Facile synthesis and electrochemical properties of two trinuclear ruthenium complexes based on star-shaped terpyridine derivatives

Article abstract of DOI:10.1039/b706419e

Two trinuclear ruthenium complexes [{TolylTerpyRu}₃(L3A)] ·(PF₆)₉ and [{TolylTerpyRu}₃(L3B)] ·(PF₆)₉ containing the star-shaped ligands 1,3,5-tris{[4′-(2,2′:6′,2″-terpyridinyl)-1-pyridinium

Full text of DOI:10.1039/b706419e

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Facile synthesis and electrochemical properties of two trinuclear
ruthenium complexes based on star-shaped terpyridine derivativesw
a
a
b
a
´
´ ´
Cedric R. Mayer,* Frederic Dumur, Fabien Miomandre, Eddy Dumas,
Thomas Devic, Celine Fosse and Francis Secheresse
a
c
a
´
´
Received (in Montpellier, France) 27th April 2007, Accepted 13th June 2007
First published as an Advance Article on the web 5th July 2007
DOI: 10.1039/b706419e
Two trinuclear ruthenium complexes [{TolylTerpyRu}3(L3A)] ꢀ (PF6)9 and
[{TolylTerpyRu}3(L3B)] ꢀ (PF6)9 containing the star-shaped ligands 1,3,5-tris{[40-(2,20:60,200-
terpyridinyl)-1-pyridiniumyl]methylphenyl}benzene (L3A) and 2,4,6-tris{[40-(2,20:60,200-
terpyridinyl)-1-pyridiniumyl]methyl}mesitylene (L3B), respectively, have been synthesized and
characterized using electrospray ionization mass spectrometry, UV-Vis spectroscopy,
1H and 13C NMR spectroscopy. These cationic ligands have been synthesized using a 1,3,5-
triphenylbenzene or a mesitylene core and the N-alkylation of the 40-pyridyl group of
40-(4-pyridyl)-2,20:60,200-terpyridine (terpy,py). The electrochemical properties of the two trinuclear
ruthenium complexes have been compared to analogous dinuclear and mononuclear ruthenium
complexes.
ified to synthesize ligands with specific characteristics. For
Introduction
example, we recently used the strong affinity of the 2-mercap-
topyridine moiety of the 40-(2-mercaptopyridyl)-2,20:60,200-
terpyridine ligand to stabilize and functionalize gold nanopar-
ticles. The nanocomposites thus generated, bearing terpy
pendant groups, have then been used as effective platform
for coordination chemistry.10 Constable et al. elegantly used
the N-alkylation of the 40-pyridyl group of terpy,py to
isolate metallamacrocycle complexes.9,11,12 In these supramo-
lecular assemblies, two {M(terpy,py)2} (M = Fe(II), Ru(II))
units were linked through aryl bridges. Inspired by these initial
studies, Kurth et al. recently reported the synthesis of electro-
chromic thin films incorporating a new cobalt(II)-metalla-
viologen.13
Polypyridyl ruthenium complexes have been extensively in-
vestigated because of their varied physicochemical properties,
leading to a wide range of successful or potential applications
such as solar energy conversion, biotechnologies or electro-
luminescent devices.1 Hundreds of such complexes have been
described due to the ability of chemists to build virtually
‘‘on-demand’’ polypyridyl ligands. Among the variety of
N-heterocycles, 2,20-bipyridine and 2,20:60,200-terpyridine
(terpy) derivatives have been thoroughly studied. Hosseini
and co-workers recently reviewed an exhaustive database of
ligands comprising two 2,20-bipyridine units.2 The tridentate
2,20:60,200-terpyridine unit has been extensively used to
generate original inorganic–organic hybrid structures due to
its high affinity toward transition metal ions. For example, the
major role of terpy in metallosupramolecular chemistry has
been recently examined.3,4 Another major feature accounting
for the tremendous interest in terpy lies in its ‘‘facile’’
40-functionalization.3–7 Among these 40-functionalized terpy,
40-(4-pyridyl)-2,20:60,200-terpyridine (terpy,py) has proven to
be a versatile heteroditopic ligand for the controlled self-
assembly of monolayers of metallic complexes on platinum
surface,8 or to build metallamacrocycles or metallapolymers.9
The pyridine pendant group of terpy,py can be further mod-
Our group recently reported the elaboration of nanocom-
posites in which gold or silver nanoparticles were stabilized
and functionalized by polypyridyl ruthenium com-
plexes.10,14,15 The aim of our studies is to induce a synergistic
effect between the intrinsic properties of the two nanocompo-
site components, namely the metallic nanoparticle and the
ruthenium complex. Polypyridyl ruthenium complexes have
been chosen due to their extensively studied optical and
electrochemical properties. Ru(II) has also been chosen be-
cause of its already established ability to generate asymmetric
salts of general formula [RuLL0]2+, L and L0 being tridentate
ligands such as terpy derivatives. Up till now, we employed
only mononuclear complexes and the next step is indeed the
extension to polynuclear ruthenium complexes. In this
context, star-shaped polynuclear metallic complexes have
been selected due to their attractive physico-chemical
properties.16–18
a
´
Institut Lavoisier de Versailles, UMR 8180, Universite de Versailles,
45 Avenue des Etats-Unis, 78035 Versailles, France. E-mail:
cmayer@chimie.uvsq.fr; Fax: +33 1 39 25 43 81; Tel: +33 1 39 25
43 97
´
Laboratoire PPSM, UMR 8531, Ecole Normale Superieure de
Cachan, 61 Avenue du Pre´sident Wilson, 94235 Cachan, France
b
c
In this paper, we took advantage of the facile N-alkylation
of the 40-pyridyl pendant group of terpy,py to synthesize two
new tripodal terpy-terminated ligands (Scheme 1, L3A and
L3B) that can further coordinate three Ru(II) centres.
Both star-shaped ligands differ by the nature of the core
´
´
Service de Spectrometrie de Masse, Ecole Nationale Superieure de
Chimie de Paris, 24 Rue Lhomond, 75231 Paris Cedex 05, France
w Electronic supplementary information (ESI) available: 13C NMR
spectra of L3B and Ru3B. Plot of peak current vs. square root of the
scan rate for compounds Ru1, Ru2 and Ru3A. See DOI: 10.1039/
b706419e
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1806 | New J. Chem., 2007, 31, 1806–1814

View Article Online
Scheme 1 Structures of the ligands used in this paper.
(1,3,5-triphenylbenzene or mesitylene) and by the presence of
an additional phenylene group in each arm of the spacer
connected to the three terpy pendant groups. Here, we report
the syntheses and the characterization (1H and 13C NMR,
ESI-MS, UV/Vis) of the two aforementioned ligands and of
the corresponding triruthenium complexes. To discuss the
electrochemical properties of the two trinuclear ruthenium
complexes Ru3A and Ru3B, analogous dinuclear (Ru2) and
mononuclear (Ru1) ruthenium complexes have been synthe-
sized and characterized.
Scheme 2 Structures of the ruthenium complexes reported in this paper.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1806–1814 | 1807

View Article Online
Results and discussion
Syntheses of ligands and ruthenium complexes
Star-shaped ligands have been synthesized using a variety of
cores such as nitrogen atom,19 benzene substituted in 1,3,5
positions16 or 1,3,5-triazine substituted in 2,4,6 positions.16,20
In this paper, ligands were obtained by using a benzene
derivative core and the N-alkylation of the 40-pyridyl group
of terpy,py.
Ligand L3A (Scheme 1) was obtained starting from 1,3,5-
tris[4-methylphenyl]benzene that was reacted with three
equivalents of NBS to give 1,3,5-tris[4-(bromomethyl)phenyl]-
benzene. The later was finally converted to L3A by reaction
with three equivalents of terpy,py. Ligands L3B and L1
(Scheme 1) were synthesized following the procedure described
for L3A but starting from the commercially available 2,4,6-
tris(bromomethyl)mesitylene, and 4-bromomethyltoluene, re-
spectively. The four ligands L3A, L3B, L213 and L1 were
reacted with three, three, two and one equivalent(s) of [{40-
(4-tolyl)-2,20,60,200-terpyridine}RuCl3]
(TolylTerpyRuCl3),
respectively, in a refluxing mixture of EtOH–H2O–DMF to
give the ruthenium complexes Ru3A, Ru3B, Ru2 and Ru1
(Scheme 2).
1
Electrospray ionization mass spectrometry, H and 13C NMR
Two types of mass spectrometers were used to characterize the
ligands and the ruthenium complexes described in this article.
The single quadrupole mass spectrometer allowed the char-
acterization of the ligands but not the identification of the final
ruthenium complexes. A systematic fragmentation of the
ruthenium complexes was detected and only two peaks were
observed at m/z = ca. 367.6 and 880.1, attributed to [(Tolyl-
Terpy)Ru(terpy,py)]2+ and [(TolylTerpy)Ru(terpy,py)
+
PF6]+, respectively. Consequently, Ru3A, Ru3B, Ru2 and
Ru1 were characterized using the triple quadrupole mass
spectrometer. The resolution of the peak at m/z = 1140.2
attributed to [{(TolylTerpy)Ru}3(L3A) + 6PF6]3+ in the ESI-
MS spectrum of Ru3A (positive ion mode) is shown in Fig.
1(a) (measured) and (b) (calculated). The ESI-MS spectrum
(positive ion mode) of Ru3B is shown in Fig. 1(c). Two peaks
attributed to the fragmentation of this trinuclear complex were
still observed at m/z = 367.9 and 880.6 (marked with an
asterisk in Fig. 1) but six other peaks were detected on the
spectrum. These six peaks were straightforwardly attributed to
the six different states of charge (z) corresponding to the ion
Fig. 1 Resolution of the peak at m/z = 1140.6 attributed to
[{(TolylTerpy)Ru}3(L3A) + 6PF6]3+ in the ESI-MS spectrum of
Ru3A, (a) measured and (b) calculated. (c) ESI-MS spectrum of
Ru3B (see text). Peaks marked with * and ** are attributed to
[(TolylTerpy)Ru(terpy,py)]2+ and [(TolylTerpy)Ru(terpy,py)
+
PF6]+, respectively.
pairs of general formula [{(TolylTerpy)Ru}3(L3B) + xPF6]z+
,
realized for d 6, 5.95 and 5.89 ppm for L3A, L2 and
L1, respectively. The N-alkylation was also confirmed by
the set of peaks located between 7 and 10 ppm attributed to
the aromatic protons, characteristic of a symmetrical complex,
and by the integration of these aromatic protons compared
with the integration of the protons H4 (see above) and
H1 (–CH3, d 2.59 ppm) of the central mesitylene derivative.
with z = 9 ꢂ x and 1 r x r 6. For all these peaks, the
envelope of the isotopic pattern was in good agreement with
the simulated ones.
The 1H and 13C NMR spectra were recorded in
DMSO-d6. A complete assignment of the 1H (Fig. 2) and
13C NMR spectra of L3B and Ru3B was realized by using
two-dimensional COSY and HMQC experiments (Fig. 3).
Characteristic features were obtained from these spectra.
For example, the N-alkylation of the three 40-pyridyl pendant
groups was evidenced by the presence of a single peak at
d = 6.28 ppm assigned to H4 (+NCH2 group) in the
1H NMR spectrum of L3B. Identical assignments were
The N-alkylation also induced
a noticeable downfield
shift of the protons of the pyridyl group (H5 and H6). The
reaction of TolylTerpyRuCl3 with L3B was indeed corro-
borated by significant chemical shifts of the protons of the
pendant terpy group of L3B (for example, Dd = ꢂ1.28 ppm
for H15).
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1808 | New J. Chem., 2007, 31, 1806–1814

View Article Online
Other ligands and ruthenium complexes reported in this
article were also characterized using ESI-MS, 1H and 13C
NMR spectroscopy. The 1H NMR spectrum of Ru3A is shown
in Fig. 4.
Electrochemistry
The electrochemical properties of Ru3A, Ru3B, Ru2 and Ru1
were investigated in acetonitrile on platinum electrode. Typi-
cal cyclic voltammograms (CVs) are shown in Fig. 5, in the
case of Ru1 (Fig. 5(a)) and Ru3A (Fig. 5(b)). Four reversible
redox waves can be clearly identified for each compound. The
values of the standard potentials associated to these redox
couples are listed in Table 1. The oxidation of Ru(II) into
Ru(III) occurs at an average standard potential of ca. +0.98 V
vs. Ag+/Ag, that is +0.90 V vs. Fc+/Fc, which is very close to
2+ 21
.
the standard potential for Ru(tpy)2
This result evidences
that oxidation is fully centred on Ru, with very weak influence
of the ligand nature. This is also confirmed by the comparison
with two similar complexes of Ru1, namely those in which the
tolylpyridinium moiety is replaced by a pyridine (Ru10) or a 2-
chloropyridine (Ru100). This comparison between three mono-
nuclear complexes is also useful to assess the reduction waves:
the most cathodic redox couple clearly involves the reduction
of the tolyl substituted terpy, which is the electron richest
ligand: its standard potential lies in the same range as the
lowest one in [Ru(tpy)2]2+, despite the donor character of the
tolyl group, counter-balanced by the electroattracting substi-
tuents on the other tpy. The intermediate redox system is
present in all the Ru1, Ru10 and Ru100 complexes, thus invol-
ving the pyridinium substituted terpy ligand. The standard
potential is positively shifted by ca. 130 mV vs. tpy due to the
electron withdrawal effect (although the now reduced pyridi-
nium function alleviates this effect). However this shift is
stronger than for pyridine substitution, as was already men-
tioned by Constable et al.22 It is noteworthy that this redox
system presents marked adsorption features, especially in the
case of Ru3A. Finally the least cathodic peak corresponds to
the reduction of the tolylpyridinium into a stable radical (note
that all these reduction waves correspond to monoelectronic
transfers, as evidenced by the comparison between peak
Fig. 2 1H NMR spectra of (a) L3B and (b) Ru3B.
Fig. 3 COSY spectra of (a) L3B and (b) Ru3B.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1806–1814 | 1809

View Article Online
oxidation waves. In the case of Ru3B, the first reduction
potential is shifted positively, as a result of stronger coulombic
repulsion between pyridinium groups within this complex
compared to the analogous Ru3A, making it easier to reduce.
It is also worth mentioning that the peak to peak separation
for the oxidation of Ru1 and Ru3A is nearly the same, which is
an evidence for the same number of electrons exchanged
(namely one) during the rate determining step of the charge
transfer kinetics: thus the three electrons in Ru3A are with-
drawn at the same potential but not at the same rate, as is
usually the case in multielectronic systems.23
Fig. 4 1H NMR spectrum of Ru3a.
Besides the determination of thermodynamic potentials,
CVs recorded at various scan rates can also give valuable
information about the diffusion features of all these com-
pounds. The plot of the anodic peak current (A) vs. the square
root of scan rate (V sꢂ1) for Ru(II)/Ru(III) leads to straight
lines (see ESIw), with slopes given by the Randles–Sevcik
equation:25
currents): this reduction is reversible, demonstrating the highly
delocalized nature of the formed radical, compared to general
alkylpyridinium compounds;23 it occurs at relatively high
potentials, actually only 100 to 150 mV more negative than
the reduction of the same function in methylpyridiniumtris-
pyridylpyrazine (see Table 1), the difference being due to the
higher attracting power of pyrazine compared to pyridine
group.
pffiffiffiffi
pffiffi
ip ¼ ð2:7 ꢃ 105ÞnAC
D
v
Table 1 also summarizes the comparison between standard
potential values for mononuclear (Ru1), dinuclear (Ru2) and
trinuclear (Ru3A and Ru3B) compounds. With the exception
of the first reduction of Ru3B, one can notice the quasi-
invariance of all these potentials with the number of Ru atoms
in the complex, as an evidence of the absence of interactions
between redox centres. This result was expected due to the
interrupted conjugation induced by the methyl group connect-
ing the terpy,py and the core of the complex and also due to
the large distance between metallic centres in the case of the
with: n = total number of exchanged electrons
A = electroactive area of the electrode (cm2)
C = bulk concentration of the electroactive species
(mol cmꢂ3
)
D = diffusion coefficient (cm2 sꢂ1
)
Recording the same curve Ip vs. v1/2 for a standard com-
pound like ferrocene (DFc = 2.3 ꢃ 10ꢂ5 cm2 sꢂ1 in acetoni-
trile)26 in the same experimental conditions allows us to get rid
of the unknown A factor, thus giving values for the product
Fig. 5 CVs (cathodic part upwards and anodic part downwards) of: (a) Ru1 (ca. 8 ꢃ 10ꢂ4 M) and (b) Ru3A (ca. 2 ꢃ 10ꢂ4 M) in acetonitrile +
TBAPF6 on Pt electrode. Scan rate: 200 mV sꢂ1. Potentials refer to Ag+/Ag reference electrode.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1810 | New J. Chem., 2007, 31, 1806–1814

View Article Online
Table 1 Electrochemical data for compounds Ru1, Ru2, Ru3A and Ru3B and similar compounds. Compounds Ru10 and Ru100 are the analogues
of Ru1 by substituting the tolylpyridinium moiety by, respectively pyridine and 2-chloropyridine at the same position. Tpy = terpyridine; Metpp =
2-[2-(1-methylpyridinium)]-3,5,6-tri(2-pyridyl)pyrazine. Potential values refer to ferrocene (+0.08 V vs. our reference)
Compound
E1ox/V
DEp/mV
E1red 1/V
DEp/mV
E1red 2/V
DEp/mV
E1red 3/V
DEp/mV
Ru10
Ru1
0.91
0.87
0.88
0.89
0.89
0.93
0.88b
1.20c
68
55
70
60
78
85
—
—
ꢂ1.10
—
—
ꢂ1.12
ꢂ1.13
ꢂ1.07a
65
—
—
60
70
—
ꢂ1.51
ꢂ1.61a
ꢂ1.56a
ꢂ1.52
ꢂ1.52
ꢂ1.53a
ꢂ1.65b
ꢂ1.15c
65
—
—
60
76
—
ꢂ1.93
ꢂ1.89a
ꢂ1.89a
ꢂ1.86
ꢂ1.87
ꢂ1.87a
ꢂ1.90b
ꢂ1.95c
100
—
—
140
140
—
Ru100
Ru2
Ru3A
Ru3B
Ru(tpy)2
2+
Ru(tpy)(Metpy)3+
ꢂ1.00c
—
—
—
a
b
Peak potentials. From ref. 20. From ref. 24.
c
nD1/2. The values of D deduced from the slopes for the four
complexes are given in Table 2, assuming that the number of
exchanged electrons is equal to the number of ruthenium
centres. These values range well with the variation of size
when the number or ruthenium centres increases. In particu-
lar, one can consider that the radius of Ru3A is approximately
equal to the diameter of Ru1, since this latter constitutes one
‘‘arm’’ of the tripodal trinuclear complex. Thus one can expect
that the diffusion coefficient of Ru3A to be around one half of
that of Ru1, based on Stokes–Einstein relationship. In the case
of Ru2, the diffusion coefficient value is found closer to that of
Ru1, probably because of the anisotropy of this complex that
has a cylindrical shape rather than a spherical one for the two
others. The good agreement between the measured values of
the diffusion coefficient and the geometries of the complexes
validates the assumption concerning the number of exchanged
electrons per ruthenium centre, as also confirmed by the
comparison of the current peaks between oxidation of ruthe-
nium and reduction of the ligand in each compound.
not strictly proportionally to that number, possibly because of
a partial overlap between the absorption cross sections of each
chromophore.
Finally, one can mention that these complexes are almost
not luminescent at room temperature, as expected for tpy
substituted ruthenium complexes although in some cases, the
modification or substitution of the tpy ligand can induce
enhanced luminescence properties.27
Conclusions
N-alkylation of the 40-pyridyl group of 40-(4-pyridyl)-
2,20:60,200-terpyridine (terpy,py) using tribromo-benzene or
-mesitylene derivative cores has been efficiently used to gen-
erate original star-shaped ligands bearing three terpyridine
pendant groups. Such tripodal ligands can straightforwardly
react with metal complexes such as [(40-functionalized-
terpy)RuCl3]. Two trinuclear ruthenium complexes have been
synthesized and characterized in this paper but the construc-
tion of star-shaped metallodendrimers can also be foreseen. In
addition, the introduction of specific pendant groups such as
thiol or mercaptopyridine would generate supramolecular
complexes adapted for the functionalization of metallic nano-
particles. Such polynuclear multifunctionalized systems are
also perfectly adapted for the self-assembly of metallic NPs
through reversible redox systems. The well-known acceptor
properties of alkylpyridinium groups present in these com-
plexes makes them attractive systems to build multifunctional
nanocomposites. The good reversibility of the electrochemical
response of these compounds is an essential point to design
nanocomposites acting as redox probes, sensitive to their
environment; this is also a key feature to investigate the
influence of the nanoparticle on the electrochemical properties
of the coordination complex in the nanocomposite.
UV-Vis spectroscopy
Absorption spectra were recorded for complexes Ru3A, Ru3B,
Ru2 and Ru1 in acetonitrile and show the classical features for
this kind of coordination compounds,22 especially the presence
of a broad MLCT absorption band in the 500 nm wavelength
region. The position of the MLCT absorption band is red
shifted by ca. 25 nm compared to Ru(tpy)2+, which seems in
agreement with the less negative reduction potential of the
pyridinium substituted tpy ligand, resulting in a stabilized
LUMO.
The maximum absorption wavelength is independent of the
number of ruthenium centres, confirming the absence of
interactions predicted by the electrochemical results. The
molar extinction coefficient values for this MLCT band in-
creases with the number of Ru centres (see Table 3), in
accordance with an increasing number of chromophores, but
Table 3 Maximum absorption wavelength and molar extinction
coefficient for solutions of Ru1, Ru2, Ru3A and Ru3B in acetonitrile
(the concentration in complex ranges from 4 ꢃ 10ꢂ6 M to 1.4 ꢃ
10ꢂ5 M for Beer–Lambert plots)
Table 2 Diffusion coefficients for compounds Ru1, Ru2 and Ru3A,
deduced from the slopes of ip vs. v1/2 for the oxidation of Ru(II),
assuming one exchanged electron per ruthenium centre
Compound
lmax/nm
e/104 L molꢂ1 cmꢂ1
Compound
D/10ꢂ5 cm2 sꢂ1
Ru1
Ru2
Ru3A
Ru3B
503
500
498
505
3.09
4.61
6.56
6.50
Ru1
Ru2
Ru3A
0.93
0.86
0.51
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1806–1814 | 1811

View Article Online
solution was then maintained at reflux overnight, leading to
the precipitation of a white solid. The reaction mixture was
cooled to room temperature. The white precipitate was filtered
off, washed with dichloromethane (20 mL), diethyl ether (20
mL) and dried under vacuum to afford L3A ꢀ Br3 in 86% yield
(Found: C, 68.7; H, 4.0; Br, 15.6; N, 10.9. C87H63Br3N12
requires C, 68.9; H, 4.2; Br, 15.8; N, 11.1%); dH (300 MHz;
DMSO-d6; Me4Si; 25 1C) 6.0 (6H, br s), 7.6 (6H, dd), 7.77 (6H,
dd), 8.0 (6H, td), 8.1 (6H, td), 8.71 (6H, td), 8.77 (3H, s), 8.79
(6H, d), 8.85 (6H, d), 8.91 (6H, s) and 9.45 (2H, d); m/z
(ESI-MS, HP 5989B) 425.1 (M3+ requires 425.5).
Experimental
General methods
ESI-MS measurements were carried out with an API 3000
(ESI/MS/MS) PE-SCIEX triple quadrupole mass spectro-
meter and HP 5989B single quadrupole mass spectrometer
equipped with an electrospray source from Analytica of
Branford. Both instruments were operated in the positive ion
mode. For the API 3000 (ESI/MS/MS) PE SCIEX triple
quadrupole mass spectrometer, the experiments were per-
formed either by direct infusion with a syringe pump with
flow rate of 10 mL minꢂ1 or flow injection acquisition with flow
rate of 200 mL minꢂ1. Standard experimental conditions were
as follows: sample concentration 10ꢂ3 to 10ꢂ5 M, nebulizing
gas N2: 7 units flow rate, ion spray voltage ꢂ5.00 kV,
temperature: 200–400 1C, declustering potential: ꢂ20 V,
focusing potential: ꢂ200 V, entrance potential: 10 V.
2,4,6-Tris{[40-(2,20:60,200-terpyridinyl)-1-pyridiniumyl]methyl}-
.
mesitylene tribromide (L3B Br3). L3B ꢀ Br3 was synthesized
from terpy,py (95 mg, 0.31 mmol) and 2,4,6-tris(bromo-
methyl)mesitylene (39.9 mg, 0.10 mmol), using the procedure
described for the synthesis of L3A ꢀ Br3, in 83% yield (Found:
C, 64.7; H, 4.4; Br, 18.4; N, 12.3. C72H57Br3N12 requires C,
65.0; H, 4.3; Br, 18.0; N, 12.6%); dH (300 MHz; DMSO-d6;
Me4Si; 25 1C) 2.43 (9H, s, H1), 6.28 (6H, br s, H4), 7.54 (6H,
UV-Vis spectra were recorded using a CARY 500 UV-Vis-
NIR spectrophotometer (Varian). Each sample was analyzed
in a similar way: about 10 mL of a 1 mM solution of the
complex were introduced into the cell (1 cm optical path) and
diluted with an appropriate volume of acetonitrile (spectro-
scopic grade), in order to obtain absorbance values lower than
1. Measurements of the molar absorption coefficient were
made by recording the spectra corresponding to several ali-
quots of 10 mL of the complex solution in 2 mL of acetonitrile,
and then drawing a Beer–Lambert plot.
1H and 13C NMR spectra were obtained at room tempera-
ture in 5 mm o.d. tubes on a Bruker Avance 300 spectrometer
equipped with a QNP probe head. A numbering scheme of the
ligands and complexes is given in ESIw when needed for the
assignment.
3
3
3
dd, J = 7.2 Hz, J0 = 6.9 Hz, H14), 8.06 (6H, td, J13–12
3J13–14 = 7.8 Hz, 4J13–15 = 1.5 Hz, H13), 8.65 (4H, d, 3J12–13
=
=
7.8 Hz, H12), 8.70–8.76 (10H, m, H15 and H5), 8.84 (6H, s, H9)
and 9.28 (6H, d, 3J5–6 = 6.5 Hz, H6); dC (75 MHz; DMSO-d6;
Me4Si; 25 1C) 16.9 (C1), 58.1 (C4), 118.02 (C9), 121.0 (C12),
124.7 (C14), 126.2 (C5), 128.9 (C2), 137.6 (C13), 143.7 (C3),
143.8 (C8), 144.8 (C6), 149.4 (C15), 152.8 (C7), 154.0 (C11) and
156.3 (C10); m/z (ESI-MS, API 3000) 363.5 (M3+ requires
363.4), 585.6 ([M + Br]2+ requires 585.1) and 1249.9
([M + 2Br]+ requires 1250.1).
4-{[40-(2,20:60,200-Terpyridinyl)-1-pyridiniumyl]methyl}toluene
.
bromide (L1 Br). L1 ꢀ Br was obtained from terpy,py (31 mg,
Electrochemical measurements were performed in a three-
electrode cell equipped with a 1 mm diameter platinum disk as
the working electrode, platinum wire as the counter electrode
and Ag+(0.01 M)/Ag as the reference electrode. The reference
potential was checked vs. ferrocene as recommended by
IUPAC (E1Fc = +86 mV). The supporting electrolyte was
tetrabutylammonium hexafluorophosphate (Fluka, puriss.)
and the solutions were deaerated by argon bubbling prior
each experiment. Cyclic voltammograms (CVs) were recorded
with a 600 CHInstruments potentiostat connected to a PC.
0.10 mmol) and 4-bromomethyltoluene (49.5 mg, 0.10 mmol)
following the procedure described for the synthesis of
L3A ꢀ Br3, in 82% yield (Found: C, 67.1; H, 4.5; Br, 16.5; N,
11.1. C28H23BrN4 requires C, 67.9; H, 4.7; Br, 16.1; N, 11.3%);
dH (300 MHz; DMSO-d6; Me4Si; 25 1C) 2.32 (3H, s), 5.89 (2H,
br s), 7.29 (2H, d), 7.5 (2H, d), 7.59 (2H, td), 8.1 (2H, td), 8.71
(2H, d), 8.77–8.81 (4H, m), 8.90 (2H, s), 9.34 (2H, d); m/z
(ESI-MS, HP 5989B) 415.6 (M+ requires 415.5).
.
.
[{TolylTerpyRu}3(L3A)] (PF6)9 (Ru3A (PF6)9). TolylTer-
pyRuCl3 (53.1 mg, 0.10 mmol) and L3A ꢀ Br3 (50 mg,
3.3 10ꢂ2 mmol) were refluxed overnight in a mixture of
EtOH–H2O–DMF (30 : 20 : 1 mL). The mixture was then
cooled to room temperature and an aqueous solution of
NH4PF6 (200 mg, 5 mL) was added. The resulting mixture
was concentrated under vacuum to ca. 25 mL and 60 mL of
water were added. The red solid obtained was filtered off and
washed with ethanol (3 ꢃ 10 mL) and diethyl ether (3 ꢃ 10
mL). Ru3A ꢀ (PF6)9 was recrystallized from acetonitrile, by
diffusion of diethyl ether, in 71% yield (Found: C, 46.9; H,
2.8; F, 27.4; N, 7.2; P, 8.0; Ru, 7.7. C153H114F54N21P9Ru3
requires C, 47.7; H, 3.0; F, 26.6; N, 7.6; P, 7.2; Ru, 7.9%); dH
(300 MHz; DMSO-d6; Me4Si; 25 1C) 2.53 (9H, s, H31), 6.10
(6H, br s, H7), 7.22–7.28 (6H, m), 7.36 (6H, t, J = 6.6 Hz),
7.49 (6H, d, J = 5.5 Hz), 7.60 (6H, d, J = 8.0 Hz), 7.65 (6H, d,
J = 5.5 Hz), 7.84–7.88 (6H, m), 8.06–8.18 (22H, m), 8.39 (6H,
d, J = 7.2 Hz), 9.07–9.21 (18H, m), 9.51 (6H, s), 9.70–9.75
Syntheses
All starting materials and solvents were purchased from
Aldrich Chemical Company and used without further purifi-
cation. 40-(4-pyridyl)-2,20:60,200-terpyridine (terpy,py),28 1,3,5-
tris[4-methylphenyl]benzene,29 1,3,5-tris[4-(bromomethyl)phe-
nyl]benzene,30 1,4-bis{[40-(2,20:60,200-terpyridinyl)-1-pyridiniu-
myl]methyl}benzene dibromide (L2),13 40-(4-tolyl)-2,20,60,200-
terpyridine (TolylTerpy) and TolylTerpyRuCl3,31 were synthe-
sized following the procedures previously reported without
modification.
1,3,5-Tris{[40-(2,20:60,200-terpyridinyl)-1-pyridiniumyl]methyl-
.
phenyl}benzene tribromide (L3A Br3). terpy,py (95 mg, 0.31
mmol) and 1,3,5-tris[4-(bromomethyl)phenyl]benzene (59 mg,
0.10 mmol) were suspended in acetonitrile (70 mL). 0.5 h at
reflux was needed to solubilize both reagents. The resulting
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1812 | New J. Chem., 2007, 31, 1806–1814

View Article Online
(11H, m); dC (75 MHz; DMSO-d6; Me4Si; 25 1C) 20.7, 20.9,
30.6, 35.7, 38.6, 38.9, 39.2, 39.5, 39.7, 40.0, 40.3, 63.0, 120.9,
122.0, 124.9, 125.0, 125.8, 127.0, 127.6, 128.1, 129.5, 130.0,
130.1, 133.0, 133.9, 138.2, 138.3, 138.9, 140.5, 140.8, 140.9,
145.7, 147.7, 151.8, 152.2, 152.3, 154.5, 155.0, 155.9, 157.5,
157.9; m/z (ESI-MS, API 3000) 283.6 ([M + PF6]8+ requires
283.4), 337.2 ([M + 2PF6]7+ requires 337.0), 497.9 ([M +
3PF6]6+ requires 497.6), 626.7 ([M + 4PF6]5+ requires 626.1),
819.0 ([M + 5PF6]4+ requires 818.9), 1140.6 ([M + 6PF6]3+
requires 1140.2).
([M + 3PF6]3+ requires 669.8), 1077.4 ([M + 4PF6]2+
requires 1077.1).
.
.
[(TolylTerpy)Ru(L1)] (PF6)3 (Ru1 (PF6)3). Ru1 ꢀ (PF6)3
was synthesized from TolylTerpyRuCl3 (26.6 mg, 5 ꢃ 10ꢂ2
mmol) and L1 ꢀ Br (63.7 mg, 5 ꢃ 10ꢂ2 mmol) following the
procedure described for the synthesis of Ru3A ꢀ (PF6)9, in 71%
yield (Found: C, 46.8; H, 3.1; F, 27.1; N, 7.4; P, 7.5; Ru, 7.6.
C50H40F18N7P3Ru requires C, 47.1; H, 3.2; F, 26.8; N, 7.7; P,
7.3; Ru, 7.9%); dH (300 MHz; DMSO-d6; Me4Si; 25 1C) 2.53
(6H, s, H30 and H17), 5.95 (2H, br s, H12), 7.25 (2H, t, 3J = 6.6
Hz, H21), 7.33–7.38 (4H, m, H15 and H3), 7.48 (2H, d, 3J = 5.4
Hz, H14), 7.55–7.61 (6H, m, H28, H4 and H11), 8.08 (2H, t,
3J = 7.7 Hz, H20), 8.15 (2H, t, 3J = 7.7 Hz, H2), 8.39 (2H, d,
.
.
[{TolylTerpyRu}3(L3B)] (PF6)9 (Ru3B (PF6)9). Ru3B ꢀ
(PF6)9 was synthesized by the reaction of TolylTerpyRuCl3
(53.1 mg, 0.10 mmol) and L3B ꢀ Br3 (44 mg, 3.3 ꢃ 10ꢂ2 mmol),
using the procedure described for the synthesis of
Ru3A ꢀ (PF6)9, in 73% yield (Found: C, 44.6; H, 3.3; F, 29.2;
N, 7.7; P, 7.9; Ru, 8.0. C138H108F54N21P9Ru3 requires C, 45.2;
H, 3.0; F, 28.0; N, 8.0; P, 7.6; Ru, 8.3%); dH (300 MHz;
DMSO-d6; Me4Si; 25 1C) 2.53 (9H, s, H28), 2.59 (9H, s, H1),
6.36 (6H, br s, H4), 7.15 (6H, t, 3J = 6.7 Hz, H14), 7.36 (6H, t,
3J = 6.3 Hz, H17), 7.45 (6H, d, 3J = 6.7 Hz, H15), 7.60 (6H, d,
3J = 8.1 Hz, H26), 7.65 (6H, d, 3J = 5.1 Hz, H16), 8.03 (6H, t,
3J = 7.8 Hz, H13), 8.13 (6H, t, 3J = 6.9 Hz, H18), 8.38 (6H, d,
3
3J = 8.1 Hz, H27), 9.07 (2H, d, J = 8.1 Hz, H1), 9.12–9.15
(4H, m, H10 and H19), 9.50 (2H, s, H24 or H7), 9.60 (2H, d, 3J
= 6.2 Hz, H18), 9.71 (2H, s, H24 or H7); dC (75 MHz; DMSO-
d6; Me4Si; 25 1C) 21.2 (C17), 21.4 (C30), 63.6 (C12), 121.4 (C24
or C7), 122.5 (C24 or C7), 125.3 (C1), 125.4 (C10 or C19), 126.3
(C10 or C19), 128.1 (C27 and C21), 128.6 (C3 or C15), 129.2 (C4,
C28 or C11), 130.4 (C3 or C15), 130.5 (C4, C28 or C11), 131.9
(C13), 133.5 (C23 or C6), 138.7 (C2), 138.8 (C20), 139.5 (C16),
139.6 (C29), 141.0 (C22), 146.0 (C18), 148.2 (C26), 152.2 (C23 or
C6), 152.7 (C14), 152.8 (C4, C28 or C11), 155.0 (C9), 156.4 (C5),
157.9 (C25), 158.4 (C8); m/z (ESI-MS, API 3000) 280.9
(M3+ requires 280.0), 492.5 ([M + PF6]2+ requires 492.5),
1130.4 ([M + 2PF6]+ requires 1129.9).
3
3J = 8.1 Hz, H25), 9.04 (6H, d, J = 6.9 Hz, H19), 9.08–9.15
3
(12H, m, H5 and H12), 9.33 (6H, d, J6–5 = 4.9 Hz, H6), 9.49
(6H, s, H22 or H9), 9.69 (6H, s, H22 or H9); dC (75 MHz;
DMSO-d6; Me4Si; 25 1C) 17.0 (C1), 20.9 (C28), 56.0 (C4), 120.9
(C9 or C22), 122.1 (C9 or C22), 123.3 (C27 or C10), 123.5 (C27 or
C10), 124.8 (C5 or C12), 124.9 (C19), 125.9 (C5 or C12), 127.6
(C25), 128.2 (C14), 129.1 (C17), 130.0 (C26), 132.9 (C2), 138.2
(C18), 138.3 (C13), 139.1 (C7), 140.5 (C24), 144.0 (C3), 144.3
(C6), 147.7 (C23), 152.1 (C16), 152.4 (C15), 154.5 (C21 or C8),
155.9 (C21 or C8), 157.4 (C20), 157.9 (C11); m/z (ESI-MS, API
3000) 313.9 ([M + PF6]8+ requires 313.7), 379.6 ([M +
2PF6]7+ requires 379.1), 467.0 ([M + 3PF6]6+ requires
466.4), 588.8 ([M + 4PF6]5+ requires 588.7), 772.7 ([M +
5PF6]4+ requires 772.1), 1078.5 ([M + 6PF6]3+ requires
1077.8).
Acknowledgements
M. Bourbin and N. Boccon, both students from the Ecole
Nationale Superieure de Chimie de Paris, are strongly ac-
´
knowledged for their participation in the electrochemical and
spectroscopic measurements during their training period in the
PPSM laboratory.
References
1 J. G. Vos and J. M. Kelly, Dalton Trans., 2006, 4869–4883.
2 C. Kaes, A. Katz and M. W. Hosseini, Chem. Rev., 2000, 100,
3553–3590.
.
.
[{TolylTerpyRu}2(L2)] (PF6)6 (Ru2 (PF6)6). Ru2 ꢀ (PF6)6
was obtained from TolylTerpyRuCl3 (53.1 mg, 0.10 mmol)
and L2 ꢀ Br2 (44 mg, 5 ꢃ 10ꢂ2 mmol) following the procedure
described for the synthesis of Ru3A ꢀ (PF6)9, in 71% yield
(Found: C, 45.6; H, 3.1; F, 28.5; N, 8.2; P, 7.9; Ru, 8.7.
C92H70F36N14P6Ru2 requires C, 45.2; H, 2.9; F, 28.0; N, 8.0;
P, 7.6; Ru, 8.3%); dH (300 MHz; DMSO-d6; Me4Si; 25 1C)
2.53 (6H, s), 6.00–6.07 (4H, br s, H12), 7.23 (4H, q, J = 6.1
Hz), 7.36 (4H, t, J = 6.2 Hz), 7.47–7.50 (6H, m), 7.60–7.67
(10H, m), 7.84 (2H, s), 8.04–8.18 (8H, m), 8.39 (4H, d, J = 8.1
Hz), 9.07–9.21 (12H, m), 9.50 (4H, s), 9.61–9.67 (4H, m), 9.73
(2H, d, J = 3.4 Hz); dC (75 MHz; DMSO-d6; Me4Si; 25 1C)
20.9 (C27), 30.6, 30.7, 35.7, 62.4, 62.8, 65.1, 71.0, 120.9, 122.0,
124.9, 125.0, 125.8, 127.2, 127.6, 128.1, 128.2, 128.5, 128.6,
129.6, 130.0, 132.7, 133.0, 133.8, 135.5, 138.2, 138.3, 138.9,
139.0, 140.2, 140.5, 144.0, 145.7, 147.7, 151.7, 151.9, 152.3,
154.5, 155.9, 157.5, 157.9, 162.2; m/z (ESI-MS, API 3000)
3 E. C. Constable, Chem. Soc. Rev., 2007, 36, 246–253.
4 H. Hofmeier and U. S. Schubert, Chem. Soc. Rev., 2004, 33,
373–399.
5 P. R. Andres and U. S. Schubert, Adv. Mater., 2004, 16,
1043–1068.
6 J. P. Sauvage, J. P. Collin, J. C. Chambron, S. Guillerez, C.
Coudret, V. Balzani, F. Barigelletti, L. D. Cola and L. Flamigni,
Chem. Rev., 1994, 94, 993–1019.
7 E. Baranoff, J.-P. Collin, L. Flamigni and J.-P. Sauvage, Chem.
Soc. Rev., 2004, 33, 147–155.
8 E. Figgemeier, L. Merz, B. A. Hermann, Y. C. Zimmermann, C. E.
Housecroft, H. J. Guntherodt and E. C. Constable, J. Phys. Chem.
B, 2003, 107, 1157–1162.
9 E. C. Constable, E. L. Dunphy, C. E. Housecroft, W. Kylberg, M.
Neuburger, S. Schaffner, E. R. Schofield and C. B. Smith, Chem.
Eur. J., 2006, 12, 4600–4610, and references therein.
10 C. R. Mayer, E. Dumas, A. Michel and F. Secheresse, Chem.
Commun., 2006, 4183–4185.
11 E. C. Constable, C. E. Housecroft, M. Neuburger, D. Phillips, P.
R. Raithby, E. Schofield, E. Sparr, D. A. Tocher, M. Zehnder and
Y. Zimmermann, J. Chem. Soc., Dalton Trans., 2000, 2219–2228.
12 E. C. Constable, Chem. Commun., 1998, 403–404.
13 D. G. Kurth, J. P. Lopez and W.-F. Dong, Chem. Commun., 2005,
2119–2121.
262.6 (M6+ requires 262.4), 343.9 ([M
requires 344.8), 466.5 ([M + 2PF6]4+ requires 466.1), 670.2
+
PF6]5+
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
New J. Chem., 2007, 31, 1806–1814 | 1813

View Article Online
14 C. R. Mayer, E. Dumas and F. Secheresse, Chem. Commun., 2005,
345–347.
23 R. J. Gale and R. A. Osteryoung, J. Electrochem. Soc., 1980, 127,
2167–2172.
15 C. R. Mayer, E. Dumas, F. Miomandre, R. Meallet-Renault, F.
Warmont, J. Vigneron, R. Pansu, A. Etcheberry and F. Secheresse,
New J. Chem., 2006, 30, 1628–1637.
16 W.-L. Jia, Y.-F. Hu, J. Gao and S. Wang, Dalton Trans., 2006,
1721–1728.
24 L. M. Vrana and K. J. Brewer, J. Photochem. Photobiol., A, 1997,
109, 201–211.
25 A. J. Bard and L. R. Faulkner, Electrochemical Methods: Funda-
mentals and Applications, John Wiley and Sons Inc., New York,
2001, ch. 6.
17 P. Belser, A. v. Zelewsky, M. Frank, C. Seel, F. Vogtle, L. D. Cola,
F. Barigelletti and V. Balzani, J. Am. Chem. Soc., 1993, 115,
4076–4086.
18 J. M. D. Wolf, R. Hage, J. G. Haasnoot, J. Reedijk and J. G. Vos,
New J. Chem., 1991, 15, 501–507.
26 M. Sharp, Electrochim. Acta, 1983, 28, 301–308.
27 E. C. Constable, C. E. Housecroft, E. R. Schofield, S. Encinas, N.
Armaroli, F. Barigelletti, L. Flamigni, E. Figgemeier and J. G.
Vos, Chem. Commun., 1999, 869–870.
28 E. C. Constable and A. M. W. C. Thompson, J. Chem. Soc.,
Dalton Trans., 1992, 2947–2950.
29 S. S. Elmorsy, A. Pelter and K. Smith, Tetrahedron Lett., 1991, 32,
4175–4176.
30 Q. He, H. Huang, J. Yang, H. Lin and F. Bai, J. Mater. Chem.,
2003, 13, 1085–1089.
19 K. Onitsuka, N. Ohara, F. Takei and S. Takahashi, Dalton Trans.,
2006, 3693–3698.
20 P. U. Maheswari, B. Modec, A. Pevec, B. Kozlevcar, C. Massera,
P. Gamez and J. Reedijk, Inorg. Chem., 2006, 45, 6637–6645.
21 A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and A.
von Zelewsky, Coord. Chem. Rev., 1988, 84, 85–277.
22 E. C. Constable, E. Figgemeier, C. E. Housecroft, J. Olsson and Y.
C. Zimmermann, Dalton Trans., 2004, 1918–1927.
31 J. P. Collin, S. Guillerez, J. P. Sauvage, F. Barigelletti, L. D.
Cola, L. Flamigni and V. Balzani, Inorg. Chem., 1991, 30,
4230–4238.
ꢁc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2007
1814 | New J. Chem., 2007, 31, 1806–1814