Electropolymerisable bipyridine ruthenium(II) complexes. Synthesis and electrochemical characterisation of 4-(3-methoxystyryl)- and 4,4′-di(3-methoxystyryl)-2,2′-bipyridine ruthenium complexes

Article abstract of DOI:10.1039/b008697p

A number of new ruthenium polypyridyl complexes with mono- or di-(3-methoxystyryl) substituted bipyridines have been synthesized. The complexes were characterised by NMR, elemental analysis, UV-vis absorption and emission spectroscopy, and cyclic voltammetry. Electroactive polymer films of these complexes have been prepared both by oxidative and reductive electropolymerisation. The polymers have been characterised by UV-vis absorption spectroscopy and cyclic voltammetry. Possible processes involved in the polymerisation and the structure of the film are discussed.

Full text of DOI:10.1039/b008697p

View Article Online / Journal Homepage / Table of Contents for this issue
Electropolymerisable bipyridine ruthenium(II) complexes. Synthesis
and electrochemical characterisation of 4-(3-methoxystyryl)- and
4,4Ј-di(3-methoxystyryl)-2,2Ј-bipyridine ruthenium complexes
Viviane Aranyos,ab Johan Hjelm,b Anders Hagfeldtb and Helena Grennberg*a
a Department of Organic Chemistry, University of Uppsala, Box 531, S-751 21, Uppsala,
Sweden. E-mail: helenag@kemi.uu.se
b Department of Physical Chemistry, University of Uppsala, Box 532, S-751 21, Uppsala,
Sweden
Received 30th October 2000, Accepted 14th February 2001
First published as an Advance Article on the web 28th March 2001
A number of new ruthenium polypyridyl complexes with mono- or di-(3-methoxystyryl) substituted bipyridines
have been synthesized. The complexes were characterised by NMR, elemental analysis, UV-vis absorption and
emission spectroscopy, and cyclic voltammetry. Electroactive polymer films of these complexes have been prepared
both by oxidative and reductive electropolymerisation. The polymers have been characterised by UV-vis absorption
spectroscopy and cyclic voltammetry. Possible processes involved in the polymerisation and the structure of the film
are discussed.
and 4,4Ј-di-(2-(3-methoxyphenyl)ethenyl)-2,2Ј-bipyridine (8).
First, nucleophilic addition of the anion of 4,4Ј-dimethyl-
Introduction
Electroactive polymers are of great interest due to their appli-
2,2Ј-bipyridine to meta-methoxybenzaldehyde gave the corre-
cations in for example electrocatalysis,1 electronic devices,2,3
sponding alcohol.14 Variation of the proportion of the aldehyde
electrochromics,4 size-selective membranes,5 and solar energy
anion allowed us to prepare either the mono-substituted (5) or
the di-substituted (6) bipyridine alcohol. Dehydration was
achieved using acetic acid with an easy separation by chromato-
graphy for 7 and precipitation in the case of 8 (Scheme 1).
conversion.6 Lately, much interest has been devoted to con-
ducting organic polymers, and to materials consisting of a
conjugated polymer linked to redox active metal centres.7–9
The conjugated backbones of the latter materials may serve
to increase the electronic coupling between the metal centres
Complexes. The homoleptic complexes tris(4-(2-(3-methoxy-
phenyl)ethenyl)-4Ј-methyl-2,2Ј-bipyridine)ruthenium bis(hexa-
fluorophosphate) (1) and tris(4,4Ј-di(2-(3-methoxyphenyl)-
ethenyl)-2,2Ј-bipyridine)ruthenium bis(hexafluorophosphate)
(2) (Fig. 1) were obtained by refluxing three equivalents of the
desired ligand (7 or 8) with RuCl3 in a mixture of EtOH and
water (Scheme 2). The complexes precipitated from water
and thus the electron transport rate through the polymer.7
Transition metal complexes of bipyridine and related ligands
are attractive because of their versatility in terms of how their
spectroscopic and chemical properties can be tuned by choice
of substituents and metal.10 Electropolymerisation of such
metal complexes may be carried out either oxidatively or
reductively depending on the polymerisable ligands. Reductive
solution when the chloride counter ions were exchanged
electropolymerisation of polypyridyl transition metal com-
to hexafluorophosphates. The heteroleptic complexes bis-
(2,2Ј-bipyridyl)(4-(2-(3-methoxyphenyl)ethenyl)-4Ј-methyl-2,2Ј-
bipyridine)ruthenium bis(hexafluorophosphate) (3) and bis-
(2,2Ј-bipyridyl)(4,4Ј-di(2-(3-methoxyphenyl)ethenyl)-2,2Ј-bipyr-
idine)ruthenium bis(hexafluorophosphate) (4) (Fig. 1) were
prepared by treating the desired ligand (7 or 8, respectively)
with one equivalent of [Ru(bpy)2Cl2] in a 1 : 1 mixture of
plexes may be carried out on complexes with a vinyl function-
ality, according to the route pioneered by Abruna et al.3 The
oxidative process has been well studied for ruthenium com-
plexes bearing aromatic amine groups or pendant pyrrole or
thiophene groups.9,11 Also, organic alkoxyaryl-containing
molecules have been reported to undergo oxidative coupling to
form electronically conducting and redox-active polyphenylene
EtOH and water, followed by exchange of the chloride counter
derivatives.12,13
ions to hexafluorophosphates (Scheme 3).15
We were interested in finding versatile complexes that allow
both oxidative and reductive electropolymerisation. In the pres-
ent paper we report the synthesis and results of the spectro-
Spectroscopic characterisation of complexes
Absorption spectroscopy. As expected for ruthenium
tris-bipyridyl complexes, the absorption spectra show a broad
visible absorbance band centred at 450–480 nm corresponding
to a metal-to-ligand charge transfer (MLCT).10a The intense
bands in the UV region are assigned to two low-lying ligand
centred (LC) π → π* transitions.16 The red-shift observed for
the MLCT maximum in the visible region (Table 1 and Fig. 2),
as compared to the MLCT maximum of Ru(bpy)3, is related to
the extent of conjugation of the ligands in the complexes.16,17
Fig. 2 shows the spectra of 2 and 4 in CH3CN. The ratio of the
values of the absorption coefficients of the two ligand centred
transitions correlates with the number of vinylic linkages (Table
scopic and electrochemical characterisation of ruthenium
tris-bipyridyl complexes with methoxystyryl substituents 1–4
(Fig. 1). The oxidative and reductive electropolymerisation of
these complexes and the properties of the resulting insoluble
polymer films are discussed.
Results and discussion
Synthesis
Ligands. A two step synthesis was used to afford the ligands
4-(2-(3-methoxyphenyl)ethenyl)-4Ј-methyl-2,2Ј-bipyridine (7)
DOI: 10.1039/b008697p
J. Chem. Soc., Dalton Trans., 2001, 1319–1325
This journal is © The Royal Society of Chemistry 2001
1319

View Article Online
Fig. 1 Complexes investigated in this work.
Scheme 1 Reaction conditions: (i) 1 equivalent LiNPri2, THF, Ϫ78 ЊC, N2; (ii) conc. AcOH, reflux.
(i)
[Ru(L)3]2ϩ 2PF6
Ϫ
3 L ϩ RuCl3ؒH2O
Scheme 2 Reaction conditions: (i) (a) 1 : 1 EtOH–water, N2, reflux;
(b) NH4PF6.
(i)
Ϫ
L ϩ [Ru(bpy)2Cl2]
[Ru(bpy)2L]2ϩ 2PF6
Scheme 3 Reaction conditions: (i) (a) complex 3 or 4, 1 : 1 EtOH–
water, N2, reflux; (b) NH4PF6.
1). The absorption coefficient of the MLCT band increases
with the number of mono- and di-substituted ligands. The
methoxystyryl groups are electron rich and make the ligand
strongly electron donating. This stabilises RuIII with respect to
RuII and shifts the HOMO to higher energy.18
Fig. 2 Absorption spectrum for complexes 2 and 4 in CH3CN. The
instrument resolution is 2 nm.
4. The formal potentials were calculated from the average of the
oxidation and reduction peak potentials. Both oxidation of the
metal and of the ligand are diffusion-controlled processes, as
seen by the identical slopes and shifts in peak potentials in the
plots of peak current upon oxidation as a function of (scan
rate)1/2 (inset in Fig. 3).13 The methyl and the methoxystyryl
units are electron-donating substituents, and have a lowering
Electrochemistry and electropolymerisation of complexes
In order to assess the effects of substitution on the formal
potentials of complexes 1–4, cyclic voltammograms of their
CH3CN solutions were recorded. The formal potentials for the
Ru2ϩ/3ϩ couple of 1 and 2 are 0.74 and 0.84 V for 3 (Fig. 3) and
1320
J. Chem. Soc., Dalton Trans., 2001, 1319–1325

View Article Online
Table 1 Optical properties (maxima values) of complexes 1–4 in
CH3CN
λabsa/nm (10Ϫ4 ε/MϪ1 cmϪ1
)
Emission
max.
(nm)
b
c
Complex
MLCT
π–π*1
π–π*2
1
2
3
4
474 (3.23)
489 (3.88)
457 (1.48)
465 (1.78)
287 (10.2)
302 (8.75)
287 (6.73)
289 (7.21)
334 (6.20)
334 (8.97)
319 (2.00)
313 (4.00)
668
673
617
661
a Resolution: 2 nm. b The first ligand centred absorption band.
c The second ligand-centred absorption band.
Fig. 4 Oxidative electropolymerisation of complex 4 from a 0.37 mM
solution of the complex in CH2Cl2 onto a platinum electrode. The scan
rate is 100 mV sϪ1. The initial scan direction is positive. The thick line
indicates the first scan. The inset is a plot of the cathodic peak current ip
vs. the number of polymerisation scans.
adherent orange film is observed. The film did not dissolve
upon storage of the electrode for hours in dichloromethane or
acetonitrile.
The increase of the peak current (inset, Fig. 4) with number
of scans is related to the rate of growth of the polymer film. The
rate decreases somewhat after the initial three to four scans, but
growth continues up to at least 20 scans. 20 scans at 100 mV sϪ1
result in a film of poly-4 with an apparent surface coverage of
about 2 × 10Ϫ9 mol cmϪ2.† Roughly the same behaviour is
observed for the electrochemical deposition of poly-1 and poly-
2, but a larger increase in surface coverage per deposition scan
is observed. The electropolymerisation rate increases with the
number of polymerisable groups and films of poly-1 and poly-2
were deposited to apparent surface coverage above 2 × 10Ϫ8 mol
cmϪ2 (≈200 monolayers). When oxidative electropolymeris-
ation was carried out with the counter electrode kept in the
same compartment as the working electrode polymer film
formation was observed also at the counter electrode, where
the converse reaction takes place.
Electron transport through the polymer film takes place
by successive electron transfer between neighbouring redox
centres. The redox reactions are accompanied by uptake (or
expulsion) of counter ions to maintain electroneutrality. It
remains unclear where the film growth takes place, at the poly-
mer film/electrolyte (i) or at the electrode/polymer film interface
(ii). Neither seems to be favoured: for (i), ruthenium() centres
at the surface would have to oxidise methoxystyryl units, which
according to our electrochemical experiments should not occur;
and for (ii), cationic monomers would have to diffuse through
the polycation network of the already formed polymer film.
Fig. 3 Cyclic voltammetry of complex 3 in a CH2Cl2–TFA–TFAA
(8 : 1 : 1) ϩ 0.1 M NBu4BF4 solution. TFA/TFAA was added to take
care of residual water. The scan rates from top to bottom are 500, 200,
100, 50 mV sϪ1. The inset is a plot of the anodic peak current (not
corrected for background current) of the metal- (᭝) and ligand-centred
(ᮀ) oxidations versus the square root of the scan rate. TFA = Trifluoro-
acetic acid, TFAA = trifluoroacetic anhydride.
effect on the formal potential of the ruthenium centre with
respect to that of [Ru(bpy)3]2ϩ 18
. The fact that the formal poten-
tial of the complexes only changes with the total number of
substituents and appears unaffected by the type of substituent
(methoxystyryl or methyl) indicates that the electron donating
effects of the two substituents are similar in magnitude.
Oxidative electropolymerisation. The oxidative polymeris-
ation of complexes 1, 2 and 4 was successfully carried out on
platinum, gold, glassy carbon and FTO (fluorine doped tin
oxide) electrodes. The films were characterised by cyclic
voltammetry and absorption spectroscopy (see below). For 3,
not even repeated scanning (>40) at 100 mV sϪ1 through both
metal- and ligand-centred oxidations led to any significant
change in the voltammetric response, indicating that no poly-
mer film is formed on the electrode surface during the experi-
ment. We thus conclude that the complex must carry at least
two polymerisable groups for polymerisation to occur.
Upon oxidation of the methoxystyryl-containing ligands a
cation radical is formed. According to the literature, this is
located mainly on the phenyl ring of the methoxystyryl group.19
The cation radical can undergo radical–radical coupling to
form dimers linked through aromatic carbons, and possibly also
through the carbons belonging to the vinyl group.12 Scanning
repeatedly through the potentials of both the metal- and the
ligand-centred oxidations of a solution of complex 4 leads to
a rapid growth of the Ru2ϩ/3ϩ wave (Fig. 4), indicating the
deposition of poly-4 onto the electrode surface. The increase in
current is due to the combined electroactivity of the polymeric
film building up on the electrode surface while monomers are
diffusing towards the electrode from the bulk of the solution.
During deposition the shape of the voltammogram changes
from the typical shape of a diffusing species into the bell-
shaped voltammogram of a surface-confined species since for
each scan the current will to an increasing proportion be due
to the oxidation and reduction of surface-confined material.
Upon removal of the electrode from the solution a smooth
Reductive electropolymerisation. It has been shown that
vinylic linkages in styryl- and dimethoxystyryl-bipyridine
ligands of ruthenium tris-bipyridyl complexes are activated
towards polymerisation upon electrochemical reduction of the
ligands.20,21 The polymerisation is believed to proceed via
several pathways, of which the radical–radical hydrodimeris-
ation pathway, leading to butyl-bridged complexes, is thought
to be the most important.22,23 The methoxystyryl ligands of the
complexes studied here also undergo coupling upon electro-
reduction. Reduction of 1, 2, 3 and 4 shows three ligand reduc-
† (a) The apparent surface coverages (Γ) were calculated using the
equation Γ = Qred/nFA, where Qred is the integrated charge under the
re-reduction wave of the metal centre recorded at a scan rate where
the film exhibited a surface confined behaviour, n the number of
electrons transferred, F Faraday’s constant, and A the projected area
of the electrode. Statements about the apparent number of “mono-
layers” of polymeric material are based on a hard-sphere model
for [Ru(bpy)3]2ϩ type complexes, in which a monolayer corresponds
to roughly 1 × 10Ϫ10 mol cmϪ2. Since the complexes studied here are
significantly larger than Ru(bpy)3, the actual number of molecular
layers in the films will be somewhat larger. (b) See ref. 7(a) and
references therein.
J. Chem. Soc., Dalton Trans., 2001, 1319–1325
1321

View Article Online
Table 2 MLCT maximum for complexes 1 and 2 in solution and in
polymer films
sol
poly
max
λmpoalxy/nm
∆λ = λ Ϫ λ
max
sol
λ
/
max
Complex
nm
Oxidative Reductive Oxidative Reductive
1
2
475
485
461
469
467
n.d.
14
16
8
n.d.
Fig. 5 Cyclic voltammograms recorded continuously at a 100 mV sϪ1
in a solution of complex 1 in acetonitrile. The scan rate is 100 mV sϪ1
and the scans were initiated at Ϫ1.0 V. The thick line indicates the first
scan.
Fig.
7
UV-Vis absorption spectra of complex
2
dissolved in
acetonitrile (plain) and poly-2-ox on FTO measured in a cell filled with
acetonitrile (dashed). The absorption spectrum of 2 was referenced to
a solvent blank and the FTO/poly-2-ox spectrum was referenced to a
blank FTO electrode in solvent.
Fig. 6 Cyclic voltammetry at different scan rates of a film of
oxidatively deposited poly-4 in complex-free electrolyte. The apparent
surface coverage is ≈2 × 10Ϫ9 mol cmϪ2 which is equal to about 20
monolayers. The scan rates, from top to bottom, are 200, 100, 50 and 20
mV sϪ1. The initial scan direction was positive.
greater electrochemical stability of poly-1-red than poly-1-ox
was observed.
tions in the interval Ϫ1.5 to Ϫ2.2 V. Sweeping the potential
repeatedly through two or all three ligand-based reductions of
1 leads to the formation of an adherent orange electroactive
film on the electrode surface. Cyclic voltammograms recorded
during the reductive deposition of poly-1 onto a platinum
electrode are shown in Fig. 5. Also 2 and 4 undergo reductive
electropolymerisation, but no polymer film is formed of 3.
Absorption spectroscopy. Red-shifts of the MLCT maximum
of ruthenium complexes with large conjugated ligands with
respect to the complexes [Ru(bpy)3]2ϩ and [Ru(dmbpy)3]2ϩ
,
where dmbpy is 4,4Ј-dimethyl-2,2Ј-bipyridine, have been
reported.10a,b This is observed for our complexes (Fig. 2, Table
1) in solution. Binuclear ruthenium polypyridyl complexes
bridged by a conjugated vinyl or a divinylphenyl unit have been
shown to have significantly red-shifted MLCT maxima com-
pared to that of the corresponding mononuclear complex.17,26
However, only blue-shifts of the MLCT maxima of the
deposited polymers were observed, regardless of the deposition
method (Table 2, Fig. 7), suggesting that the conjugated
systems in the ligands are not as extended in the polymer films
as in an unchanged complex in solution. This is as expected for
the reductively prepared polymer, but surprising in the case
of the oxidatively prepared polymers.12b,c,13 According to the
literature, the consumption of vinylic linkages during poly-
merisation of transition metal complexes of 4-methyl-4Ј-
vinylbipyridine leads to a decreased extent of conjugation for
the ligands and consequently a blue-shift of the MLCT maxi-
mum.10b The blue-shift observed for our oxidatively prepared
polymers is likely due to the presence of similar saturated links
between the complexes as when the reductive route is employed.
As the shift is larger for poly-ox films than for poly-red ones,
the vinyl consumption is probably higher during oxidative
polymerisation.
Characterisation of the polymers
Electrochemistry. In Fig. 6 cyclic voltammograms of oxid-
atively deposited poly-4 (poly-4-ox) in complex-free electrolyte
are shown. The film of poly-4-ox displays one metal centred
oxidation and gives a surface wave type response up to 100 mV
sϪ1 with a peak separation of approximately 10 mV and a full
width at half-maximum (FWHM) of 120 mV.24
At higher scan rates the voltammetric response of the poly-
mer film changes from the surface-confined thin film behaviour
to a response characteristic of semi-infinite diffusion condi-
tions. This occurs at scan rates where not all the metal centres in
the film are oxidised during one scan. The redox reactions in the
polymer films are accompanied by migration of counter ions to
maintain electroneutrality. Also solvent can be transferred and
polymer configurational changes can occur. It is common that
the overall electron transport rate is governed by the counter
ion migration rate.1
Films of reductively deposited poly-1 (poly-1-red) exhibit
similar electrochemical properties to those of poly-1-ox, except
for a large irreversible wave slightly positive of the Ru2ϩ/3ϩ wave
observed in the first anodic scan. This is likely due to oxidation
of unchanged methoxy-phenyl units in the film.21 The formal
potentials of poly-1-ox and poly-1-red films are within 50 mV
of each other. The electrochemical stability of poly-1-red and
poly-1-ox is largely dependent on the potential limits of the
experiment. Both polymers are stable during extensive cycling
of the potential within 0.20 V of EЊЉ(Ru2ϩ/3ϩ). Extending the
potential window to 0.50 V leads to losses of 10–30% of
the initial electroactivity over 50 scans at a 100 mV sϪ1, but
variations between the electrodes were large and no signs of
Stability tests. In an attempt to dissolve the oxidatively and
reductively formed films of poly-1, and to explore the chemical
stability of the polymer films, polymer films on FTO were
soaked and left standing in toluene, pyridine, dimethyl sulfoxide
(DMSO) or dimethylformamide (DMF) during ten days, after
which absorption spectra of the electrodes were recorded. The
stability of the films in DMSO and DMF were significantly
higher for the reductively deposited films. For instance, in
DMF, 97% of the initial absorbance remained after ten days
for poly-1-red whereas for poly-1-ox only 76% remained. The
films were also left standing for up to 18 hours in neutral
1322
J. Chem. Soc., Dalton Trans., 2001, 1319–1325

View Article Online
and basic water, MeOH or EtOH. Also in these experiments,
poly-1-ox proved less stable than poly-1-red. The stability of
poly-1-ox in pure water was poor with a 10% loss of absorb-
ance over 18 hours whereas no significant loss of absorbance
was detected for poly-1-red in water. In 0.1 M NaOH water
solution the differences in stability of poly-1-red and poly-1-
ox are even more pronounced, with an almost complete loss
of absorbance of poly-1-ox but no significant loss at all of
poly-1-red.
of the respective complex was chosen as excitation wavelength.
The samples were diluted until the absorbance at the maximum
of the metal-to-ligand charge-transfer band was approximately
0.1.
Electrochemistry
Electrochemical experiments were performed using an ECO
Chemie Autolab PGSTAT 10 potentiostat and a three-electrode
system. Potentials are quoted with respect to the ferrocene
redox couple unless otherwise noted. All electrochemical
experiments were performed inside an argon filled drybox at
room temperature unless otherwise noted. The supporting
electrolyte in all experiments was 0.1 M NBu4BF4. The counter
electrode was a coiled platinum wire kept in a separate com-
partment fitted with a ceramic frit. The reference for meas-
urements in acetonitrile was an Ag–0.01 M AgNO3 ϩ 0.1 M
NBu4BF4 electrode, separated from the solution with an 8 cm
salt bridge filled with electrolyte and fitted with a ceramic frit.
The potential of this electrode was Ϫ0.08 V vs. the ferrocene
couple (ϩ0.24 V vs. SCE).25 A silver wire was used as quasi-
reference in all experiments carried out in dichloromethane;
its potential was measured versus the ferrocene couple after
experiments. Working electrodes of gold, platinum and glassy
carbon shrouded in Kel-F were employed. The diameters of the
electrodes were 2 mm except for the glassy carbon electrodes
which had a diameter of 4 mm. The electrodes were polished in
an aqueous suspension of 0.3 µm diameter alumina particles
(Buehler) on a polishing cloth (Buehler), sonicated in water
(deionised to a resistivity of 18 MΩ using a Millipore purifi-
cation pak) for at least 15 min to remove alumina particles
followed by 15 min sonication in dichloromethane to remove
organic impurities. The electrodes were then taken into the
drybox. Transparent conducting oxide electrodes were cut from
large FTO sheets into 10 × 25 mm pieces. The FTO electrodes
were pre-treated by sonication in acetone for at least 10 min,
rinsed under a stream of millipore water (0.1–0.5 L) and then
heated to 400 ЊC in a stream of air for 30 min. The electrodes
were taken into the drybox while still warm (>100 ЊC). Glass-
ware was heated to ≥110 ЊC overnight and brought into the
drybox while still warm. All electropolymerisation was carried
out in dilute (0.3–1 mM) solutions of the complexes at scan
rates of 50–100 mV sϪ1. Reductive electropolymerisation was
carried out in CH3CN by scanning the potential through at
least two of the three ligands reductions (all in the region Ϫ1.5
to Ϫ2.2 V vs. ferrocene). All oxidative electropolymerisation
was carried out in CH2Cl2 or CH2Cl2–TFA–TFAA (8 : 1 : 1)
mixtures. The TFAA–TFA was added to the mixture to take
care of residual water.
Conclusion
Bipyridine ligands with methoxystyryl units that can be
activated towards electropolymerisation both at positive and
negative potentials (capable of both oxidative and reductive
polymerisation) have been prepared and their ruthenium com-
plexes synthesized. Two polymerisable groups on the same
complex are necessary for the formation of an insoluble
polymeric film on an electrode surface. Oxidative electro-
polymerisation was carried out in CH2Cl2 and reductive
electropolymerisation in CH3CN. It was possible to deposit
relatively thick (≈200 layers of complex) polymer films by
oxidative deposition. The spectroscopic measurements indi-
cate a lower degree of conjugation of the complexes in the
oxidatively deposited film compared to that of the unchanged
monomer in solution. This is pointing towards a coupling,
which, to a large extent involves the carbon atoms belonging to
the vinyl groups.
Experimental
Materials
All reagents were from commercial sources used without fur-
ther purification. Solvents were HPLC grade used without
further purification with the following exceptions: acetonitrile
(CH3CN) was purchased (Aldrich) anhydrous and further dried
statically for several weeks over 3 Å molecular sieves previously
regenerated by heating to 350 ЊC under vacuum for 48 hours;
tetrahydrofuran (THF) was distilled from sodium; and diiso-
propylamine (DIPA), triethylamine (Et3N) and dichlorometh-
ane (CH2Cl2) were distilled from CaH2. Butyllithium (BuLi)
was stored in a desiccator under argon, and used as fresh (no
precipitate) 1.6 M solutions in hexane. Trifluoroacetic acid and
trifluoroacetic acid anhydride were stored at Ϫ18 ЊC and used
as received. Ferrocene was purified by sublimation and tetra-
butylammonium tetrafluoroborate was dried under vacuum at
140 ЊC for 48 hours. Syntheses were performed under a nitro-
gen atmosphere. Column chromatography was performed with
Merck silica gel (230–400 mesh). Thin-layer chromatography
(TLC) was run on Merck precoated silica gel 60-F254 plates.
Transparent conducting oxide electrodes of fluorine-doped tin
oxide on a 3 mm glass support (sheet resistance 8 Ω) were
purchased from Liffey-Owens-Ford. Elemental analyses were
performed by Analytische Laboratorien, Lindlar, Germany.
Synthesis
Substituted 4,4Ј-dimethyl-2,2Ј-bipyridines 5 and 6. Lithium
diisopropylamide was formed by mixing BuLi (2.48 × 10Ϫ3
mol for monodeprotonation (5) and 5.92 × 10Ϫ3 mol for bis-
deprotonation (6)) with DIPA (0.39 mL for monodeproton-
ation and 0.78 mL for bis-deprotonation) in 20 mL THF at
Ϫ60 ЊC under N2. The LiNPri2 solution was stirred at room
temperature for 30 minutes and cooled to Ϫ40 ЊC. 4,4Ј-
Dimethyl-2,2Ј-bipyridine (0.5 g, 2.48 × 10Ϫ3 mol) dissolved in
20 mL THF was added to the LiNPri2 through a double tipped
needle, the cooling bath removed and the black reaction
mixture then stirred at room temperature during one hour.
After cooling to Ϫ60 ЊC a solution of 3-methoxybenzaldehyde
(0.727 mL, 2.48 × 10Ϫ3 mol) in 10 mL THF was added through
a double tipped needle. A yellow precipitate was formed.
Stirring was continued for three hours, then the mixture was
stirred into water and extracted with CH2Cl2 (1 × 100 mL). The
organic phase was washed with brine (150 mL), dried over
MgSO4 and concentrated.
Physical characterisation
NMR spectra (1H at 400 MHz and 13C at 100 MHz) were
recorded on a Varian Unity 400 spectrometer for CDCl3 solu-
tions unless otherwise stated. The chloroform signals at δ 7.26
(for 1H) or 77.0 (for 13C) were used as indirect reference to TMS.
UV–Vis absorption spectra were recorded using a Hewlett-
Packard 8453 diode-array spectrophotometer and referenced
against a solvent blank (CH3CN). The fluorine-doped tin oxide
electrodes with polymer films deposited on the surface were
positioned against the wall of a 1 cm quartz cuvette containing
CH3CN. Emission spectroscopy measurements were performed
with a Spex Fluorolog Model 212 spectrometer with photon
counting detection. The emission was monitored at right angles
with respect to the excitation. The MLCT absorption maximum
J. Chem. Soc., Dalton Trans., 2001, 1319–1325
1323

View Article Online
4-(2-Hydroxy-2-(3-methoxyphenyl)ethyl)-4-methyl-2,2Ј-
bipyridine (5). The crude product was chromatographed using
ethyl acetate–pentane–Et3N (50 : 48 : 2) to give compound 5 in
46% yield (0.364 g, 1.14 × 10Ϫ3 mol). 1H NMR (CDCl3): δ 8.54
(1H, J = 5.2, d, pyridine), 8.51 (1H, J = 5.2, d, pyridine), 8.31
(1H, s, pyridine), 8.21 (1H, s, pyridine), 7.25 (1H, t, J = 8 Hz,
methoxyphenyl), 7.12 (2H, m, pyridine ϩ methoxyphenyl), 6.94
(2H, m, pyridine ϩ methoxyphenyl), 6.82 (1H, dd, J1 = 8,
J2 = 2.4 Hz, methoxyphenyl), 4.99 (1H, m, benzylic CH(OH)),
3.93 (3H, s, benzylic CH3), 3.08 (2H, m, benzylic CH2) and 2.43
(3H, s, OCH3). 13C NMR: δ 160.0, 156.2, 156.0, 149.2, 149.1,
148.7, 148.5, 145.6, 129.8, 125.2, 125.0, 122.3, 122.2, 118.3,
113.7, 111.4, 74.6, 55.5, 45.6 and 21.4.
tered off, washed with water (100 mL) and diethyl ether (100
mL). The product was dried under reduced pressure during
24 h.
Tris(4-(2-(3-methoxyphenyl)ethenyl)-4Ј-methyl-2,2Ј-
bipyridine)ruthenium bis(hexafluorophosphate) (1). 37% yield
(0.111 g, 85.5 × 10Ϫ6 mol). 1H NMR (CDCl3): δ 8.97 (3H,
pyridine), 8.76 (3H, d, pyridine), 8.04–7.67 (12H, m, methoxy-
phenyl), 7.43–7.34 (15 H, m, methoxyphenyl), 7.26 (3H, m,
methoxyphenyl), 6.94 (3H, m, methoxyphenyl), 3.84 (9H, s,
OCH3) and 2.59 (s, 9H, CH3). Calc. for C60H54F12N6O3P2Ru:
C, 55.52; H, 4.19; N, 6.47. Found: C, 55.83; H, 4.69; N,
6.14%.
Tris(4,4Ј-di(2-(3-methoxyphenyl)ethenyl)-2,2Ј-bipyrid-
ine)ruthenium bis(hexafluorophosphate) (2). Compound 2
was obtained in 41% yield (0.147 g, 86.9 × 10Ϫ6 mol). 1H
NMR (CDCl3): δ 9.15 (6H, m, pyridine), 8.10 (6H, d, J =
6.4, pyridine), 7.83 (6H, d, J = 16.4, vinylic), 7.54 (6H, d, J = 6,
pyridine), 7.46 (6H, d, J = 16.4, vinylic), 7.37 (12H, m, meth-
oxyphenyl), 7.27 (12H, m, methoxyphenyl), 6.99 (6H, dd,
J1 = 7.6, J2 = 1.6 Hz, methoxyphenyl) and 3.86 (18H, s, OCH3).
13C NMR: δ 160.5, 157.7, 151.7, 147.0, 137.6, 136.8, 130.2,
124.7, 121.3, 120.4, 115.5, 112.6 and 55.0. Calc. for C84H72-
F12N6O6P2Ru: C, 61.05; H, 4.39; N, 5.09. Found: C, 60.78; H,
4.65; N, 4.92%.
4,4Ј-Di(2-hydroxy-2-(3-methoxyphenyl)ethyl)-2,2Ј-
bipyridine (6). NMR of the crude product showed more than
85% of conversion compared to the aldehyde and complete
conversion compared to the bipyridine. The crude product 6
1
was used without further purification. H NMR (200 MHz)
(CDCl3): δ 8.37 (2H, d, J = 5, pyridine), 8.21 (2H, broad s,
pyridine), 7.24 (2H, t, J = 8.4, methoxyphenyl), 7.06 (2H, dd,
J1 = 5, J2 = 1.6 Hz, methoxyphenyl), 6.93 (4H, m, methoxy-
phenyl), 6.80 (2H, m, pyridine), 4.98 (2H, m, benzylic
CH(OH)), 3.77 (6H, s, OCH3) and 3.03 (4H, m, benzylic CH2).
13C NMR: δ 159.7, 155.6, 148.9, 148.8, 145.4, 129.5, 125.0,
122.2, 118.1, 113.3, 111.1, 74.2, 55.2 and 45.5.
Heteroleptic complexes. Bis(2,2Ј-bipyridyl)(4-(2-(3-meth-
oxyphenyl)ethenyl)-4Ј-methyl-2,2Ј-bipyridine)ruthenium
bis-
4-(2-(3-Methoxyphenyl)ethenyl)-4Ј-methyl-2,2Ј-bipyridine (7).
Compound 5 (0.367 g, 1.14 × 10Ϫ3 mol) was dissolved in conc.
AcOH (5 mL, excess) and stirred under reflux during 18 hours.
The reaction mixture was poured into sat. aq. NaHCO3 (100
mL) and extracted with CH2Cl2 (100 mL), dried over MgSO4
and concentrated to give 7 in 74% yield (0.255 g, 0.84 ×
10Ϫ3 mol) from 5 and 34% yield from 4,4Ј-dimethyl-2,2Ј-
bipyridine. 1H NMR (CDCl3): δ 8.637 (1H, d, J = 5.2, pyridine),
8.57 (1H, d, J = 4.8, pyridine), 8.51 (1H, d, J = 1.2, pyridine),
8.26 (1H, broad s, pyridine), 7.42 (1H, d, J = 16.4, vinylic), 7.37
(1H, dd, J1 = 5.2, J2 = 1.2, pyridine), 7.31 (1H, t, J = 8, meth-
oxyphenyl), 7.16 (2H, m, methoxyphenyl ϩ pyridine), 7.11
(2H, m, vinylic ϩ methoxyphenyl), 6.88 (1H, dd, J1 = 8, J2 = 2.4
Hz, methoxyphenyl), 3.86 (s, 3H, OCH3) and 2.45 (s, 3H, CH3).
13C NMR: δ 160.2, 156.8, 156.0, 149.7, 149.1, 148.5, 137.9,
133.5, 130.0, 126.7, 125.1, 122.3, 121.2, 120.0, 118.6, 114.7,
112.4, 55.5 and 21.5. Calc. for C28H18N2O: C, 79.44; H, 6.00;
N, 9.26. Found: C, 78.24; H, 6.17; N, 9.63%.
(hexafluorophosphate) (3). [Ru(bpy)2Cl2]14 (0.035 g, 66 × 10Ϫ6
mol) and compound 7 (20 mg, 66 × 10Ϫ6 mol) were stirred at
reflux in a 10 mL mixture of EtOH and water (1 : 1), under
argon for one hour. The solution turned deep red. The EtOH
was distilled off and a solution of NH4PF6 (0.215 g, 1.32 × 10Ϫ3
mol) in water (15 mL) added and stirred overnight (18 h), after
which time an orange precipitate had appeared. The precipitate
was filtered off and dried under vacuum overnight to give 3 in
94% yield (0.063 g, 62.5 × 10Ϫ6 mol). 1H NMR (acetone): δ 9.00
(1H, s, substituted pyridine), 8.84–8.78 (4H, m, non-substituted
bipyridines), 8.70 (1H, s, substituted pyridine), 8.22–8.16 (5H,
m, 4H non-substituted bipyridines ϩ 1H substituted bipyrid-
ines), 8.09–8.04 (5H, m, non-substituted bipyridines ϩ 1H
substituted bipyridines), 7.94 (1H, d, J = 5.6, substituted
pyridine), 7.77 (1H, d, J = 16, vinylic), 7.67 (1H, m, substituted
pyridine), 7.59 (4H, m, non-substituted bipyridines), 7.42 (1H,
d, J = 16 Hz, vinylic), 7.36 (1H, m, methoxyphenyl), 7.23 (2H,
m, methoxyphenyl), 6.98 (1H, m, methoxyphenyl), 3.85 (3H, s,
OCH3), 2.57 (3H, s, py CH3). Calc. for C40H34F12N6OP2Ruؒ
7H2O: C, 42.45; H, 4.27; N, 7.42. Found: C, 41.86; H, 3.79;
N, 8.00%.
4,4Ј-Di(2-(3-methoxyphenyl)ethenyl)-2,2Ј-bipyridine
(8).
Crude compound 6 was dissolved in conc. AcOH (10 mL,
excess) and the mixture stirred under reflux for 18 hours.
When the reaction mixture was cooled to room temperature
the product precipitated and 6 was obtained with a yield of 36%
(0.378 g, 0.898 × 10Ϫ3 mol) from 4,4Ј-dimethyl-2,2Ј-bipyridine.
1H NMR (CDCl3): δ 8.69 (2H, d, J = 4.8, pyridine), 8.61 (2H,
s, pyridine), 7.47 (2H, J = 16.4, d, vinylic), 7.43 (2H, d, J = 4.8,
pyridine), 7.32 (2H, t, J = 7.6 Hz, methoxyphenyl), 7.18 (2H, d,
methoxyphenyl), 7.14 (4H, m, vinylic ϩ methoxyphenyl), 6.90
(2H, m, pyridine) and 3.87 (s, 6H, OCH3). 13C NMR: δ 149.5,
148.9, 133.3, 129.8, 126.4, 124.9, 122.2, 121.1, 119.7, 118.4,
114.5, 112.1, 55.3 and 21.2. IR (KBr): 3306, 2937, 1598, 1487,
1462, 1435, 1263, 1149, 1046, 783 and 700 cmϪ1. Calc. for
C14H12NO: C, 79.98; H, 5.75; N, 6.66. Found: C, 79.11; H, 5.81;
N, 6.56%.
Bis(2,2Ј-bipyridyl)(4,4Ј-di(2-(3-methoxyphenyl)ethenyl)-
2,2Ј-bipyridine)ruthenium
bis(hexafluorophosphate)
(4).
[Ru(bpy)2Cl2]14 (0.025 g, 47.5 × 10Ϫ6 mol) and compound 8
(0.020 g, 66 × 10Ϫ6 mol) were stirred at reflux in a 10 mL
mixture of EtOH and water (1 : 1), under argon for one hour.
The solution turned deep red. The EtOH was distilled off and
a solution of NH4PF6 (0.080 g, 0.48 × 10Ϫ3 mol) in water (10
mL) added and stirred overnight (18 h), after which time an
orange precipitate had appeared. The precipitate was filtered
off and dried under vacuum overnight to afford 4 in 46%
1
yield (0.024 g, 22.3 × 10Ϫ6 mol). H NMR (acetone): δ 9.08
(2H, s, substituted pyridine), 8.84 (4H, d, J = 8.4, non-
substituted bipyridines), 8.23 (4H, t, J = 8, non-substituted
bipyridines), 8.19 (2H, d, J = 5.6, methoxyphenyl), 8.08 (2H,
d, J = 5.6, substituted pyridine), 7.96 (2H, d, J = 6, methoxy-
phenyl), 7.79 (2H, d, J = 16, vinylic), 7.71 (2H, d, J = 6, substi-
tuted pyridine), 7.61 (4H, m, non-substituted bipyridines), 7.45
(2H, d, J = 16 Hz, vinylic), 7.37 (2H, m, methoxyphenyl), 7.27
(4H, m, non-substituted bipyridines), 6.98 (2H, m, methoxy-
phenyl) and 3.86 (6H, s, OCH3). Calc. for C48H40F12N6O2P2-
Ruؒ2Et2O: C, 52.87; H, 4.75; N, 6.61. Found: C, 53.74; H, 4.46;
N, 6.77%.
Homoleptic complexes. RuCl3ؒxH2O (48 mg, 0.23 × 10Ϫ3 mol)
and compound 7 or 8 (0.69 × 10Ϫ3 mol) were stirred at reflux in
a 20 mL mixture of EtOH and water (1 : 1) under nitrogen for
24 hours during which time the solution turned dark red. The
EtOH was distilled off and a solution of NH4PF6 (1.24 g,
7.66 × 10Ϫ3 mol) in water (25 mL) added, after which an
orange-brown precipitate appeared. The precipitate was fil-
1324
J. Chem. Soc., Dalton Trans., 2001, 1319–1325

View Article Online
A. Von Zelewsky, Coord. Chem. Rev., 1988, 84, 85; (b) S. L.
Acknowledgements
Bommarito, S. P. Lowery-Bretz and H. D. Abruna, Inorg. Chem.,
1992, 31, 495; (c) C. Kaes, A. Katz and M. W. Hosseini, Chem. Rev.,
2000, 100, 3553.
The Swedish Research Council for Engineering Sciences
(TFR), the Foundation for strategic environmental research
(MISTRA), and the Swedish Natural Science Research Council
(NFR) are acknowledged for financial support.
11 M.-N. C. Dunand-Sauthier, A. Deronzier, J.-C. Moutet and
S. Tingry, J. Chem. Soc., Dalton Trans., 1996, 2503.
12 (a) L. M. Goldenberg and P. C. Lacaze, Synth. Met., 1993, 58, 271;
(b) Rault-Berthelot, M. M. Granger and E. Raoult, J. Electroanal.
Chem., 2000, 486, 40; (c) A. Haj Said, C. Dridi, S. Roudesli
and F. Matoussi Mhalla, Eur. Polym. J., 2000, 36, 909; (d) B. Fabre,
K. Michelet, N. Simonet and J. Simonet, J. Electronanal. Chem.,
1997, 425, 67.
13 A. Haj Said, F. Matoussi Mhalla, C. Amatore and J.-N. Verpeaux,
J. Electroanal. Chem., 1999, 464, 85.
14 B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978, 17,
3334.
15 T. A. Heimer, S. T. D’Arcangelis, F. Farzad, J. M. Stipkala and
G. J. Meyer, Inorg. Chem., 1996, 35, 5319.
16 P. D. Beer, O. Kocian, R. J. Mortimer and C. Ridgeway, J. Chem.
Soc., Dalton Trans., 1993, 2629.
17 J. R. Shaw, R. T. Webb and R. H. Schmehl, J. Am. Chem. Soc., 1990,
112, 1117.
References
1 C. P. Andrieux and J.-M. Savéant, Molecular Design of Electrode
Surfaces, Wiley-Interscience, New York, 1992, vol. 22 and references
therein; M. E. G. Lyons, Electroactive Polymer Electrochemistry:
Part 1—Fundamentals, 2nd edn., Plenum Press, New York, 1994
and references therein.
2 P. Denisevich, K. W. Willman and R. W. Murray, J. Am. Chem. Soc.,
1981, 103, 4727.
3 H. D. Abruna, P. Denisevich, M. Umana, T. J. Meyer and R. W.
Murray, J. Am. Chem. Soc., 1981, 103, 1.
4 P. M. S. Monk, R. J. Mortimer and D. R. Rosseinsky,
Electrochromism: Fundamentals and Applications, VCH, Weinheim,
1995.
5 T. Ikeda, R. Schmehl, P. Denisevich, K. Willman and R. W. Murray,
J. Am. Chem. Soc., 1982, 104, 2683; A. G. Ewing, B. J. Feldman
and R. W. Murray, J. Phys. Chem., 1985, 89, 1263; S. Gould,
K. H. Gray, R. W. Linton and T. J. Meyer, J. Phys. Chem., 1995,
99, 16052.
6 H. D. Abruna and A. J. Bard, J. Am. Chem. Soc., 1981, 103, 6898;
J. A. Moss, J. M. Stipkala, J. C. Yang, C. A. Bignozzi, G. J. Meyer,
T. J. Meyer, X. Wen and R. W. Linton, Chem. Mater., 1998, 10, 1748;
M. A. Fox and P. F. Britt, New J. Chem., 1992, 16, 113; E. M.
Girotto, W. A. Gazotti and M.-A. De Paoli, J. Phys. Chem. B, 2000,
104, 6124.
7 (a) C. G. Cameron and P. G. Pickup, J. Am. Chem. Soc., 1999, 121,
11773; (b) G. Zotti, G. Schiavon, S. Zecchin, A. Berlin, G. Pagani
and A. Canavesi, Synth. Met., 1996, 76, 255.
8 S. S. Zhu and T. M. Swager, Adv. Mater., 1996, 8, 497; B. J. MacLean
and P. G. Pickup, Chem. Commun., 1999, 2471.
18 V. Skarda, M. J. Cook, A. P. Lewis, G. S. G. McAuliffe, A. J.
Thomson and D. J. Robbins, J. Chem. Soc., Perkin Trans. 2, 1984,
1293.
19 A. J. Bard, Encyclopedia of Electrochemistry of the Elements,
vol. XI, Marcel Dekker, New York, 1978.
20 C. R. Leidner, B. P. Sullivan, R. A. Reed, B. A. White, M. T.
Crimmins, R. W. Murray and T. J. Meyer, Inorg. Chem., 1987, 26,
882.
21 P. D. Beer, O. Kocian, R. J. Mortimer and C. Ridgeway, J. Chem.
Soc., Faraday Trans., 1993, 333.
22 J. M. Calvert, R. H. Schmehl, B. P. Sullivan, J. S. Facci, T. J. Meyer
and R. W. Murray, Inorg. Chem., 1983, 22, 2151.
23 T. F. Guarr and F. C. Anson, J. Phys. Chem., 1987, 91, 4037.
24 Theoretically, at room temperature, a 0 V peak separation and a
90.6 mV FWHM is expected for a one-electron reaction of an
ideal surface-immobilised redox couple experiencing no lateral
intermolecular interactions. See for example ref. 25.
25 A. J. Bard and L. R. Faulkner, Electrochemical Methods:
Fundamentals and Applications, John Wiley & Sons, New York,
1980.
9 J. G. Eaves, H. S. Munro and D. Parker, Inorg. Chem., 1987, 26, 644;
J. A. Crayston, A. Iraqi, J. J. Morrison and J. C. Walton, Synth.
Met., 1997, 84, 441; J. Wang and F. R. Keene, J. Electroanal. Chem.,
1996, 405, 71; C. D. Ellis, L. D. Margerum, R. W. Murray and
T. J. Meyer, Inorg. Chem., 1983, 22, 1283.
26 S. Boyde, G. F. Strouse, W. E. Jones and T. J. Meyer, J. Am. Chem.
Soc., 1990, 112, 7395.
10 (a) A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and
J. Chem. Soc., Dalton Trans., 2001, 1319–1325
1325