Syntheses, and electro(co)polymerization of novel thiophene- and 2,2′:5′,2′′-terthiophene-functionalized metal-tetraazamacrocycle complexes, and electrochemical and spectroelectrochemical characterization of the resulting polythiophenes

Article abstract of DOI:10.1039/b103284b

We have prepared novel tetraazacyclotetradecane derivatives, functionalised at nitrogen with a pendant thiophene or oligothiophene group, for electropolymerisation. Square planar Ni(II) complexes of these ligands, [Ni(L)](CIO₄)₂, hav

Full text of DOI:10.1039/b103284b

View Article Online / Journal Homepage / Table of Contents for this issue
Syntheses and electro(co)polymerization of novel thiophene- and
2,2’:5’,2@-terthiophene-functionalized metal–tetraazamacrocycle
complexes, and electrochemical and spectroelectrochemical
characterization of the resulting polythiophenes
Simon J. Higgins,a Thomas J. Poundsa and Paul A. Christensenb
aDepartment of Chemistry, University of Liverpool, Crown Street, Liverpool, UK L69 7ZD
bDepartment of Chemistry, University of Newcastle upon Tyne, Newcastle, UK NE1 7RU
Received 11th April 2001, Accepted 12th June 2001
First published as an Advance Article on the web 17th July 2001
We have prepared novel tetraazacyclotetradecane derivatives, functionalised at nitrogen with a pendant
thiophene or oligothiophene group, for electropolymerisation. Square planar Ni(II) complexes of these ligands,
[Ni(L)](ClO4)2, have been prepared, and their electropolymerisation has been examined. The terthiophene-
bearing complex was electrooxidised in CH3CN–0.2 M Et4NBF4 to give soluble oligomers, which are likely to
be the products of dimerisation (sexithiophenes). However, by employing a more positive potential limit, they
are electropolymerised, to afford polythiophene films on indium-doped tin oxide (ITO)-coated glass or platinum
electrodes, and the Ni(II)/Ni(III) wave is superimposed upon the polythiophene redox process. Attempts to
isolate films of the dimerised terthiophene by the oxidation of neutral [NiCl2(L)] complexes in less polar media
were unsuccessful. A polymer film could not be generated from the thiophene-bearing complex alone, even
though a ‘spacer’ of five carbon atoms was incorporated between the thiophene and the bulky metal centre.
However, copolymerisation with 3-methylthiophene successfully gave polythiophene copolymers incorporating
the nickel(II) complex. In situ FTIR reflectance spectroelectrochemistry shows that in the early stages of film
oxidation, the film behaves as a conventional polythiophene, but at higher potentials, the electronic band
attributable to the transition from the valence band to the first intergap state undergoes some very unusual
changes. Possible explanations for these are suggested.
polymer backbone by an alkyl ‘spacer’ of at least four
Introduction
carbon atoms to minimise this.1 We have therefore prepared
a thiophene-appended cyclam with a longer alkyl ‘spacer’, and
have examined the electropolymerisation of its square planar
Ni(II) complex.
The generation of conjugated polymer-modified electrodes
containing covalently-attached functional groups of various
kinds has attracted much attention, and issues in the design and
syntheses of these systems have recently been reviewed.1,2
Metal complexes and potential ligands have been particularly
attractive targets, and polythiophenes and polypyrroles bear-
ing crown ethers,3,4 redox-active pyridine and bipy (2,2’-
bipyridine) complexes,5,6 and ferrocenes,7,8 have all been well
studied. Potential applications in electrocatalysis and in sensing
drive this work. Recently, there has been a tendency to employ
metal complexes with free or potentially free coordination
sites,9 including metal–phosphine complexes.10a,10b
For applications, polypyrroles have the potential advantage
that they can be cycled in aqueous electrolytes. However, the
electroactivity of polypyrrole films bearing pendant metal
complexes is destroyed when, as is often the case, the redox
potential of the metal complex is positive of that of the
polypyrrole backbone.1 Therefore, the study of possible
interactions between a metal centre and a conjugated polymer
backbone is better addressed with polythiophenes. We have
been interested in the incorporation of the well-known
electrocatalyst [Ni(cyclam)]2z (cyclam~1,4,8,11-tetraazacyclo-
tetradecane) into poly-3-methylthiophene,11 and have used
FTIR spectroscopy to probe, in situ, the effect on the polymer
properties of the incorporation of the metal complex.12 Bulky
substituents have adverse effects upon important polythio-
phene properties such as mean conjugation length,13 and it is
preferable to separate substituents from the conjugated
Copolymerisation has the advantage that thiophene-func-
tionalised metal complexes are relatively straightforward to
synthesise, but the disadvantage that control over the copolymer
composition is not straightforward. Some control over the
monomer ratio in the electrogenerated polymer can be exerted
by varying the monomer ratio in the electrolyte solution.
However, although an idea of monomer ratio in the resulting
copolymers can be obtained by comparing charge passed for
the polymer redox process with that for the redox process(es) of
the metal complex,11 there is no control over, or means of
determining, whether the copolymer is random or block.
Ba¨uerle and co-workers previously showed that while
thiophene-functionalised crown ethers often cannot be electro-
polymerised because of the solubility of the electrogenerated
species, the corresponding 2,2’-bithiophene- and 2,2’:5’,2@-
terthiophene-functionalised crowns readily afforded polythio-
phene films containing covalently-attached crown ether
moieties.3,4 Therefore, an alternative method of avoiding
copolymerisation is to use an oligothiophene-appended metal
complex as monomer. We have therefore investigated the
synthesis of an oligothiophene-functionalised cyclam and its
Ni(II) complexes. In this paper, we report these syntheses, the
electro(co)polymerisation of the complexes to give polymer-
modified electrodes, and preliminary in situ FTIR spectro-
scopic data on one of the polymer films.
DOI: 10.1039/b103284b
J. Mater. Chem., 2001, 11, 2253–2261
This journal is # The Royal Society of Chemistry 2001
2253

View Article Online
C4HBr2RS), 4.59 (t, 1H, pyranyl -OCHO-), 3.92–3.44 (over-
lapping m, 4H, -CH2CH2O- and pyranyl CH2O), 2.82 (t, 2H,
J~6.7 Hz, -CH2CH2O-). MS (EI): m/z 370 ([Mz], 1), 268
([M2OTHPz], 27).
Experimental
General
Instrumentation and general methods were as previously
described,14 with the following exceptions and additions. Pet.
ether refers to the petroleum ether fraction boiling between 40–
60 uC. Fast atom bombardment mass spectra were recorded
either in house by Mr Allan Mills, or by the EPSRC Mass
Spectrometry Service at Swansea. Electrochemical experiments
were conducted as previously described.10a
3’-[2-(Tetrahydropyranyloxy)ethyl]-2,2’:5’,2@-terthiophene, 8.
A solution of 2-thienylmagnesium bromide (from 2-bromo-
thiophene [1.96 g, 12.0 mmol] and Mg [0.36 g, 15 mmol] in
diethyl ether [10 cm3]) was added dropwise via cannula to a
solution of
7 (1.75 g, 4.73 mmol) and [NiCl2(Ph2PCH2-
CH2CH2PPh2)] (0.13 g, 2 mol. %) at 0 uC. The mixture was
then refluxed for 24 h, before being cooled to 0 uC and
hydrolysed with 0.5 M HCl (10 cm3). The organic phase was
extracted with CH2Cl2 (3620 cm3) and the extracts were
washed with water (20 cm3) before being dried over Na2SO4
and evaporated to dryness. The crude product was purified by
repeated column chromatography (hexanes, then pet. ether–
ethyl acetate, 9 : 1). Yield 1.08 g, 61%. Elemental analysis, calc.
for C19H20O2S3: C, 60.61; H, 5.35. Found: C, 60.80; H, 5.44%.
Syntheses
1-[5-(3-Thienyl)pentyl]-1,4,8,11-tetraazacyclotetradecane, 3.
To a solution of 1,4,8,11-tetraazacyclotetradecane (cyclam)
(1.55 g, 7.76 mmol) in refluxing benzene (50 cm3) and pyridine
(15 cm3) was added a solution of 3-(v-bromopentyl)thio-
phene15 (0.9 g, 3.88 mmol) dropwise with stirring. After
being refluxed for a further 4 h, the reaction mixture was
allowed to cool and the solvent was removed under reduced
pressure. The residue was dissolved in 0.2 M aqueous NaOH
(60 cm3) and extracted with diethyl ether (4630 cm3). The
ether extracts were dried over MgSO4. The ether was
evaporated off to leave a waxy oil. Yield 0.73 g, 53%. Selected
1H NMR (CDCl3): d~7.18 (dd, 1H, H5), 6.74 (m, 2H, H4,
H2), 3.82 (t, 2H, CH2C4H3S). MS (EI): m/z 352 ([Mz], 1), 199
([M2C5H10C4H3Sz], 95).
1
Selected H NMR (CDCl3): d~7.31 and 7.20 (dd’s, 1H, H5
and H5@, J~5.25, 1.05 Hz), 7.18 and 7.15 (dd’s, 1H, H3 and
H3@, J~3.60, 1.20 Hz), 7.11 (s, 1H, H4’), 7.06 and 7.01 (dd’s,
1H, H4 and H4@, J~5.10, 3.60 Hz), 4.63 (t, 1H, -OCH). MS
(EI): m/z 376 ([Mz], 16), 274 ([M2OTHPz], 23).
3’-(2-Hydroxyethyl)-2,2’:5’,2@-terthiophene, 9. Compound 8
(1.0 g, 2.66 mmol) in THF (20 cm3) and 0.5 M HCl (10 cm3)
was gently refluxed with monitoring of the reaction by TLC
(pet. ether–ethyl acetate, 8 : 2). Reaction was complete in 2 h.
The volume of the solution was reduced and CH2Cl2 (50 cm3)
was added. The layers were separated, the organic layer was
washed with water and dried over MgSO4, then evaporated.
Flash chromatography (silica gel, pet. ether–ethyl acetate 8 : 2)
gave the product as a yellow oil which crystallised on standing.
Yield 0.71 g, 91%. Elemental analysis, calc. for C14H12OS3: C,
57.50; H, 4.14. Found: C, 57.42; H, 4.14%. 1H NMR (CDCl3):
d~7.29–6.97 (m, 7H, thienyl H), 3.87 (t, 2H, -CH2CH2OH),
2.98 (t, 2H, -CH2CH2OH). MS (EI): m/z 292 ([Mz], 87), 261
([M2CH2OHz], 100).
{1-[5-(3-Thienyl)pentyl]-1,4,8,11-tetraazacyclotetradecane}-
nickel(II) perchlorate, [Ni(3)](ClO4)2, 4. The ligand 3 (0.32 g,
0.653 mmol) in MeOH (20 cm3) was added to a solution of
[Ni(H2O)6](ClO4)2 (0.32 g, 0.87 mmol) in hot methanol
(30 cm3). After 15 min, the hot solution was filtered from a
little insoluble impurity, and allowed to cool. An oil separated.
This was decanted off and washed with diethyl ether (10 cm3),
whereupon it crystallised. The solid was filtered off and
recrystallised three times from hot MeOH–EtOH, then dried in
vacuo. Elemental analysis, calc. for C19H36Cl2N4NiO8S: C,
37.40; H, 5.95; N, 9.18. Found: C, 37.74; H, 6.12; N, 8.05%. MS
(FAB positive ion): m/z 510 ([M2ClO4z], 88), 409
([M2HClO42ClO4z], 100). E1/2 [Ni(II)/Ni(III)] z 0.70 V (vs.
Fc/Fcz).
Toluene-p-sulfonic acid 2-(2,2’:5’,2@-terthiophen-3’-yl)ethyl
ester, 10. Toluene-p-sulfonyl chloride (0.86 g, 4.5 mmol) was
dissolved in anhydrous pyridine (10 cm3), left for 10 min, then
cooled to 0 uC. To this solution was added 9 (1.0 g, 3.42 mmol)
in pyridine (10 cm3) dropwise with stirring. After a further 2 h
at 0 uC, ice (100 g) was added, the solution was extracted with
CH2Cl2, the organic phase was washed with 2 M HCl, then
water, dried over MgSO4 and evaporated in vacuo. The residue
was purified by flash chromatography (CH2Cl2 eluant). Yield
1.40 g, 70%. Elemental analysis, calc. for C21H18O3S4: C, 56.48;
H, 4.06. Found: C, 55.98; H, 3.72%. 1H NMR (CDCl3):
d~7.69 (m, 4H, Ph), 7.32–7.02 (m’s, 7H, thienyl H), 4.23 (t,
2H, -CH2CH2O-, J~6.87 Hz), 3.06 (t, 2H, -CH2CH2O-), 2.36
(s, 3H, -CH3). MS (EI): m/z 446 ([Mz], 76), 273 ([M2OTsz],
100).
2-(2,5-Dibromo-3-thienyl)ethanol, 6. In the absence of light,
and under dinitrogen, N-bromosuccinimide (8.00 g, 45 mmol)
in dry DMF (30 cm3) was added dropwise to a stirred solution
of 2-(3-thienyl)ethanol (2.5 g, 19.5 mmol) in DMF (35 cm3).
The reaction mixture was stirred for 3 h at 40 uC before being
poured onto crushed ice (100 g). The mixture was extracted
with diethyl ether (3640 cm3). The ether phase was washed
with water, dried (MgSO4) and treated with decolourising
charcoal (1 g). Removal of the solvent afforded the product as
a yellow oil. Yield 4.70 g, 85%. Elemental analysis, calc. for
C6H6Br2OS: C, 25.20; H, 2.11. Found: C, 25.17; H, 2.11%. 1H
NMR (CDCl3): d~6.80 (s, 1H, C4HBr2RS), 3.78, 2.78 (t’s, 2H
each, J~6.5 Hz, -CH2CH2-), 2.05 (s, 1H, -OH). MS (EI): m/z
286 ([Mz], 47), 255 ([M2CH2OHz], 100).
1-[2-(2,2’:5’,2@-Terthiophen-3’-yl)ethyl]-1,4,8,11-tetraaza-
cyclotetradecane, 11. Prepared as for 3, using compound 10
(0.68 g, 1.5 mmol) and cyclam (0.74 g, 3.7 mmol). Crude yield
0.41 g, 59%, used directly to make 12. 1H NMR (CDCl3):
d~7.24–6.94 (m’s, 7H, thienyl H), 2.88–2.49 (overlapping
m’s, 23H, cyclam ring and ethyl CH2’s), 1.64 (m, 4H,
-CH2CH2CH2-). MS (EI): m/z 474 ([Mz], 2).
2-[2-(2,5-Dibromo-3-thienyl)ethoxy]tetrahydropyran,
7. A
solution of 6 (2.00 g, 7.27 mmol) in CH2Cl2 (50 cm3), contain-
ing pyridinium toluene-p-sulfonate (0.18 g, 0.70 mmol) and
dihydropyran (0.91 g, 10.85 mmol), was stirred for 36 h under
dinitrogen. The reaction mixture was diluted with diethyl ether,
washed with 50 : 50 sat. brine–water, and dried over Na2SO4.
Removal of the solvent in vacuo yielded the crude product as a
yellow oil. Column chromatography (petroleum ether–ethyl
acetate, 9 : 1) gave pure 7. Yield 1.95 g, 73%. Elemental
analysis, calc. for C11H16Br2O2S: C, 35.69; H, 3.81. Found:
C, 35.80; H, 3.84%. Selected 1H NMR (CDCl3): d~6.88 (s, 1H,
[Ni(11)](ClO4)2, 12. Prepared from 11 (1.00 g, 2.11 mmol)
and [Ni(H2O)6](ClO4)2 (0.91 g, 2.50 mmol) using the method
described for complex 4. Yield, after three recrystallisations
2254
J. Mater. Chem., 2001, 11, 2253–2261

View Article Online
from hot MeOH–EtOH, 0.68 g, 44%. Elemental analysis, calc.
for C24H34Cl2N4NiO8S3: C, 39.36; H, 4.68, N~7.65. Found: C,
39.34; H, 4.76, N~7.37%. MS (FAB): m/z 631 ([M2ClO4z],
76), 531 ([M22ClO42Hz], 100).
use 20 mM 4 and 50 mM 3-methylthiophene. The electrolyte
employed was 0.2 M Et4NBF4–CH3CN. Polymer films were
grown with the working electrode pulled back from the CaF2
window. The growth solution was then flushed from the cell
with fresh electrolyte and the film was cycled between 20.25
and z0.75 V until no further changes occurred (ca. 3 cycles). It
was then held at z0.2 V for several minutes before being
secured against the CaF2 window for FTIR experiments.
[Ni(11)Cl2], 13. The ligand 11 (0.80 g, 1.68 mmol) was
dissolved in MeOH (20 cm3) and added dropwise to a hot
solution of [Ni(H2O)6]Cl2 (0.45 g, 1.90 mmol) in MeOH
(20 cm3). The mixture was stirred and refluxed for 15 min.
The solution was filtered hot, and evaporated to dryness. The
residue was extracted with CH2Cl2. Removal of the CH2Cl2 in
vacuo gave the green product, which was recrystallised once
from CH2Cl2–diethyl ether. Yield 0.42 g, 41.5%. Elemental
analysis, calc. for C24H34Cl2N4NiS3: C, 47.70; H, 5.67, N, 9.27.
Found: C, 47.27; H, 5.26, N, 8.83%. MS (FAB): m/z 567
([M2Clz]), 531 ([M22Cl2Hz]). MS (FAB, accurate mass):
Found, 567.104224, calcd. for ([M2Clz]) 567.098763.
Results and discussion
Syntheses
Treatment of excess 1,4,8,11-tetraazacyclotetradecane (‘cyclam’)
with a suitable electrophile, followed by selective extraction of
the less-polar functionalised derivative from aqueous base with
a solvent such as diethyl ether, is a reliable and rapid method for
mono-functionalising cyclam.17 The expensive excess cyclam
can be recycled by subsequent extraction with dichloro-
methane. The required 5-(3-thienyl)-1-bromopentane, 2, was
made from 3-bromothiophene using the method developed by
Ba¨uerle.15 Treatment of excess cyclam with 2, followed by
extraction from aqueous base using diethyl ether, afforded
crude 1-[5-(3-thienyl)pentyl]-1,4,8,11-tetraazacyclotetradecane
3 as a waxy solid. Mass spectra of the crude product suggested
the presence of small amounts of unfunctionalised and di-
functionalised cyclam impurities in addition to 3. Treatment of
[Ni(H2O)6](ClO4)2 in MeOH with crude 3 gave [Ni(3)](ClO4)2,
4, which was recrystallised from hot MeOH–EtOH.{ As with
the ligand in complex 1, which was similarly prepared by
reacting excess cyclam with the toluene-p-sulfonate of com-
mercially-available 2-(3-thienyl)ethanol,11 we found that it was
more satisfactory to use crude 3 to prepare the complex 4, and
subsequently to purify 4 by recrystallisation, than it was to
purify 3.
Electrochemical polymer syntheses and characterisation
Electrochemical copolymerisation of 4 with 3-methylthio-
phene. A Pt disc (0.3 cm2) or ITO-coated glass (2 cm2) working
electrode was employed, in a standard 3-electrode cell with Pt
gauze counter electrode and saturated calomel reference
electrode separated from the working compartment by a salt
bridge. The stability of the reference electrode was checked by
periodically examining the ferrocene/ferrocinium redox pro-
cess, which occurred at z0.45 V in this set-up. A solution of
complex 4 (0.012 M) and 3-methylthiophene (0.025 M) in
0.2 M Et4NBF4–CH3CN electrolyte was employed. The poten-
tial of the working electrode was cycled repeatedly, initially
between 20.45 and z1.35 V (1 scan; z1.55 V in in situ FTIR
experiments), then subsequently to z1.25 V, at 0.1 V s21
.
Electrochemical polymerisation of complex 12. A 5 mM
solution of complex 12 in 0.2 M Et4NBF4–CH3CN electrolyte
was employed for the electropolymerisation experiments.
The onset of terthiophene oxidation occurred at ca.
z0.75 V. Adherent, redox-active films were grown by repeti-
tive cycling between 20.35 and z1.05 V. Similar experiments
were also performed using 0.2 M Bu4NBF4–CH2Cl2 electro-
lyte.
Electrochemical polymerisation of complex 13. A 5 mM
solution of complex 13 in 0.2 M Bu4NBF4–CH2Cl2 electrolyte
was employed for the electropolymerisation experiments. The
onset of oxidation occurred at ca. z0.45 V. Adherent, redox-
active films were grown by repetitive cycling between 20.35
and z0.95 V.
Treatment of 2-(3-thienyl)ethanol, 5, with two equiv. of NBS
gave 2-(2,5-dibromo-3-thienyl)ethanol 6. This was converted to
the tetrahydropyranyl (THP) derivative 7. Although protection
as the THP derivative followed by bromination was unsatis-
factory, and other protection strategies failed, conversion of 5
to 7 was straightforward. Nickel-catalysed cross-coupling of 7
with two equiv. of 2-thienylmagnesium bromide ([NiCl2(dppp)],
Et2O) afforded 8, in only moderate yield following the
extensive column chromatography necessary for its purifica-
tion. Although we did not attempt to identify byproducts,
related coupling reactions between 3-alkyl-2,5-dibromothio-
phene and 2-thienylmagnesium bromide give significant
amounts of 2,2’-bithiophene, 2,2’:3’,2@-terthiophene and 5’-(2-
thienyl)-2,2’:3’,2@-terthiophene,18 and similar side-reactions
may occur during the preparation of 8. Compound 8 was
deprotected by treatment with acid in aqueous THF to give the
free alcohol 9, and this was converted to the toluene-p-
sulfonate 10. Reaction of 10 with excess cyclam gave 11. Once
again, crude 11 contained trace amounts of unfunctionalised
Polymer cyclic voltammetry
Modified electrodes were removed from the growth solution
with the film in the reduced form, washed with the appropriate
pure solvent, and stored in dust-free conditions prior to redox
cycling experiments, which were conducted in fresh back-
ground electrolyte (0.2 M Et4NBF4–CH3CN) as described in
the Results section.
Infrared spectroscopy
The electrochemical cell and FTIR instrumentation, and
general methods, were as described recently.16 All electrolyte
solutions were prepared under Schlenk conditions in flame-
dried glassware, and were injected into the cell under nitrogen.
The reference electrode (Ag/AgCl) was separated from the
sample compartment by a salt bridge containing the same
electrolyte as used in the sample compartment. To grow
copolymer films of 4 with 3-methylthiophene in such a way that
the highly polished bulk Pt working electrode was uniformly
coated for reflectance measurements, it was found necessary to
{WARNING! Perchlorate salts are highly oxidising and potentially
explosive. No accidents were encountered with the compounds in this
study, but perchlorate salts should be isolated in small quantities, and
should never be scratched or subjected to shock.
J. Mater. Chem., 2001, 11, 2253–2261
2255

View Article Online
and difunctionalised cyclams, and was used directly to prepare
the complex [Ni(11)](ClO4)2, 12, which was purified by
recrystallisation from hot MeOH. Crude 11 was also used to
prepare the neutral complex [Ni(11)Cl2] (13) by treatment with
[Ni(H2O)6]Cl2 in MeOH, followed by removal of the solvent
and extraction into hot CH2Cl2. The electronic spectrum of a
CH2Cl2 solution of 13 was consistent with six-coordinate Ni(II).
Fig. 1 Growth of a 4–3-methylthiophene copolymer by repetitive cyclic
voltammetry at a Pt disk working electrode. The arrows on the first
scan illustrate the nucleation loop seen on the first scan, indicative of
the formation of a conducting phase. Solution composition: 0.2 M
Et4NBF4–CH3CN, 12 mM 4, 25 mM 3-methylthiophene. Scan rate
0.1 V s21. Potentials vs. internal ferrocene/ferrocinium.
Electropolymerisation of complex 4
A solution of complex 4 (5 mM) in 0.2 M Et4NBF4–CH3CN
showed a chemically reversible one-electron oxidation wave at
z0.70 V (vs. internal ferrocene/ferrocinium) due to the Ni(III)/
Ni(II) couple. As observed for other square planar nickel(II)
complexes of monofunctionalised cyclams,11 this is slightly
more positive than the redox potential for [Ni(cyclam)]2z
itself, as tertiary amines are significantly weaker donors than
secondary amines, destabilising the higher oxidation state. The
positive potential limit of consecutive scans was then increased
in 0.1 V steps until the onset of a further, irreversible, oxidation
was seen commencing at z1.4 V. On repeated cycling between
0.00 and z1.45 V, there was no evidence for the formation
of a conducting polymer film. Varying the concentration
(1–20 mM), electrolyte concentration (0.05–0.4 M), working
electrode (Pt disk, Au disk, ITO-coated glass) or positive limit
(up to z1.6 V) also failed to yield a polymer film. As with
the electro-oxidation of 1 (perchlorate salt),11 dark blue
material was seen in solution around the working electrode
at wz1.45 V, indicating that oxidised oligothiophenes were
being generated, but that these were soluble. Clearly, the effect
of the longer ‘spacer’ on the electropolymerisation of 4
compared with 1 is negligible.
We therefore studied co-polymerisation with 3-methylthio-
phene. Fig. 1 shows a typical cyclic voltammogram for the
initial stage of the growth of a copolymer of 3-methylthiophene
(25 mM) and complex 4 (12 mM) on a Pt disk working
electrode. The peak currents for both the polythiophene and
Ni(III)/Ni(II) redox waves increase with cycle number, suggest-
ing the growth of a copolymer film. Copolymer films were
grown on both Pt disk and ITO electrodes. After film growth,
the electrodes were removed from the solution with the poly-
thiophene film in its insulating form, washed with acetonitrile,
Fig. 2 Cyclic voltammograms of the 4–3-methylthiophene copolymer,
the early stages of whose growth is shown in Fig. 1, recorded in 0.2 M
Et4NBF4–CH3CN. In order of increasing peak currents, scan rates
were 0.025, 0.05, 0.075, 0.10, 0.15 V s21
.
and transferred to a monomer-free electrolyte solution. Fig. 2
shows a typical cyclic voltammogram for such a film, on Pt. As
expected for a surface-localised redox process, peak current
varies linearly with scan rate.19 Peak-to-peak separation for the
Ni(III)/Ni(II) redox wave (Pt) are small, but non-zero, and
increase with scan rate (0.02 V at 25 mV s21; 0.09 V at
150 mV s21), presumably because of kinetic limitations to the
movement of charge-balancing anions and associated
solvent.20 As expected for a polythiophene, the films were
electrochromic, being red in the reduced state and dark blue
when oxidised.
We conducted an experiment to check that the nickel
complex is not simply physically trapped within the growing
polymer film. A film of poly-3-methylthiophene was grown by
potential cycling on Pt under the same conditions as used for
the copolymer, using 35 mM 3-methylthiophene, but in the
21
presence of the complex {Ni[1-(n-octyl)cyclam]}(BF4)2
(5 mM). This complex was chosen to have similar charge and
solubility properties to 4, but without a polymerisable 3-thienyl
group. No Ni(III)/Ni(II) wave could be detected in the cyclic
2256
J. Mater. Chem., 2001, 11, 2253–2261

View Article Online
solution), the degree of doping, the nature of the substituents
(if any), etc.
In the spectral region accessible in our experiments (1000–
6000 cm21), a strong electronic band is generally observed on
oxidation of a conducting polymer film, owing to the lowest-
energy transition (from the valence band to the lowest inter-gap
level; transition 1 in Fig. 3) regardless of whether the charge
carrier is a polaron or bipolaron. In addition, the movement of
charge carriers along polymer chains gives rise to large dipole
changes in those vibrations that can couple to the carrier
motion. The corresponding infrared bands are selectively
enhanced, such that difference spectra such as those in Fig. 4
show only the gain of these very intense infrared activated
vibrations (IRAV’s).
Earlier, we studied the copolymer of 1 with 3-methylthio-
phene, and compared it with poly-3-methylthiophene.12 The
evolution of the spectra as a function of potential for poly-3-
methylthiophene could be interpreted in terms of polaron
formation early in the oxidation, then co-existence of polarons
and bipolarons, and finally, bipolaron formation at the most
positive potentials. In contrast, for the copolymer, while
spectra early in the oxidation were little different from those of
poly-3-methylthiophene, at higher potentials, different beha-
viour was observed. In particular, the successive increases in
the electronic band intensity, which occurred at constant energy
(ca. 5000 cm21) in the early stages of oxidation, appeared to
shift to progressively lower energy thereafter. However, this
was hard to quantify owing to the very low throughput at ca.
4500 cm21 of the BioRad FTS-40 spectrometer employed.
There also appeared to be a slight loss in intensity of the
electronic band at high energy at the most positive potentials.
In the current study, we employed an FTS-60 spectrometer,
which does not have the throughput limitations of the FTS-40.
Moreover, we carefully excluded moisture from the working
compartment by using Schlenk techniques and employing a salt
bridge to separate the reference electrode from the working
compartment, and this has enabled us to access somewhat more
positive potentials without degrading the polymer.
Scheme 1
voltammogram of the resulting polymer film in background
electrolyte.
As noted earlier for copolymers of [Ni(1)]2z and 3-methyl-
thiophene, the significance of the fact that the Ni(III)/Ni(II)
wave occurs at almost the same potential in the polymer film as
for the monomer in solution is that there must therefore be
sufficient acetonitrile in the polymer film to allow the chemistry
in Scheme 1 to occur unchanged.
The unusually symmetrical and highly reversible polythio-
phene redox chemistry seen is a consequence of the presence of
the high ionic strength in the film, even for the undoped form of
the polythiophene, owing to the presence of the anchored
[Ni(cyclam)]2z moieties. This effect has been noted before, for
polypyrrole bearing covalently-tethered pyridinium ions.22
Infrared spectroscopy of 4–3-methylthiophene copolymers in situ
An intense p–p* transition dominates the visible region of the
electronic spectrum for the neutral form of polythiophenes
(Fig. 3). On removal of an electron from the HOMO of a
polythiophene chain, the new p SOMO moves up in energy,
and the corresponding p* orbital moves down in energy, into
the gap.23 There has been much controversy over whether the
removal of further electrons results in the generation of further
SOMO’s (polaron carriers), or whether spinless dicationic
charge carriers (bipolarons) are formed, in which case the
intragap levels are both empty.24–28 More recently, compelling
evidence from oligomer studies has shown that it is also
possible for polarons on neighboring chains to form p-dimers,
which are also spinless.29 This may be more likely to occur in
regioregular polyalkylthiophene films, in which the polymer
chains can self-assemble into ordered layers.30 Thus, the
situation is complicated. Theoretical results indicate that the
energy difference between polaron and bipolaron carriers is
quite small.31 Therefore, the preferred carrier for a given
polythiophene (polaron, bipolaron, polaron p-dimer) may
depend upon criteria such as the origin of the polymer
(e.g. whether it was generated chemically or electrochemically
by oxidative coupling, or by organometallic aryl–aryl coupl-
ing), how the carrier is generated (e.g. electrochemical or
chemical oxidation; photoexcitation), the phase (solid or
A copolymer of 4 (perchlorate salt) and 3-methylthiophene
was grown on a polished bulk Pt electrode (area 0.64 cm2) for
the FTIR experiments. We found that it was necessary to
use a solution of 20 mM 4 and 50 mM 3-methylthiophene to
promote the required even coverage of this electrode with the
copolymer. However, cyclic voltammograms in background
electrolyte established that the copolymer composition was
similar to that observed in the earlier experiments described
above. The polymer-coated working electrode was then held at
20.4 V, and a reference spectrum was collected at this
potential. The potential was then increased in 0.1 V intervals
up to z0.8 V, and then at 0.1 V intervals down to 20.4 V once
more, and spectra were collected at each potential.
The spectral changes seen on progressively oxidising the
4–3-methylthiophene copolymer film are shown in Fig. 4 (a–d).
Fig. 4a shows all these spectra collected up to z0.8 V over the
full spectral range. It is clear from Fig. 4a that these changes
are not uniform as a function of potential. Fig. 4b shows the
spectra taken from 20.3 to 0.1 V with respect to the spectrum
taken at 20.4 V, Fig. 4c shows the spectra taken at 0.2 and
0.3 V with respect to the spectrum taken at 0.1 V, and Fig. 4d
shows the spectra taken from 0.4 to 0.8 V with respect to the
spectrum taken at 0.3 V.
During the first stage of oxidation distinguishable in the IR
spectra, from 20.4 to z0.1 V, ca. 15% of the total oxidative
charge is passed, calculated from the charge passed in the
course of FTIR data collection. This region is associated with
the pre-peak in the cyclic voltammogram seen in Fig. 2. It is
generally accepted, even by workers who hold that bipolarons
are the main charge carrier in polythiophenes, that polarons
can form at low doping levels,13 and we therefore propose that
early in film oxidation, polarons are forming. The IRAV bands
Fig. 3 Band diagram for polythiophene, showing the p–p* transition
for the neutral polymer (left), the effect of removing an electron to form
a polaron (centre) together with the resulting additional electronic
transitions (1–3), and the formation of a bipolaron with its resulting
additional electronic transitions (right). Note that in the polaron,
transitions labelled 2 are expected at equal energy by electron-hole
symmetry, and that for the bipolaron, transition 3 is absent.
J. Mater. Chem., 2001, 11, 2253–2261
2257

View Article Online
Fig. 4 (a) In situ FTIR spectra (8 cm21 resolution, 100 co-added and averaged scans per spectrum), collected from a 0.64 cm2 Pt electrode coated
with the 4–3-methylthiophene copolymer (see text for details), and immersed in 0.2 M Et4NBF4–CH3CN. The spectra were collected in an
experiment in which the potential of the coated electrode was increased in 100 mV steps from (i) 20.3 V to (xi) z0.8 V with a spectrum taken at each
step and normalised to the reference taken at 20.4 V. The spectra are displayed as ‘Abs’ vs. cm21, where Abs~log10 (S[n]/S[20.4 V]). (b) The spectra
collected at (i) 20.3 V to (v) z0.1 V in the experiment depicted in (a). (c) The spectra collected at (i) z0.2 V and (ii) z0.3 V in (a) normalised to that
taken at z0.1 V. (d) The spectra taken at (i) z0.4 V to (iv) z0.8 V in (a) normalised to that taken at z0.3 V; the spectrum collected at z0.7 V is not
shown, for clarity.
seen for the copolymer (Fig. 4b; 1167, 1203 sh, 1308, 1375–1392
and 1440 sh cm21) are very similar in position and relative
intensities to those seen earlier for polythiophene32 and poly-3-
methylthiophene itself in this potential region,12,33 and a gain is
seen at 1050 cm21 which is assigned to tetrafluoroborate
anions. The electronic band at ¢5000 cm21 is attributed to the
transition from the valence band to the lowest intergap level
(transition 1 in Fig. 3).
In the spectra taken at z0.2 and z0.3 V, referenced to that
taken at z0.1 V, there is clearly a transition in behaviour
(Fig. 4c). Further intensity increases across the electronic band
occur, but the peak of these increases moves to lower energy.
There are also significant changes in the IRAV bands seen in
this region, which suggest the evolution of a different type of
carrier. However, the absence of significant loss features in the
IRAV region suggests that the carriers generated during the
first stage of oxidation are not lost here.
Examining the spectra taken at wz0.4 V, referenced to that
taken at z0.3 V, the gains in the electronic band continue to
move to lower energy, while there is now significant loss of
intensity at higher energy. Furthermore, there is now definite
evidence for loss of the carriers generated in the early stage of
oxidation; clear loss features are seen at 1395–1375 cm21, and
in addition at 1310 cm21 and 1172 cm21, although the latter
two features are less certain because of the proximity of other
gain peaks. These loss features coincide with the onset of the
Ni(II)/Ni(III) process within the film, and they are seen much
more clearly here than in our experiments with copolymers of 1
and 3-methylthiophene.12 Although the potential limit used in
these experiments is somewhat more positive than that
commonly used for polythiophene films,2,13 it is important to
emphasise that the losses are not due to some irreversible
degradation process, since on stepping down progressively
from z0.8 V, the low-energy electronic band is lost and the
high energy band intensity is re-gained. The IRAV band
changes are also reversible. This process could be repeated
several times with no significant irreversible spectroscopic
changes.
It is difficult to be certain of the origin of the behaviour of the
electronic bands, which is unlike anything observed previously.
It is clear that the balance between charge carriers on
polyalkylthiophenes in solution can be substantially altered
by a small change of relative permittivity of the solvent,31 and
therefore it may be that relative permittivity change in the
copolymer film, as a result of accessing the Ni(II)/Ni(III) redox
process, may likewise alter this balance. Given that the
copolymer films are less conjugated than poly-3-methylthio-
phene itself, leading to more extensive early polaron formation,
it is possible that the loss feature seen at higher energy later in
oxidation might be due to bleaching of the polaron transition
involving both intergap levels (transition 3 in Fig. 3), which
could tail into the near-IR. The observed IRAV changes
support this possibility.
The new electronic band at low energy may be due to a
photoassisted electron transfer from Ni(II) to the lowest
interband state of the oxidised polythiophene. There is much
interest in photoinduced charge separation in neutral poly-
thiophenes containing electron acceptors, such as C60.2,34–36
The potential at which the new electronic band is first
apparent, z0.2 V, is some 0.45 V negative of the Ni(II)/
Ni(III) half-wave potential. This translates to 3640 cm21, close
to the maximum of the new band. The movement of the band
to lower energy as a function of increasing potential can then be
understood as due to the decreasing energy needed to promote
the electron from the Ni(II) to the increasingly oxidised
polythiophene. To test these ideas further, we plan to examine
2258
J. Mater. Chem., 2001, 11, 2253–2261

View Article Online
related polymers using similar Ni(II) complexes with less
positive redox potentials.
Electrochemistry of terthiophene-functionalised complexes 12
and 13
Before outlining the electrochemistry of 12 and 13, it is worth
mentioning briefly the electropolymerisation of terthiophenes
which do not have a cationic substituent, for example 2,2’:5’,2@-
terthiophene itself, 14,37 3’-n-hexyl-2,2’:5’,2@-terthiophene,
15,14 and 3,3@-di-n-octyl-2,2’:5’,2@-terthiophene 16.38 In 0.2 M
tetraethylammonium tetrafluoroborate (TEAT)–acetonitrile,
the onset of oxidation of 15 occurs at ca. z0.35 V, and on
cycling between 20.45 V and z0.55 V, a well-defined, quite
sharp, chemically reversible redox system grows with cycle
number, at z0.26 V. This was assigned to the reversible
oxidation of a film consisting mainly of the electrodeposited
sexithiophene formed by a,a’-dimerisation of 15 on formation
of its radical cation, by analogy with the work of Henderson
and Collard on 16 which behaves similarly.38 To electrodeposit
a redox-active, adherent polythiophene using 15 as monomer,
it was necessary to employ a more positive potential limit
(z0.85 V), probably because at this potential, formation of the
terthiophene dication begins, which undergoes a much more
rapid irreversible polymerisation reaction.14,38
Fig. 5 Cyclic voltammograms of a film of poly-12 (grown from a
solution of 5 mM 12 in 0.2 M Et4NBF4–CH3CN) on an ITO-coated
glass electrode, in 0.2 M Et4NBF4–CH3CN. Scan rates, in order of
increasing peak currents, were 0.01, 0.02, 0.03 and 0.04 V s21
.
Turning now to the oxidation of the terthiophene-bearing
complex 12, on the first oxidative sweep, the onset of oxidation
was at z0.55 V (Pt electrode), some 0.2 V more positive than
for 15.14 When the positive limit was set just beyond this
potential, intensely blue material was seen in solution around
the electrode surface, and a chemically reversible redox wave
developed at z0.41 V, but after 12 cycles the peak currents for
this process reached constant values, and no polymer layer
formed. The evolving redox process at z0.41 V suggests that
dimerisation of the electrogenerated terthiophene cation
radical is occurring, and that a sexithiophene product (two
possible regioisomers) analogous to that generated in similar
experiments with 1514,38 is formed. However, in this case the
sexithiophenes are soluble in the electrolyte, giving the intense
blue colour in their oxidised form. The redox process at
z0.41 V is therefore due to oxidation of sexithiophenes to the
radical cation, made more positive than for 15 by the close
proximity of the cationic Ni(II) centres. Although the oxidation
process at wz0.55 V in these experiments is expected to
overlap the Ni(II)/Ni(III) wave, this does not affect the outcome.
This is supported by the observation that when the non-
coordinating electrolyte CH2Cl2–0.2 M Bu4NBF4 was employed,
almost identical results were obtained, although in this
medium, the Ni(II)/Ni(III) redox wave would not be expected
until wz1.0 V.17
When the positive limit was increased to z1.05 V, we were
able to obtain an adherent and electrochromic polymer layer,
and new broad redox waves due to polythiophene redox
chemistry developed, a minor wave centred at z0.35 V and a
major wave, which overlapped the Ni(III)/Ni(II) system, at
z0.70 V. Films of varying thickness could be deposited by
repetitive redox cycling, on either ITO or Pt working
electrodes.
After growth experiments, the films were held at the negative
potential limit, then removed from the cell at open circuit,
washed with acetonitrile and transferred to fresh background
electrolyte. The voltammogram of a typical film, grown by 20
potential cycles between 20.25 and z1.05 V on an ITO-coated
glass electrode, is shown in Fig. 5. It was noticeable that some
material dissolved when the films were washed. A greater
proportion of the material appeared to dissolve for nominally
thicker films: when a film (grown using 20 redox cycles) was
cycled in background electrolyte after washing (Fig. 5), the
charge passed was only 2.9 times that for a nominally much
thinner film, grown using 5 cycles.
Interestingly, the voltammograms of these polyterthio-
phene films are remarkably symmetrical, and are very different
from the usual polyterthiophene voltammograms (see Fig. 4
of ref. 36).14,37,38 They are notably similar in profile to the CV’s
of the poly-4–poly-3-methylthiophene copolymer, except that
the onset of the polyterthiophene oxidation, z0.30 V, is more
positive than that for the copolymer oxidation (z0.15 V on
Pt), suggesting that the copolymer film contains at least some
polymer chains with more extensive conjugation. It is well-
established that polythiophene films grown using thiophene
oligomers often have shorter conjugation lengths than films
grown using thiophenes as monomers.38 The Ni(III)/Ni(II) redox
potential, at z 0.75 V, is ca. 0.1 V more positive than that
observed in solution and in the copolymer. We suggest that the
high ionic strength in poly-12 films regardless of the
polythiophene redox state, due to the presence of the cationic
metal complex, makes the redox processes more reversible than
those of the hydrophobic poly-14, -15 or -16, owing to easier
ion ingress and egress.
Because the thiophene : nickel stoichiometry is fixed at 3 : 1
by polymerising 12, we can use the voltammograms in Fig. 3
to estimate the ratio of Ni(II) to thiophene rings in
4–3-methylthiophene copolymers. Earlier, we estimated the
composition of 1–3-methylthiophene copolymers by integrat-
ing the charge under the Ni(II)/Ni(III) wave,11,12 and under the
‘polythiophene’ redox wave, and assuming that fully-oxidised
polythiophene chains have one positive charge per four
thiophene rings.13 This consistently gave a composition of
approximately one Ni(II) site per ten thiophene rings. If we
apply the same procedure to our poly-4–3-methylthiophene
films (Fig. 2), we find that Ni(II) : thiophene is slightly
smaller, 1 : 12. However, for poly-12 the same procedure
gives a composition of Ni(II) : thiophene 1 : 6, which is clearly a
serious underestimate. This suggests that we have also
previously underestimated the Ni(II) content of our copoly-
mers.11,12
The ‘polythiophene’ redox wave of the copolymers is clearly
different from that of poly-3-methylthiophene itself, and this
may invalidate simple extrapolation to estimate the charge
under the ‘polythiophene’ redox wave; the current due to
polythiophene oxidation may, in fact, fall as the Ni(II)/Ni(III)
J. Mater. Chem., 2001, 11, 2253–2261
2259

View Article Online
wave is accessed. Moreover, the assumption that one charge is
removed per four thiophene rings for fully oxidised poly-
thiophene films13 may be an underestimate at our positive
potential limit, which must be significantly more positive than
those normally employed for polythiophenes13 to allow us to
access the Ni(II)/Ni(III) redox wave. Wrighton et al. examined
the electrochemistry of some polythiophenes up to z1.8 V vs.
Ag wire quasi-reference in liquid SO2–Bu4NPF6 electrolyte
under anhydrous conditions, and found that they could be
oxidised to the extent of approximately 0.5–1.0 electrons per
thiophene ring at the most positive potential.39 Using these
results, we estimate that our polythiophene may, in fact, be
oxidised to the extent of up to 0.35 electrons per thiophene ring
at the positive limit.
Given the observations of Henderson et al. that it is possible
to deposit electrochemically dimerised terthiophene films,38 we
were interested in the possibility of electrodepositing the
soluble oligomers generated at low potentials in the above
experiments, by using a less polar electrolyte in which the
charged oligomer solubility might be smaller. Electropoly-
merisation of 12 in CH2Cl2–0.2 M tetrabutylammonium
tetrafluoroborate (TBAT) resulted in essentially the same
voltammetric behaviour as in acetonitrile.
transition as a function of potential in the copolymer, possibly
owing to charge transfer from the nickel(II) centre to the
oxidised polymer backbone.
Acknowledgements
We thank EPSRC for a project studentship (GR/J68625) and
equipment grants, and for access to the Mass Spectrometry
service at Swansea, and Professor Peter Pickup (Memorial
University of Newfoundland) for helpful discussions.
References
1
2
3
4
5
S. J. Higgins, Chem. Soc. Rev., 1997, 26, 247.
J. Roncali, J. Mater. Chem., 1999, 9, 1875.
S. Scheib and P. Ba¨uerle, J. Mater. Chem., 1999, 9, 2139.
P. Ba¨uerle and S. Scheib, Adv. Mater., 1993, 5, 848.
C. Lopez, J. C. Moutet and E. Saint-Aman, J. Chem. Soc.,
Faraday Trans., 1996, 92, 1527.
6
7
8
9
D. Curran, J. Grimshaw and S. D. Perera, Chem. Soc. Rev., 1992,
20, 391.
T. Inagaki, M. Hunter, X. Q. Yang, T. A. Skotheim and
Y. Okamoto, J. Chem. Soc., Chem. Commun., 1988, 126.
G. Zotti, S. Zecchin, G. Schiavon, A. Berlin, G. Pagani and
A. Canavesi, Chem. Mater., 1995, 7, 2309.
We also prepared the neutral, pseudo-octahedral trans-
[Ni(11)Cl2] (13), with the idea that it might be possible to ‘tune’
the Ni(II)/Ni(III) potential by controlling the coordination
chemistry at Ni(II); a related complex (CH2Cl2–0.2 M Bu4NCl)
shows the Ni(II)/Ni(III) process at z0.175 V.17 Moreover, we
hoped that by using a neutral complex, we could move to a less
polar solvent, and so isolate the terthiophene dimers in film
form.38
H. Cano-Yelo Bettega, J.-C. Moutet and S. Tingry, J. Electroanal.
Chem., 1995, 391, 51.
10 (a) S. J. Higgins, H. L. Jones, M. K. McCart and T. J. Pounds,
Chem. Commun., 1997, 1907; (b) T. B. Higgins and C. A. Mirkin,
Chem. Mater., 1998, 10, 1589.
11 G. King, S. J. Higgins and N. Price, Analyst (London), 1992, 117,
1243.
However, the fact that we wished to electro-oxidise the
terthiophene unit meant that we were unable to use Bu4NCl as
the electrolyte salt.40 When we investigated the electrochem-
istry of 13 in CH2Cl2–0.2 M TBAT, we found that the onset of
oxidation did not occur until z0.30 V. Moreover, when the
dark green solid 13 was dissolved in CH2Cl2–0.2 M TBAT, it
gave an orange–yellow solution, the electronic spectrum of
12 P. A. Christensen, A. Hamnett and S. J. Higgins, J. Chem. Soc.,
Faraday Trans., 1996, 92, 773.
13 J. Roncali, Chem. Rev., 1992, 92, 711.
14 C. L. Jones and S. J. Higgins, J. Mater. Chem., 1999, 9, 865.
15 P. Ba¨uerle, F. Wurthner and S. Heid, Angew. Chem., Int. Ed. Engl.,
1990, 29, 419.
16 M. Vilas-Boas, C. Freire, B. d. Castro, P. A. Christensen and
A. R. Hillman, Chem. Eur. J., 2001, 7, 139.
17 M. Di Casa, L. Fabbrizzi, M. Mariani and B. Seghi, J. Chem. Soc.,
Dalton Trans., 1990, 55.
18 P. A. Chaloner, S. R. Gunatunga and P. B. Hitchcock, J. Chem.
Soc., Perkin Trans. 2, 1997, 1597.
19 A. C. Fisher, in Electrode Dynamics, ed. R. G. Compton, Oxford
University Press, Oxford, 1996.
20 M. E. G. Lyons, Adv. Chem. Phys., 1996, 94, 297.
21 G. D. Santis, M. D. Casa, M. Martani, B. Seghi and L. Fabbrizzi,
J. Am. Chem. Soc., 1989, 111, 2422.
22 H. Mao and P. G. Pickup, J. Electroanal. Chem., 1989, 265, 127.
23 H. E. Schaffer and A. J. Heeger, Solid State Comm., 1986, 59, 415.
24 M. J. Nowak, D. Spiegel, S. Hotta, A. J. Heeger and P. A. Pincus,
Macromolecules, 1989, 22, 2917.
25 Z. W. Sun and A. J. Frank, J. Chem. Phys., 1991, 94, 4600.
26 Y. Furukawa, J. Phys. Chem., 1996, 100, 15644.
27 P. A. Christensen, A. Hamnett, A. R. Hillman, M. J. Swann and
S. J. Higgins, J. Chem. Soc., Faraday Trans., 1993, 89, 921.
28 P. A. Christensen, A. Hamnett, A. R. Hillman, M. J. Swann and
S. J. Higgins, J. Chem. Soc., Faraday Trans., 1992, 88, 595.
29 D. D. Graf, R. G. Duan, J. P. Campbell, L. L. Miller and
K. R. Mann, J. Am. Chem. Soc., 1997, 119, 5888.
30 H. Sirringhaus, P. J. Brown, R. H. Friend, M. M. Nielsen,
K. Bechgaard, M. M. W. Langeweld-Voss, A. J. H. Spiering,
R. A. J. Janssen, E. W. Meijer, P. Herwig and D. M. D. Leeuw,
Nature, 1999, 401, 685.
which was consistent with the square planar form, [Ni(12)]2z
.
In cases where the anion is moderately coordinating, such as
for chloride, there is an equilibrium between octahedral,
paramagnetic [Ni(L)X2] and square planar, diamagnetic
[Ni(L)]X2, and high ionic strength drives this to the square
planar form.17 Assuming that the large excess of tetrafluoro-
borate anions leads to metathesis from [Ni(12)]Cl2 to
[Ni(12)](BF4)2, this irreversible oxidation must be due to the
generation of the terthiophene radical cation, since the Ni(III)/
Ni(II) redox potential for [Ni(12)](BF4)2 would not be expected
until wz1 V in CH2Cl2.17 Thus, our efforts were defeated by
solution equilibria involving the coordination geometry of the
Ni(II) site. Once more, dark blue material was seen streaming
away from the working electrode surface positive of z0.3 V,
and no polymer film was deposited.
When a higher potential limit (z0.9 V) was employed, an
electrochromic, redox-active polymer film was deposited. The
electrochemistry of these films, cycled in 0.2 M TEAT–CH3CN
as background electrolyte, was essentially indistinguishable
from that of films grown using complex 12 in 0.2 M TEAT–
CH3CN. Interestingly, the presence of chloride ions in the
electrodeposition solution does not appear adversely to affect
the polymer formation.
31 M. J. Nowak, D. Spiegel, S. Hotta, A. J. Heeger and P. A. Pincus,
Macromolecules, 1989, 22, 2917.
In summary, we have shown that the short alkyl ‘spacer’
between the thiophene and the bulky metal complex substituent
in 1 is not a significant factor in preventing its electrochemical
homopolymerisation, since electropolymerisation of 4 did not
afford homopolymer films, whereas electropolymerisation of
12 and 13 was successful. Copolymerisation of 4 with 3-methyl-
thiophene gave polythiophene films bearing covalently-
attached [Ni(cyclam)]2z units. In situ infrared spectroscopy
reveals unusual evolution of the lowest energy electronic
32 P. A. Christensen, A. Hamnett and D. C. Read, Synth. Met., 1994,
62, 141.
33 E. Lankinen, G. Sundholm, P. Talonen, T. Laitinen and T. Saario,
J. Electroanal. Chem., 1998, 447, 135.
34 N. S. Sariciftci, L. Smilowitz, A. J. Heeger and F. Wudl, Science,
1992, 258, 1474.
35 T. Benincori, E. Brenna, F. Sannicolo, L. Trimarco, G. Zotti and
P. Sozzani, Angew. Chem., Int. Ed. Engl., 1996, 35, 648.
36 J. P. Ferraris, A. Yassar, D. C. Loveday and M. Hmyene, Opt.
Mater., 1998, 9, 34.
2260
J. Mater. Chem., 2001, 11, 2253–2261

View Article Online
37 H. L. Bandey, P. Cremins, S. E. Garner, A. R. Hillman,
J. B. Raynor and A. D. Workman, J. Electrochem. Soc., 1995,
142, 2111.
39 D. Ofer, R. M. Crooks and M. S. Wrighton, J. Am. Chem. Soc.,
1990, 112, 7869.
40 J. M. Margolis (ed.), in Conducting Polymers, New York, 1989.
38 P. T. Henderson and D. M. Collard, Chem. Mater., 1995, 7, 1879.
J. Mater. Chem., 2001, 11, 2253–2261
2261