 together with those we have pre-
viously reported for the related azaparacyclophanes L4–L6 and
those for the non-cyclic dibenzylated polyamines L7–L9.21,22
In Fig. 1 is plotted the distribution diagram for L3.
and 13C NMR chemical shifts with pH. The H NMR spec-
trum of L1 at pH 12.0, where the non-protonated free amine
predominates, consists of singlet signals at 3.78 and 2.28 ppm
(protons H1 and H4, respectively; for the numbering see
Scheme 1) and of two triplet signals at 2.57 and 2.35 ppm (pro-
tons H2 and H3, respectively). The aromatic protons (HT2)
appear at this pH as a singlet signal at 6.80 ppm. The 13C NMR
spectrum consists at the same pH of six signals at 46.27, 47.49,
47.76, 48.43, 128.28 and 144.32 ppm that can be attributed to
carbon atoms C2, C4, C1, C3, CT2 and CT1 respectively. The
number of 1H and 13C signals remains constant throughout the
whole of the studied pH range indicating that the two-fold
symmetry presented by these ligands is preserved upon proton-
ation of the free amines. Similar spectral characteristics are dis-
played by L2 and L3 which both present seven 13C NMR signals.
Nevertheless, the protonation of the nitrogens produces signifi-
cant changes in the 1H spin-systems that will be discussed
below.
L1 displays a group of two relatively high constants (log
KHL = 9.37(1), log KH L = 8.35(2)), an intermediate one (log
2
KH L = 5.04(2)) and a much lower constant for the last proton-
3
ation step (log KH L = 2.69(4)). L2, with a symmetric set of two
propylenic chains4 and one central ethylenic chain, presents
relatively high basicity in its three first protonation steps and
much reduced basicity in its last protonation step (Table 1).
All this grouping of constants can be explained by consider-
ing the electrostatic repulsion between charged ammonium
groups to be the major factor affecting the successive proton-
ation steps. For instance, in the case of L1, while both the first
two protonations may occur on non-adjacent nitrogen atoms,
the third one should take place on a nitrogen atom adjacent to
one already protonated nitrogen and the last one on a nitrogen
next to two already protonated nitrogens producing, therefore,
a drastic reduction in the basicity constant. Similar consider-
ations would explain the sequence of protonation constants
found for L2 although, in this case, the presence of the larger
propylenic chains yields more moderate reductions in basicity
in the different steps. As a matter of fact, the constant found for
the last protonation of L2 (log KH L = 4.20(6)), which has to
occur on a nitrogen atom separated4 from the adjacent ammo-
nium groups by an ethylenic and a propylenic chain, is clearly
All the assignments have been made on the basis of two-
1
1
dimensional H–1H and H–13C correlations as well as taking
into account the data in the bibliography for similar systems. It
is to be noted that, for these compounds, the 13C chemical shifts
of the thenylic (thenyl = thienylmethyl) carbon atoms show a
very important upfield shift with respect to the analogue ben-
zylic carbons of the benzene counterparts; their chemical shifts
being similar to those observed for the methylenic carbons of
the ethylenic chains.
As is well known, upon protonation the signals of the pro-
tons placed α and of the carbon atoms placed β to the nitrogen
atoms bearing protonation are the ones shifting most.23,24 In
fact, in this kind of compound the variations with pH of the
chemical shifts of the quaternary carbon atoms of the aromatic
rings are particularly useful to establish their average proton-
ation patterns, since they are only in the β-position with respect
to nitrogen atoms labelled N1 (Scheme 1).6,21,22 In Fig. 2 such
variations are plotted for all three ligands. For L1 the largest
upfield shifts are produced on going from pH 10 to 5 and for L2
from pH 10 to 7, suggesting that for both ligands the first two
higher than the corresponding one for L1 (log KH L = 2.69(4)).
4
These basicity trends parallel the behaviour displayed by the
related cyclophanes L4 and L5 as is reflected by the similar bas-
icity constants found for both types of macrocycles (Table 1).21
The effects of the cyclic topology on basicity can be analysed
by comparing the basicities of the different cyclophane families
with those of their dibenzylated open-chain counterparts.
A larger charge accumulation due to the cyclic topology is
only observed for the smallest thiophenophane L1 containing
just ethylenic chains, which displays significant lower overall
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168
1161

View Article Online
Scheme 1
protonations are mainly affecting the thenylic nitrogen atoms
(see Scheme 2). In L1, this is further confirmed by the upfield
shift experienced over the same pH range by the signal of
carbon C3, also in a position β to N1, and by the signals of
carbons C2 and C4, both β to the central nitrogen of the
polyamine chain (N2) which only shift significantly upfield
below pH 5, when the third proton binds to L1 (see supplemen-
tary Figure S2).† Similarly, for L2 the signal assigned to C5, β to
N2, only moves upfield below pH 7, again corresponding to the
third protonation. The signal of the central carbon of the pro-
pylenic chain (C3) continuously shifts upfield throughout all
the pH range in agreement with its location β with respect to all
nitrogen atoms of L2. All these data lead us to infer the average
protonation mechanism proposed in Scheme 2.
part of the polyamine chain (N2 and N3). The signal of C5
shifts upfield from pH 12 to 7.5, remains almost steady from pH
7.5 to 4.5 and again moves upfield below pH 4.5 when the
fourth protonation occurs (see Fig. 3). The third protonation
causes an important downfield shift of the signal C2, probably
suggesting that at this stage a reorganisation of the molecule is
produced in such a way that a minimum electrostatic repulsion
situation is achieved. This would imply the location of the three
protons on non-adjacent nitrogens throughout the molecule
with the consequent change in the initial position of the two
first protons (see Scheme 2).
1
The H NMR spectra of L3 recorded at different pH values
also reflect this situation (Fig. 3). In this sense, the changes that
protonation produces in the spin-systems of the protons of the
ethylenic chains somehow reveal the actual protonation state of
the compounds. At pH 12, where the free amine predominates,
the proton signals of the two ethylenic chains present A4 spin-
systems. The first protonation yields a change in the spin-
system from A4 to AAЈBBЈ of the protons of the ethylenic
chains defined by carbon atoms C4–C5. This lower equivalence
The situation for L3 is however somewhat different. In Fig.
2C it can be seen that, in this case, the signal of the carbon atom
(CT1) just moves upfield on going from pH 8 to 6 which, as
shown in the distribution diagram (see Fig. 1), corresponds to
the pH range when the third protonation takes place. Therefore,
both of the first two protonations should occur in the central
1162
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168

View Article Online
Table 2 Logarithms of the formation constants of the Cu2ϩ complexes of the receptors L1–L3, determined in 0.15 mol dmϪ3 NaCl at 298.1 K.
Stability constants for the copper() complexes of paraazacyclophanes L4–L6 are also included
Reactiona
L1
L2
L3
L4c
L5c
L6
Cu ϩ L = CuL
11.89(3)b
18.48(1)
22.24(3)
3.14(4)
6.59
12.39(2)
20.13(5)
26.62(4)
3.26(5)
7.74
14.46(4)
21.86(6)
27.25(8)
—
7.4
5.39
10.41
16.92
—
2.27
6.51
13.02
20.82
—
3.92
7.80
17.73
26.86
33.28
—
Cu ϩ L ϩ H = CuHL
Cu ϩ L ϩ 2H = CuH2L
Cu ϩ L ϩ H2O = CuL(OH) ϩ H
CuL ϩ H = CuHL
9.13
CuHL ϩ H = CuH2L
3.76
6.49
CuL ϩ H2O = CuL(OH) ϩ H
2Cu ϩ L = Cu2L
2Cu ϩ L ϩ H2O = Cu2L(OH) ϩ H
2Cu ϩ L ϩ 2H2O = Cu2L(OH)2 ϩ 2H
2Cu ϩ L ϩ 3H2O = Cu2L(OH)3 ϩ 3H
CuL ϩ Cu = Cu2L
Ϫ8.75
—
10.26(2)
3.76(2)
Ϫ6.58(3)
Ϫ9.13
—
—
—
—
Ϫ8.14
—
—
3.44
—
Ϫ9.10
—
—
—
—
Ϫ11.08
24.29
16.95
7.50
—
6.56
19.23(6)
12.54(5)
2.45(8)
—
4.8
Ϫ6.7
Cu2L ϩ H2O = Cu2L(OH) ϩ H
Cu2L(OH) ϩ H2O = Cu2L(OH)2 ϩ H
Ϫ7.34
Ϫ9.45
Ϫ6.5
Ϫ9.89
a Charges omitted for clarity. b Numbers in parentheses are standard deviations of the last significant figure. c Constants taken from references 4 and
7 and written without standard deviations.
of the protons is due to protonation of N2. Between pH 7.5 and
4.5 both different ethylenic chains display an AAЈBBЈ spin-
system in agreement with the proposed alternate disposition of
charged nitrogens in the macrocycle. At lower pH, the observed
spin-systems reflect the greater equivalence of the chemical
environment of the molecule and protons of carbons C4 and
C5 recover the A4 pattern already observed in the pH values
range where all nitrogens were non-protonated.
The protonation behaviours observed for these compounds
correspond quite closely to those displayed by their related
azaparacyclophanes and dibenzylated open-chain polyamines.
Namely, L1 behaves similarly to L4 and L7 while L2 presents an
analogous protonation pattern to those of L5 and L8. On the
other hand, at least in its first protonation steps, L3 resembles
L6 and L9.
Whether the thenylic or benzylic nitrogens protonate in the
first place or not should depend on the local electron densities
of the different nitrogens as well as on solvation effects. Since
an acid–base process implies electron sharing between the
HOMO of the base and the LUMO of the acid, identifying the
location of the HOMO of the base could help in understanding
the discussed protonation behaviours. Protonation would be
favoured on the nitrogen atoms whose atomic orbitals contrib-
ute to the HOMO. Indeed, molecular orbital calculations per-
formed with the program CACAO17 show the participation of
atomic orbitals from the aromatic ring and from the thenylic or
benzylic nitrogens in the HOMO of the free amines L1, L2, L4
and L5 for which the NMR study supported a first protonation
occurring on such nitrogens (see Fig. 4).
Also in agreement with the spectroscopic results, the molecu-
lar orbital calculations show that, in the case of L3 the HOMO
is made up exclusively from the atomic orbitals of the aromatic
unit suggesting a more random first protonation. Similar
treatment shows for L6 an important contribution of the atomic
orbitals of the central nitrogen atoms to the HOMO, again in
good agreement with the protonation pattern found.21 An
analogous situation is observed for L3 in the highest molecular
orbital occupied by an amine lone pair, which is similar to
the HOMO of L3 being also made up preferentially from the
atomic orbitals of the central nitrogen atoms.
Cu2ϩ co-ordination
Stability constants. Table 2 shows the logarithms of the
stability constants for the formation of Cu2ϩ complexes of
thiophenophanes L1–L3. By means of comparison, in the same
Table 2 are also included the formation constants of the Cu2ϩ
complexes of the related paraazacyclophanes L4–L6, taken
from the literature.4,7 For molar ratios [Cu2ϩ]/[L] = 1 the speci-
Fig. 2 Plots of the 13C NMR chemical shifts of the aromatic carbon
atoms CT1 and CT2 as a function of pH. A) L1, B) L2 and C) L3.
ation model of all three receptors is rather similar, and in the
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168
1163

View Article Online
Fig. 3 1H and 13C NMR spectra of L3 recorded at different pH values.
Scheme 2
1164
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168

View Article Online
Fig. 4 HOMO orbitals of ligands L1–L6.
species are detected and, in this case, formation of Cu2ϩ hydrox-
ide can be observed above pH 8.
The stability constants of the [CuL]2ϩ species of all three
thiophenophanes are rather low in comparison with related
tetraaamines or pentaazacycloalkanes either of an open-chain
or a cyclic nature being, however, much closer to those reported
for triamines.25 On the other hand, the first protonation con-
stants of the [CuL]2ϩ species (see Table 2) are very large and
present comparable or even higher magnitudes than the proton-
ation constants of the corresponding doubly charged cation
[H2L]2ϩ in the free receptors (Table 1), suggesting that proton-
ation of the Cu2ϩ complexes occurs on non-co-ordinated nitro-
gen atoms. Therefore, all these data support an incomplete
involvement of all the nitrogen donors of the receptors in the
co-ordination to a single metal ion. The UV-Vis spectra of
these systems also give support to this conclusion. For instance,
in the system Cu2ϩ–L2 the visible spectra of the [CuL]2ϩ and
[CuHL]3ϩ species are identical (λ = 597 nm, ε = 153 mol dmϪ3).
Probably the most noticeable feature in the chemistry of
these ligands is the formation of Cu2ϩ binuclear complexes by
L1 and L3. Interestingly, the receptor L2 of intermediate size
between L1 and L3 does not form this kind of complex. This
behaviour, which parallels that found for paraazacyclophanes
L4–L6,7 can be explained in terms of a balance between flex-
ibility, size and the number of nitrogen atoms of the receptors.
The larger the adaptability of the polyamine bridge the better
it can fit the coordination requirements of a single metal ion,
whereas a large number of nitrogen donors will tend to favour
the formation of polynuclear complexes.
Fig. 5 Distribution diagrams for the systems Cu2ϩ–L1 (A) and Cu2ϩ
L3 (B) ([Cu2ϩ] = 2 × 10Ϫ3 mol dmϪ3, [L] = 10Ϫ3 mol dmϪ3).
–
The important Cu2ϩ hydroxylated species formed (see Fig. 5)
together with the low number of nitrogen donors involved in
the first co-ordination sphere of the metals, make these systems
potential catalysts for the activation of small molecules. Indeed,
many metallobiomolecules with hydrolytic or redox catalytic
activity present co-ordination environments displaying similar
features.
pH range 2.5–10.5 formation of [CuH2L]2ϩ, [CuHL]3ϩ and
[CuL]2ϩ species is detected. The only difference among them
comes from the fact that while L1 and L2 also form a hydroxyl-
ated [CuL(OH)]ϩ species, no such species are detected for L3, at
least in the pH range explored. However the situation is quite
different for molar ratios [Cu2ϩ]/[L] = 2, since in these condi-
tions binuclear species [Cu2L(OH)]3ϩ, [Cu2L(OH)2]2ϩ and
[Cu2L(OH)2]2ϩ are found for L1 and [Cu2L]4ϩ, [Cu2L(OH)]3ϩ
and [Cu2L(OH)2]2ϩ for L3. These binuclear species, as shown in
the distribution diagram in Fig. 5, are predominant in solution
over a wide pH range. Interestingly enough, for L2 no such
Electrochemistry. The electrochemical behaviour of the
Cu2ϩ–thiophenophane systems is sensitive to the electrode
material, solution composition, pH changes and voltammetric
parameters. In acidic media no adduct species are formed and
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168
1165

View Article Online
Fig. 7 Cathodic differential pulse voltammograms for the Cu–L3
system in NaClO4 0.15 mol dmϪ3. A) Cu2ϩ 0.40 × 10Ϫ3 mol dmϪ3
plus L3 0.20 × 10Ϫ3 mol dmϪ3, pH 3.4; B) as for A, pH 5.0; C) as for A,
pH 10.7; D) Cu2ϩ 0.40 × 10Ϫ3 mol dmϪ3 plus L3 0.45 × 10Ϫ3 mol dmϪ3
,
pH 11.4. GCE, v = 20 mV sϪ1; pulse width 10 mV.
Fig. 6 Cyclic voltammograms at the hanging drop mercury electrode
(scan rate, 0.15 V sϪ1) for NaClO4 0.15 M solutions of A) Cu(ClO4)2
1.07 × 10Ϫ3 mol dmϪ3, pH = 2.35; B) as for A plus L5 1.12 × 10Ϫ3 mol
dmϪ3, pH = 8.60; C) as for A plus L2 1.22 × 10Ϫ3 mol dmϪ3, pH = 9.95.
CuI(HpL)(p ϩ 1) (sol) = CuI(HpL)(p ϩ 1) (ads)
CuI(HpL)(p ϩ 1) (ads) ϩ eϪ = Cu0 (Hg) ϩ HpLpϩ (5)
(4)
the CV is identical to that obtained in the absence of the
macrocyclic ligand, consisting of a single two-electron revers-
ible couple at the potential expected for the Nernstian Cu2ϩ/
Cu(Hg) reduction (Fig. 6A). As demonstrated by cyclic vol-
tammetry at fast scan rates, the electrochemical behaviour,
however, is more complicated, involving two successive one-
electron steps coupled with dehydration and hydrolysis reac-
tions of the parent Cu2ϩ and the intermediate Cuϩ species
(electrochemical chemical electrochemical (ECE) and dispro-
portionation mechanisms).26
be considered as operative (ECE mechanism). The intermediate
step must correspond to an adsorption process that can be
accompanied by a rearrangement of the coordination sites.
This scheme is supported by cyclic voltammetry at glassy
carbon electrodes. Although at solid electrodes the voltam-
metric response of Cu2ϩ solutions is complicated by adatom
diffusion on the electrode,29 formation of film oxides,30,31 etc.,
significant differences were obtained between macrocyclic
ligands containing benzene and thiophene groups.
For L1 and L3 dinuclear species are formed in neutral and
alkaline media. As can be seen in Fig. 7, differential pulse
voltammograms for solutions containing Cu2ϩ and the ligand in
a 2:1 stoichiometric ratio exhibit significant variations upon
pH changes. Below pH 4, two well-defined peaks at Ϫ0.20 and
Ϫ0.87 V appear, corresponding to successive one-electron
reductions of copper (curve A). Upon increasing the pH, the
first cathodic peak is followed by a second overlapped peak at
Ϫ0.45 V (curve B). At alkaline pH values (curve C), two
overlapped peaks at Ϫ0.45 and Ϫ0.60 V appear. This electro-
chemical response suggests that relatively stable Cuϩ intermedi-
ate species are involved. In solutions containing a 1:1 Cu:L
stoichiometric ratio (curve D), only one peak close to Ϫ0.65 V
is recorded suggesting a chemical electrochemical (CE) mech-
anism involving the dissociation of the copper–macrocycle
adduct preceding a two-electron transfer reaction as described,
for instance, for hydroxylate and carbonate adducts.32
As previously reported,9,10 for neutral and alkaline media
solutions containing equimolar amounts of Cu2ϩ and tetra-
azacyclophanes with benzene or durene spacers two nearly
identical couples appear that can be assigned to the electrode
processes (eqns. (1) and (2)). This voltammetric pattern is repre-
CuIIHpL(p ϩ 2) ϩ eϪ = CuIHpL(pϩ1)
CuIHpL(p ϩ 1) ϩ eϪ = Cu0 ϩ HpLpϩ
(1)
(2)
sentative of the stabilisation of the Cuϩ species with respect to
its disproportionation into Cu2ϩ and Cu0 caused by co-
ordination to macrocyclic receptors (Fig. 6B).
This behaviour contrasts with that exhibited by the systems
consisting of Cu2ϩ–thiophenophane receptors, which is
depicted in Fig. 6C. At a mercury electrode, neutral and
alkaline solutions containing equimolar amounts of Cu2ϩ and
of the thiophenophane receptors display two quasi-reversible
overlapped couples (X,XЈ, Y,YЈ) in the potential region
between 0.2 and Ϫ0.2 V vs. SCE. In successive scans, both
cathodic peaks are enhanced. This feature is especially relevant
for peak Y, denoting that the electrode reaction is mediated by
adsorption of the electroactive species on the mercury elec-
trode.27 These voltammetric data may be adjusted to an ECE
mechanism28 in which a relatively slow reaction is coupled
between two successive electron charge transfers assuming
that the second transfer occurs at a potential less negative than
the first one, resulting in an apparent two-electron couple.
Accordingly, the overall electrochemical reactions (3)–(5) can
The cyclic voltammetric pattern previously described denotes
the stabilization of Cuϩ species with disproportionation into
Cu2ϩ and Cu0 (eqn. (6)). As previously described,9,10 a direct
I
2 Cum (HpL)(p ϩ m) = CumII(HpL)(2m ϩ p) ϩ mCu0 ϩ HpLpϩ (6)
estimate of the formation constants of CuI(HpL)(p ϩ 1) species
can be made, providing the formation constants of the CuII-
(HpL)(p ϩ 2) species are known, from the relationship (at 298 K)
(7); βI and βII refer to the cumulative formation constants of
the CuI and CuII complexes, respectively.
log βI = log βII ϩ m[Eo Ϫ Eo(Cu2ϩ/Cuϩ)]/59
(7)
CuII(HpL)(p ϩ 2) (sol) ϩ eϪ = CuI(HpL)(p ϩ1 ) (sol) (3)
Eqn. (7) holds at selected pH values for which only one
1166
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168

View Article Online
Table 3 Formal potential for the Cu2ϩ/Cuϩ couple and stability constants (0.15 mol dmϪ3 NaClO4 at 298 K) for the Cu2ϩ and Cuϩ complexes
with polyazamacrocyclic ligands containing aromatic spacers L. Equilibrium constants for the disproportionation process calculated from cyclic
voltammetric data
Species
pH
8.0
5.5
10.5
9.0
8.0
7.0
10.0
6.5
9.0
11.5
8.4
7.2
8.5
7.0
EoЈ/mV
log βII
log βI
log KЈd
Ϫ0.1
1.6
Ϫ4.2
Ϫ2.0
Ϫ2.9
Ϫ1.2
Ϫ1.1
Ϫ1.3
Ϫ0.6
Ϫ3.8
Ϫ4.5(5)
Ϫ2.2(5)
Ϫ5.6(4)
Ϫ2.0(4)
[Cu(L1)]2ϩ
Ϫ250
11.89
18.48
3.14
9.1
16.2
Ϫ0.08
Ϫ2.0
10.7
18.7
10.9
19.5
5.9
Ϫ4.6
11.9(2)
19.7(2)
12.6(2)
19.8(2)
[CuH(L1)]3ϩ
[Cu(L1)(OH)]ϩ
[Cu2(L1)(OH)2]2ϩ
[Cu(L2)]2ϩ
Ϫ220
Ϫ275
Ϫ255
Ϫ180
Ϫ170
Ϫ290
Ϫ220
Ϫ260
Ϫ280
Ϫ150
Ϫ150
Ϫ140
Ϫ140
3.76
12.28
20.16
14.37
21.77
12.54
2.0
13.02(1)
20.82(1)
13.35(3)
20.79(1)
[CuH(L2)]3ϩ
[Cu(L3)]2ϩ
[CuH(L3)]3ϩ
[Cu2(L3)(OH)]3ϩ
[Cu2(L3)(OH)2]2ϩ
[Cu(L5)]2ϩ a
[CuH(L5)]3ϩ a
[Cu(L10)]2ϩ a
[CuH(L11)]3ϩ a
a Values taken from references 9,10.
adduct species predominates in solution and that is also valid
for the hydroxylated complexes. The cumulative formation con-
stants of the Cuϩ complexes (βI) calculated from cyclic voltam-
metric data are presented in Table 3 in logarithmic form. An
estimate of the equilibrium constant, KЈd for the dispropor-
tionation reaction (6) can be obtained through the relation-
ship (8),9,10 where Kd (log Kd = 6.24 at 298 K) represents the
atom could also contribute to the observed change in the
mechanism.
We are currently extending these studies to unravel the
electrochemical reduction mechanism and to test the possible
use of these complexes as catalysts assisting a variety of
processes.
log KЈd = 2log βI Ϫ log βII Ϫ mlog Kd ϩ log βp (8)
Acknowledgements
We thank DGICYT Project No. PB96-0792-CO2 and General-
itat Valenciana Project No. GV-D-CN-09-140–16 for financial
support.
equilibrium constant for the disproportionation of the Cuϩ(aq)
into Cu2ϩ(aq) and Cu0. βp represents the cumulative proton-
ation constant for the HpLpϩ species. For the dismutation of the
Cu–L hydroxylated complexes, eqn. (9) holds, and the equation
References
(m Ϫ z)
2 CumI L(OH)z
=
1 A. Bencini, M. I. Burguete, E. García-España, S. V. Luis,
J. F. Miravet and C. Soriano, J. Org. Chem., 1993, 58, 4749.
2 M. I. Burguete, B. Escuder, E. García-España, S. V. Luis and
J. F. Miravet, J. Org. Chem., 1994, 59, 1067.
CumIIL(OH)x(2m Ϫ x) ϩ mCu0 ϩ L ϩ (2z Ϫ x)OHϪ (9)
for KЈd becomes eqn. (10).9,10 The KЈd values estimated for the
3 M. I. Burguete, B. Escuder, E. García-España, S. V. Luis and
J. F. Miravet, Tetrahedron Lett., 1994, 9075.
equilibrium constants are also summarized in Table 3.
4 A. Andrés, M. I. Burguete, E. García-España, S. V. Luis,
J. F. Miravet and C. Soriano, J. Chem. Soc., Perkin Trans. 2, 1993,
749.
log KЈd =
2log βI Ϫ log βII Ϫ mlog Kd Ϫ (2z Ϫ x)log Kw (10)
5 M. A. Bernardo, J. A. Parola, F. Pina, E. García-España,
V. Marcelino, S. V. Luis and J. F. Miravet, J. Chem. Soc., Dalton
Trans., 1995, 993.
6 E. García-España and S. V. Luis, Supramol. Chem., 1996, 6, 257.
7 A. Andrés, C. Bazzicalupi, A. Bianchi, E. García-España, S. V. Luis,
J. F. Miravet and J. A. Ramírez, J. Chem. Soc., Dalton Trans., 1994,
1995.
On comparing the stability constants βI and βII in Table 3 for
the different compounds, it can be concluded that substitution
of benzene or durene spacers by a thiophene group does not
increase the binding strength towards both Cu2ϩ and Cuϩ oxid-
ation states.
In this respect, it should be noted that contrary to the popu-
lar belief that sulfur donors stabilise the Cuϩ oxidation state,
Cuϩ exhibits little preference between saturated N and S donor
atoms.33 Thus, whereas the stability constants for a series of
Cu2ϩ complexes with polyamino polythiaether ligands increase
significantly upon increasing the number of sulfur atoms, the
stability constants for Cuϩ complexes remain constant. Accord-
ingly, eventual coordination with an aromatic S atom of low
nucleophilic character should not produce any particular
enhancement in the binding affinity of the metal ion.
The cyclic voltammetric data suggest that two different
coordinative arrangements of metallic centre are possible for
thiophene ligands. For Cu2ϩ adducts, the coordination takes
place mainly via nitrogen atoms as described for ligands con-
taining benzene and durene spacers and the electrochemical
reduction occurs in two reversible one-electron steps. Although,
as previously discussed, adsorption-mediated electrochemical
processes are involved for ligands containing thiophene spacers,
a possible weak co-ordination of the reduced Cuϩ by the sulfur
8 B. Escuder, E. García-España, J. Latorre, S. V. Luis and J. A.
Ramírez, in preparation.
9 A. Doménech, J. V. Folgado, E. García-España, S. V. Luís, J. M.
Llinares, J. F. Miravet and J. A. Ramírez, J. Chem. Soc., Dalton
Trans., 1995, 541.
10 A. Doménech, E. García-España, V. Marcelino, B. Altava, S. V.
Luís, J. F. Miravet, A. Bianchi and L. Ferrini, Inorg. Chim. Acta,
1996, 252, 123.
11 E. Garcia-España, M. J. Ballester, F. LLoret, J. M. Moratal, J. Faus
and A. Bianchi, J. Chem. Soc., Dalton Trans., 1988, 101.
12 M. Fontanelli and M. Micheloni, Proceedings of the I Spanish-
Italian Congress on Thermodynamics of Metal Complexes,
Peñíscola, Spain, 1990. Program for the automatic control of the
burette additions and electromotive force readings.
13 G. Gran, Analyst (London), 1952, 77, 661; F. J. Rossoti and
H. Rossoti, J. Chem. Educ., 1965, 42, 375.
14 P. Gans, A. Sabatini and A. Vacca, J. Chem. Soc., Dalton Trans.,
1985, 1195.
15 A. Vacca, University of Florence, unpublished work, a FORTRAN
Program to determine the distribution of the species in
multiequilibria systems from the stability constants and mass
balance equations.
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168
1167

View Article Online
16 A. K. Convington, M. Paabo, R. A. Robinson and R. G. Bates,
Anal. Chem., 1968, 40, 700.
17 C. Mealli and D. Proserpio, J. Chem. Educ., 1990, 67, 399.
18 HYPERCHEM 3.0 from Hypercube, Inc., Waterloo, ON, Canada.
VRL. http://www.hyper.com.
19 J. E. Richman and T. J. Atkins, J. Am. Chem. Soc., 1974, 96, 2268;
J. E. Richman, T. J. Atkins and W. F. Oetle, Organic Synthesis,
Wiley, New York, Collect. Vol. VI., p. 652.
20 F. Chavez and D. Sherry, J. Org. Chem., 1989, 54, 2990; B. K.
Vriesema, J. Buter and R. M. Kellog, J. Org. Chem., 1994, 49, 110;
D. M. Trost, H. C. Arndt, P. E. Strege and T. R. Verhoeven,
Tetrahedron Lett., 1976, 3477.
21 A. Bianchi, B. Escuder, E. García-España, S. V. Luis, V. Marcelino,
J. F. Miravet and J. A. Ramírez, J. Chem. Soc., Perkin Trans. 2, 1994,
1253.
22 M. A. Bernardo, J. A. Guerrero, E. García-España, S. V. Luis,
J. M. Llinares, F. Pina, S. V. Luis, J. A. Ramírez and C. Soriano,
J. Chem. Soc., Perkin Trans. 2, 1996, 2335.
25 A. E. Martell, R. M. Smith and R. M. Moteikaitis, NIST Critical
Stability Constants Database; Texas A&M, University College
Station, TX, 1993.
26 (a) M. Mastragostino, L. Nadjo and J. M. Saveant, Electrochim.
Acta, 1968, 13, 721; (b) M. Mastragostino and J. M. Saveant,
Electrochim. Acta, 1968, 13, 751; (c) F. M. Hawkridge and H. H.
Bauer, Anal. Chem., 1972, 44, 364; (d) J. L. Anderson and I. Shain,
Anal. Chem., 1976, 48, 1274; (e) J. L. Anderson and I. Shain, Anal.
Chem., 1978, 50, 163.
27 A. J. Bard and L. R. Faulkner, Electrochemical Methods, Wiley &
Sons, New York, 1980.
28 R. S. Nicholson and I. Shain, Anal. Chem., 1965, 37, 178.
29 L. H. Jenkins, J. Electrochem. Soc., 1970, 117, 630.
30 J. J. Kruger, J. Electrochem. Soc., 1959, 106, 847.
31 G. T. Miller and K. R. Lawless, J. Electrochem. Soc., 1959, 106, 854.
32 M. S. Shuman and L. C. Michael, Anal. Chem., 1978, 50, 2104.
33 M. M. Bernardo, M. J. Heeg, R. R. Schroeder, L. A. Ochrymowycz
and D. A. Rorabacher, Inorg. Chem., 1992, 31, 191.
23 J. E. Sarnesky, H. L. Surprenant, F. K. Molen and C. N. Reileg,
Anal. Chem., 1975, 47, 2116.
24 D. N. Hague and A. D. Moreton, J. Chem. Soc., Perkin Trans. 2,
1994, 265.
Paper 9/00700H
1168
J. Chem. Soc., Perkin Trans. 2, 1999, 1159–1168