Synthesis, spectroscopic and electrochemical properties of mononuclear and dinuclear bis(bipy)ruthenium(II) complexes containing dimethoxyphenyl(pyridin-2-yl)-1,2,4-triazole ligands

Article abstract of DOI:10.1039/b108728m

The ligands HL1 and H₂L2 and the complexes [Ru(bipy)₂L1]PF₆·2H₂O 1, [(Ru(bipy)₂)₂L2](PF₆)₂·7H ₂O 2, { where HL1 = 3-(2′,5′-dimethoxyphenyl)-5-(pyridin-2″-yl)-1 H-1,2,4-triazole, H₂L2 = 1,4-bis(5′-(pyridin-2″-yl)- 1′H-1′,2′,4′-triazol-3′-yl)-2,5-dimethoxybenzene and bipy = 2,2′-bipyridyl}, have been prepared and characterised, by NMR, UV-vis and emission spectroscopies and by electrochemical measurements. X-Ray crystal structures of ligands HL1, H₂L2 and of the complex 1 are also reported. The dinuclear complex (2) exhibits a weak electronic interaction between the metal centres, which is modulated by the protonation state of the 1,2,4-triazole rings. The extent of the metal-metal interaction in these systems is compared with that observed in other pyridyl-1,2,4-triazole based dinuclear compounds of differing metal-metal distances.

Full text of DOI:10.1039/b108728m

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis, spectroscopic and electrochemical properties of
mononuclear and dinuclear bis(bipy)ruthenium(II) complexes
containing dimethoxyphenyl(pyridin-2-yl)-1,2,4-triazole ligands†
Paolo Passaniti,a Wesley R. Browne,b Fiona C. Lynch,b Donal Hughes,b Mark Nieuwenhuyzen,c
Paraic James,d Mauro Maestria and Johannes G. Vos*b
a Dipartimento di Chimica “G.Ciamician”, University of Bologna Via Selmi 2, 40126 Bologna,
Italy
b National Centre for Sensor Research, School of Chemical Sciences, Dublin City University,
Dublin 9, Ireland. E-mail: johannes.vos@dcu.ie
c School of Chemistry, Queen’s University of Belfast, Belfast, Northern Ireland, UK BT9 5AG
d National Institute for Cellular Biotechnology, Dublin City University, Dublin 9, Ireland
Received 25th September 2001, Accepted 12th February 2002
First published as an Advance Article on the web 26th March 2002
The ligands HL1 and H2L2 and the complexes [Ru(bipy)2L1]PF6ؒ2H2O 1, [(Ru(bipy)2)2L2](PF6)2ؒ7H2O 2, {where
HL1 = 3-(2Ј,5Ј-dimethoxyphenyl)-5-(pyridin-2Љ-yl)-1H-1,2,4-triazole, H2L2 = 1,4-bis(5Ј-(pyridin-2Љ-yl)-1ЈH-1Ј,2Ј,4Ј-
triazol-3Ј-yl)-2,5-dimethoxybenzene and bipy = 2,2Ј-bipyridyl}, have been prepared and characterised, by NMR,
UV–vis and emission spectroscopies and by electrochemical measurements. X-Ray crystal structures of ligands HL1,
H2L2 and of the complex 1 are also reported. The dinuclear complex (2) exhibits a weak electronic interaction
between the metal centres, which is modulated by the protonation state of the 1,2,4-triazole rings. The extent of the
metal–metal interaction in these systems is compared with that observed in other pyridyl-1,2,4-triazole based
dinuclear compounds of differing metal–metal distances.
Due to its role as redox relay in biological systems, the
hydroquinone/quinone redox couple is of particular interest. In
Introduction
Ruthenium() polypyridyl complexes are of interest for their
an earlier study on ruthenium polypyridyl complexes incorpor-
spectroscopic, photophysical, photochemical, and electro-
ating pyridyltriazole ligands with pendent hydroquinone group-
chemical properties.1–3 These properties are of particular use
ings an electrochemically induced intramolecular protonation
in the construction of supramolecular systems and photo-
of the ruthenium centre was observed upon oxidation of the
chemically driven molecular devices.4,5 Ruthenium() poly-
hydroquinone.13 We are presently involved in a systematic study
pyridyl complexes have also received extensive attention as
of this unusual observation and also of the photophysical
models for photo-system II and in the catalytic photochemical
properties of molecular dyads of this type. In the present
contribution, the spectroscopic and electrochemical properties
of the mono-nuclear complex 1 based on the Ru(bipy)2-
complex of 3-(2Ј,5Ј-dimethoxyphenyl)-5-(pyridin-2Љ-yl)-1H-
1,2,4-triazole (HL1) and the dinuclear complex 2 based on the
Ru(bipy)2-complex of 1,4-bis(5Ј-(pyridin-2Љ-yl)-1ЈH-1Ј,2Ј,4Ј-
triazol-3Ј-yl)-(2,5-dimethoxybenzene) (H2L2) are reported
(bipy = 2,2Ј-bipyridyl). The electrochemical, spectroelectro-
chemical and photophysical properties of the compounds are
compared with those reported for analogous mononuclear
and dinuclear complexes based on 3-(pyridin-2Ј-yl)-1H-1,2,4-
triazole (Hpytr) (3) and 3,5-bis(pyridin-2Ј-yl)-1H-1,2,4-triazole
(Hbpt) (4). For structures of complexes 1–4 see Fig. 1. The
compounds reported are synthetic precursors to the analogous
cleavage of water.6
Of particular interest is the design of multinuclear structures
capable of directing and modulating electron and energy trans-
fer processes.4,5 In this regard many studies of ruthenium
complexes covalently bound to electron acceptors or donors
have been reported. The general approach taken in electron
transfer studies has been to bind the electron acceptor/donor to
the ruthenium polypyridyl centre via the polypyridyl ligands.7–9
However, since the excited state in such compounds is normally
based on these ligands a strong coupling between the metal
centre and the electron donor/acceptor is usually observed,
which precludes a long-lived charge separation.9,10 In our
laboratories we have taken a different approach aimed at weak-
ening the direct electronic coupling between the electron donor/
hydroquinone/quinone compounds and serve as model com-
acceptor and the sensitiser, consequently slowing down the
pounds. The results obtained for the hydroquinone/quinone
back reaction and achieving a longer lived charge separated
analogues will be reported in a further publication.
state.11 This approach is based on the attachment of groupings
to spectator ligands, such as 3-(pyridin-2Ј-yl)-1,2,4-triazoles.
In addition, the pH dependent properties of the 1,2,4-triazole
ligands are shown to be useful in applications such as molecular
switches.12
Experimental
Materials
All solvents employed were of HPLC grade or better and used
as received unless otherwise stated. For all spectroscopic
measurements Uvasol (Merck) grade solvents were employed.
All reagents employed in synthetic procedures were of reagent
grade or better. Cis-[Ru(bipy)2Cl2]ؒ2H2O,14 (pyridin-2-yl)-
† Electronic supplementary information (ESI) available: figures show-
ing the molecular structures and intermolecular interactions for HL1
and H2L2; 1H COSY NMR spectrum of 2. See http://www.rsc.org/
suppdata/dt/b1/b108728m/
1740
J. Chem. Soc., Dalton Trans., 2002, 1740–1746
This journal is © The Royal Society of Chemistry 2002
DOI: 10.1039/b108728m

View Article Online
1,4-Bis(5Ј-(pyridin-2Љ-yl)-1ЈH-1Ј,2Ј,4Ј-triazol-3Ј-yl)-2,5-
dimethoxybenzene (H2L2). Diethyl-2,5-dihydroxyterephthalate
(5.09 g, 20 mmol) and potassium carbonate (13.8 g, 100 mmol)
were stirred with 100 cm3 dry acetone in a flask fitted with reflux
condenser and CaCl2 drying tube. Iodomethane (14.2 g,
100 mmol) was added and the mixture heated to reflux for 48 h.
The reaction was followed by TLC (mobile phase: pet. ether
(40/60) : diethyl ether 2 : 1 v/v). On cooling, the white precipi-
tate was filtered and washed with acetone. Yield of diethyl-
2,5-dimethoxyterephthalate: 5.36 g (95%). 1H NMR (d1-chloro-
form): δ 7.36 (1H, s), 4.36 (2H, q, 7.4 Hz), 3.87 (3H, s), 1.37
(3H, t, 7.4 Hz).
Diethyl-2,5-dimethoxyterephthalate (5 g, 18 mmol) was
added to 5 g (9 mmol) of potassium hydroxide in 15 cm3 ethyl-
ene glycol, and heated at 100 ЊC until the mixture became
homogenous. The ethanol formed during heating was removed
by distillation at reduced pressure. The reaction mixture was
then cooled, H2O (20 cm3) was then added and the reaction
mixture was made slightly acidic by the addition of 20% H2SO4,
which resulted in a white precipitate of 2,5-dimethoxytere-
phthalatic acid forming. This was filtered and washed with
water (10 cm3) and then acetone (5 cm3). The 2,5-dimethoxy-
terephthalatic acid was dried in a Kugelrohr apparatus in vacuo
1
over P2O5 at 100 ЊC. Yield: 3.9 g (94%), mp 265 ЊC. H NMR
(d6-dimethylsulfoxide): δ 7.28 (1H, s), 3.77 (3H, s).
Phosphorous pentachloride (6.25 g, 30 mmol) was added to
2,5-dimethoxyterephthalic acid (3.5 g, 15 mmol) and the solids
stirred with gentle heating. After evolution of gas had ceased,
the phosphoryl oxychloride was removed by distillation at
normal pressure and then under reduced pressure. The 2,5-
dimethoxyterephthaloyl dichloride (3.95 g, 15 mmol) formed
was dissolved in 50 cm3 tetrahydrofuran and added with stirring
to a solution of (pyridin-2-yl)amidrazone (4.76 g, 35 mmol) and
sodium carbonate (1.59 g, 15 mmol) in 30 cm3 of dry tetra-
hydrofuran at 0 ЊC. After stirring for 24 h, the reaction contents
were added to 100 g of ice water and stirred for 30 min. The
yellow solid was filtered, washed with water and acetone, and
dried under vacuum. Impurities were removed by heating the
crude product in dimethylsulfoxide, filtering and washing with
acetone (10 cm3). Yield of 1,4-bis(acylpyridin-2Ј-ylamidra-
zone)-2,5-dimethoxybenzene: 5 g (10 mmol, 67%), mp 189 ЊC.
1H NMR (d6-dimethylsulfoxide): δ 10.21 (2H, s, –NH), 8.61
(2H, d, 5.6 Hz, H6), 8.17 (2H, d, 7.4 Hz, H3), 7.76 (2H, t,
7.4 Hz, H4), 7.50(2H, t, 7.4 Hz, H5), 7.38 (2H, s, Benz-H),
6.81(4H, s, –NH2), 4.03 (6H, s, –OCH3).
1,4-Bis(acylpyridin-2Ј-ylamidrazone)-2,5-dimethoxybenzene
(5 g, 10 mmol) was heated to reflux in ethylene glycol until the
reaction mixture turned to a clear brown colour. The product
(H2L2), a tan solid, precipitated from solution overnight and
was recrystallised from dimethylsulfoxide. Yield of H2L2: 2.3 g
(5.4 mmol, 54%). 1H NMR (d6-dimethylsulfoxide): δ 14.11 (2H,
broad s, –NH), 8.70 (2H, broad s, H6), 8.18 (2H, d, H3),
7.94 (4H, broad s, H4 and Benz-H); 7.47 (2H, s, H5); 4.03
(6H, broad s, –OCH3). Elemental analysis: Found (Calc. for
C22H18N8O2ؒ1H2O) C 59.46 (59.46), H 4.20 (4.50), N 25.42
(25.23)%.
Fig. 1 Structures of compounds 1–4.
amidrazone and tetraethylammoniumperchlorate (TEAP)15
were prepared by previously reported procedures.
Syntheses
3-(2Ј,5Ј-Dimethoxyphenyl)-5-(pyridin-2Љ-yl)-1H-1,2,4-triazole
(HL1). 2,5-Dimethoxybenzoic acid (10 g, 55 mmol) and PCl5
(11.45 g, 55 mmol) were reacted together to form a clear
yellow product. The POCl3 formed was removed by distil-
lation with slight heating (40 ЊC) under a weak vacuum
leaving a clear yellow liquid, which crystallised overnight
under vacuum. The acid chloride (11.03 g, 50 mmol) formed
was dissolved in 30 cm3 of dry tetrahydrofuran and added
dropwise to (pyridin-2-yl)amidrazone (6.85 g, 50 mmol)
and sodium carbonate (2.65 g, 25 mmol) in 40 cm3 of tetra-
hydrofuran, resulting in immediate precipitation of the pale
yellow acylamidrazone. The reaction mixture was stirred for
a further 4 h and the yellow product collected by vacuum
filtration and stirred in ice water for 30 min followed by
recovery using vacuum filtration. The yellow product was dried
overnight at 80 ЊC. Yield of 1-(2Ј,5Ј-dimethoxyphenyl)-4-
(pyridin-2Љ-yl)acylamidrazone: 10.95 g, (73%), mp 190 ЊC. The
amidrazone (6 g, 20 mmol) was dissolved in a minimum of
ethylene glycol and heated at reflux for 30 min. On cooling, the
ligand HL1 precipitated from solution and was collected by
filtration. The product was decolourised over charcoal and
crystallised from dichloromethane. Yield: 3.4 g, 60%. 1H NMR
(d6-acetone): δ 12.54 (1H, s), 8.69 (1H, d, 3.9 Hz), 8.20 (1H, d,
7.9 Hz), 7.87 (1H, t, 7.9 Hz), 7.82 (1H, s, 3.0 Hz), 7.39 (1H,
t, 4.9 Hz), 7.12 (1H, d, 9.9 Hz), 7.04 (1H, dd, 3.0 Hz, 8.9
Hz), 3.99 (3H, broad s), 3.83 (3H, s). Elemental analysis:
Found (Calc. for C15H14N4O2) C 63.62 (63.83), H 5.15 (4.96),
N 18.74 (19.86)%.
[Ru(bipy)2(L1)]PF6ؒ2H2O (1). Cis-[Ru(bipy)2Cl2]ؒ2H2O (370
mg, 0.71 mmol) and HL1 (200 mg, 0.71 mmol) were heated at
reflux for 8 h in 150 cm3 EtOH–H2O (2 : 1 v/v). The hot solution
was filtered and evaporated to dryness after which 10 cm3 of
water was added to the dark red product. 1 was precipitated
with an excess aqueous solution of NH4PF6. The product was
purified by column chromatography with activated neutral
alumina (Brockmann 1, std grade, 150 mesh, Aldrich) and
acetonitrile as eluent. The product obtained was recrystallised
from acetone–H2O (with 1 drop of conc. ammonia). Single
crystals for X-ray structure determination were grown from an
acetone solution of the product. Yield: 495 mg (82%). 1H NMR
(d3-acetonitrile). δ 8.45 (2H, t, 8.9 Hz), 8.38 (2H, t, 8.9 Hz), 8.08
J. Chem. Soc., Dalton Trans., 2002, 1740–1746
1741

View Article Online
Table 1 X-Ray experimental crystal data for HL1, H2L2 and 1
HL1
H2L2
1
Empirical formula
M/g molϪ1
T /K
C15H14N4O2
282.30
293(2)
C22H18N8O2
426.44
153(2)
C38H35F6N8O3PRu
897.78
293(2)
Wavelength/Å
Crystal system, space group
a/Å, α/Њ
1.54178
0.71073
1.54178
Triclinic, P1
¯
Monoclinic, P21/n
10.664(2), 90
17.689(7), 96.96
15.057(3), 90
2819.4(14), 8
0.85 × 0.19 × 0.12
0.754
Monoclinic, P21/c
9.235(3), 90
11.636(3), 106.44
10.096(3), 90
1040.5(5), 2
0.76 × 0.21 × 0.30
0.093
11.704(7), 69.92
12.915(7), 88.60
15.631(9), 63.14
1956(2), 2
0.32 × 0.11 × 0.15
4.358
b/Å, β/Њ
c/Å, γ/Њ
V/Å3, Z
Crystal dimensions/mm
µ/mmϪ1
No. reflections collected
3763
1853
7006
Independent reflec. (Rint
)
3540 (0.0549)
0.065 (0.1599)
0.1103 (0.1952)
1359 (0.0528)
0.0610 (0.0991)
0.1398 (0.1258)
4917 (0.0501)
0.0655 (0.1639)
0.0761 (0.1724)
Final R indices [I > 2σ(I )] R1 (wR2)
R indices all data R1 (wR2)
(1H, d, 7.9Hz, H3), 7.95 (7H, m, containing H4), 7.79 (2H, t,
4.9 Hz), 7.51 (1H, d, 5.9 Hz, H6), 7.42(1H, t, 6.9 Hz), 7.38 (2H,
t, 5.9 Hz), 7.27 (1H, t, 5.9 Hz), 7.20 (1H, d, 2.9 Hz, H3Љ), 7.12
(1H, t, 6.9 Hz, H5), 6.93 (1H, d, 8.9 Hz, H6Љ), 6.83 (1H, dd, 2.9
Hz, 8.9, H5Љ), 3.71(3H, s, –OCH3 1Љ), 3.61(3H, s, –OCH3 4Љ).
Elemental analysis: Found (Calc. for C35H33N8O4PF6Ru) C 48.2
(48.0), H 4.0 (3.8), N 12.4 (12.8)%.
uncorrected for photomultiplier response. 1 cm path length
quartz cells were used for recording spectra. Luminescence
lifetime measurements were obtained using an Edinburgh
Analytical Instruments (EAI) Time Correlated Single Photon
Counting apparatus (TCSPC) comprising of two model J-yA
monochromators (emission and excitation), a single photon
photomultiplier detection system model 5300, and a F900
nanosecond flashlamp (N2 filled at 1.1 atm pressure), interfaced
with a personal computer via a Norland MCA card. Data
correlation and manipulation was carried out using EAI F900
software version 5.1.3. Emission lifetimes were calculated using
a single exponential fitting function (Edinburgh instruments
F900 software). pH titrations of 1 and 2 were carried out in
Britton–Robinson buffer (0.04 M H3BO3, 0.04 M H3PO4, 0.04
M CH3CO2H) (pH was adjusted using concentrated sulfuric
acid or sodium hydroxide solution). The appropriate isosbestic
point from the absorption spectra was used as the excitation
wavelength for emission titrations.
[(Ru(bipy)2)2(L2)](PF6)2ؒ7H2O (2). 2 was prepared in a simi-
lar manner to 1, except that 0.25 g (0.48 mmol) of cis-
[Ru(bipy)2(Cl)2]ؒ2H2O was heated at reflux with 0.1 g (0.24
mmol) of H2L2. Yield: 0.32 g (89%). 1H NMR (d3-acetonitrile):
δ 8.47 (2H, d, 7.4 Hz), 8.68 (2H, d, 7.4Hz), 8.59 (4H, t, 7.4 Hz),
8.29 (2H, d, 7.4 Hz, H3), 8.07 (8H, m), 7.96 (10H, m, H4), 7.71
(2H, d, 5.5 Hz, H6), 7.50 (8H, m, Benz-H), 7.37 (2H, t, 7.4 Hz),
7.24 (2H, t, 7.4 Hz, H5), 3.54 (6H,s,–OCH3). Elemental
analysis: Found (Calc. for C62H62N18O9P2F12Ru2) C 44.7
(44.65), H 3.4 (3.72), N 13.1 (13.44)%.
Elemental analysis was performed at the Microanalytical
Laboratory at University College Dublin.
Electrochemical and spectroelectrochemical measurements
Electrochemical measurements were carried out on a Model
660 Electrochemical Workstation (CH Instruments). Typical
complex concentrations were 0.5 to 1 mM in anhydrous
acetonitrile (Aldrich 99.8%) containing 0.1 M TEAP. A Teflon
shrouded glassy carbon working electrode, a Pt wire auxiliary
electrode and SCE reference electrode were employed. Solu-
tions for reduction measurements were deoxygenated by
purging with N2 or Ar gas for 15 min prior to the measure-
ment. Measurements were made in the range of Ϫ2.0 to 2.0 V
(w.r.t SCE electrode). Protonation of complexes was achieved
by addition of 0.1 M trifluoroacetic acid (in acetonitrile) to the
electrolyte solution. The scan rates used were typically 100 or
200 mV sϪ1. Spectroelectrochemistry was carried out using an
OTTLE set-up comprising of a home made Pyrex glass, thin
layer cell (1 mm pathlength). The optically transparent working
electrode was made from platinum–rhodium gauze, the counter
electrode used was a platinum wire, and the reference electrode
was a pseudo Ag/AgCl reference electrode. 0.1 M TEAP in
anhydrous acetonitrile was used as electrolyte. The working
electrode was held at the required potential throughout the
measurement using an EG&G PAR Model 363 potentiostat.
Absorption and emission spectra were recorded as described
above.
X-Ray crystallography
Data for HL1, H2L2 and 1 were collected on a Siemens P4
diffractometer using the XSCANS16 software with graphite
monochromated Mo-Kα radiation (Table 1) Relevant experi-
mental data is presented in Table 1.
The structures were solved using direct methods and refined
with the SHELXTLPC and SHELXL-93 program packages17
and the non-hydrogen atoms were refined with anisotropic
thermal parameters.
CCDC reference numbers 171755–171757.
See http://www.rsc.org/suppdata/dt/b1/b108728m/ for crys-
tallographic data in CIF or other electronic format.
NMR spectroscopy
1
1H and H COSY spectra were recorded on a Bruker AC400
(400 MHz) NMR spectrometer. All measurements were carried
out in d6-dimethylsulfoxide, d1-chloroform or d6-acetone for
ligands and d3-acetonitrile for complexes. Peak positions are
relative to residual solvent peaks.
Photophysical measurements
UV-vis absorption spectra were recorded on a Shimadzu
UV.vis-NIR 3100 spectrophotometer interfaced with an
Elonex-466 PC using UV-vis data manager software. Emis-
sion spectra were recorded using a LS50-B Luminescence
spectrophotometer, equipped with a red sensitive Hamamatsu
R928 PMT detector, interfaced with an Elonex-466 PC
using Windows based fluorescence data manager software.
Emission/excitation slit widths were 5 nm. Emission spectra are
Results and discussion
As noted in the introduction, the ultimate aim of our investi-
gations is the study of metal complexes having pendent hydro-
quinone or quinone groupings. In earlier studies hydroquinone
type ligands such as the hydroquinone analogue of HL1,
3-(2Ј,5Ј-dihydroxyphenyl)-5-(pyridin-2Љ-yl)-1,2,4-triazole were
1742
J. Chem. Soc., Dalton Trans., 2002, 1740–1746

View Article Online
prepared directly using 2,5-dihydroxybenzoic acid.13 With this
approach however, low yields were generally obtained. For this
reason a new synthesis has been developed based on methoxy
precursors with subsequent deprotection of these groups to
form the required hydroquinone complex. In this contribution
we report on the synthesis and properties of these dimethoxy-
‘protected’ complexes. The synthesis of the ligands HL1 and
H2L2 and complexes 1 and 2 (Fig. 1) have been carried out by
modification of general procedures.18 In this manner dimethoxy
precursors can be obtained in high yield.
dimensional network via π ؒ ؒ ؒ π interactions and hydrogen
bonding involving the bipy and pyridyltriazole ligands. The
packing of these dimers creates cavities within the lattice in
which the PF6 anions and acetone molecules are located.
1H NMR spectroscopy
1H NMR spectra have been used extensively to determine the
coordination mode of pyridyltriazoles and other types of
chelating ligands.21 The spectra obtained for 1 and 2 are shown
in Fig. 4. Due to the complexity of the 1H NMR spectra of the
X-Ray crystallography
The molecular structures obtained for HL1 and H2L2 confirm
the structural features of the two ligands. Figures showing the
molecular structure and intermolecular interactions are given
as ESI.† The molecular structure of 1 is shown in Fig. 2. In the
Fig. 4 1H NMR spectra of 1 (upper spectrum) and 2 (lower spectrum)
in d3-acetonitrile (400 MHz). –OCH3 signals are shown as insets.
1
complexes H COSY NMR spectroscopy (see ESI†) has been
performed of both ligands and complexes and the spectra have
been assigned by comparison with data previously reported for
similar ligands and complexes.21 It is worth pointing out that
for 1 two non-equivalent methoxy signals are observed at 3.61
and 3.74 ppm. The spectrum obtained for 2 is strikingly simple.
The dinuclear compound obtained is clearly highly symmetric
as indicated by the presence of only one methoxy signal at 3.64
ppm and by the fact that the phenyl protons are equivalent and
found as a singlet at 7.40 ppm. In contrast for the mononuclear
complex 1 the three phenyl protons are all non-equivalent and
appear at 6.83, 6.93 and 7.20 ppm. The crystal structure
obtained for 1 (Fig. 2) shows that the coordination mode of the
triazole ring in this compound is via the N2 nitrogen atom. For
3 it has been shown previously that the 1H NMR spectra of the
N2 and N4 bound coordination isomers are significantly differ-
Fig. 2 Molecular structure and labelling scheme for 1.
X-ray analysis of 1 the protons on one of the methoxy groups
are disordered. Complex 1 cocrystallised with a molecule of
acetone and a hexafluorophosphate counter anion. From the
crystal structure it is clear that the ligand is bound through the
pyridine-N and N2 of the triazole ring (or via N(25) and
N(32)). The bite angle of the N(25)–Ru(1)–N(32) is 77.8(2)Њ,
which corresponds well with the bite angle of 77.9(1)Њ obtained
by Hage et al. for [Ru(bipy)2(3-(2-hydroxy-phenyl)-5-(pyridin-
2-yl)-1,2,4-triazole)]PF6ؒCH3COCH3.19 Bite angles of 79.5(3)
and 78.8(3)Њ for bipyridine ligands and Ru–N distances of
2.033(7)–2.098(7) Å are also comparable to those found in
other complexes.19,20 Ru(1)–N(25) at 2.098(7) Å is the longest
Ru–N bond in the complex. One factor contributing to this
increased length is limited π-backbonding to the π-electron rich
1,2,4-triazole ring from the metal centre. The intermolecular
X-ray structure is dominated by interaction between two C(3)–
H atoms and the N2 nitrogen of the triazole ring forming a
hydrogen bonded dimer (Fig. 3). The dimers form a three-
1
ent.21 The symmetric H NMR of 2 together with steric con-
siderations and its similarity with the spectrum of 1, point
therefore strongly to a N2 coordination of both the 1,2,4-
triazole rings in the dinuclear species.
Photophysical and acid–base properties
The photophysical data obtained for 1 and 2 are summarised in
Table 2, together with data for [Ru(bipy)3]2ϩ and complexes 3
and 4 for comparison. For both 1 and 2, the absorption and
emission maxima are red-shifted with respect to [Ru(bipy)3]2ϩ 1
but are in agreement with those reported for 3 and 4.21 This is
not unexpected as π-electron rich 1,2,4-triazole containing
ligands such as HL1 and H2L are stronger σ-donors and weaker
π-acceptors than 2,2Ј-bipyridine and destabilise the filled metal
d-orbitals and hence raise the energy of the ground state metal
based orbitals.21 On protonation of both 1 and 2 a blue shift of
both the lowest energy absorption bands and the emission
maxima are observed, as would be expected due to the stabilis-
ation of the filled metal d-orbitals by the reduction in the
σ-donor strength of the protonated triazole ligand. In addi-
tion, on protonation of both 1 and 2 the emission lifetime
decreases to the sub-nanosecond timescale. This decrease (in
Fig. 3 Diagram of the hydrogen bonded dimer for 1.
J. Chem. Soc., Dalton Trans., 2002, 1740–1746
1743

View Article Online
Table 2 Spectroscopic and redox data for complexes described in text
Abs. λ max/
Em. λ max
298 K/nm
bLifetime at
298 K/ns
cRuII/RuIII oxid.
(ligand)/V
nm (ε/10Ϫ3 MϪ1 cmϪ1
)
Ligand based red./V
a[Ru(bipy)3]2ϩ
452
608
685
612
683
612
650
620
648
1000
110
—
105
—
145
—
100
1.26
Ϫ1.33, Ϫ1.55, Ϫ1.8
Ϫ1.48, Ϫ1.76
Not measured
Ϫ1.48, Ϫ1.73
Ϫ1.49, Ϫ1.73
Ϫ1.47, Ϫ1.72, Ϫ2.25
Ϫ1.49, Ϫ1.73, Ϫ2.25
Ϫ1.40, Ϫ1.62, Ϫ1.67
1 [Ru(bipy)2(L1)]ϩ
[Ru(bipy)2(HL1)]2ϩ
2 [(Ru(bipy)2)2(L2)]2ϩ
[(Ru(bipy)2)2HL2]4ϩ
3 [Ru(bipy)2(pytr)]ϩ
[Ru(bipy)2(pytr)]2ϩ
d4 [(Ru(bipy)2)2(bpt)]3ϩ
485 (10.7) 290 (80.7)
441 (15.3) 286 (86.8)
481 (18.9) 290 (148)
412 (28.3) 285 (153)
465 (11.0)
437 (12.9)
453 (22.6)
0.80 (1.20, 1.40)
1.2 (1.45)
0.82 (1.26, 1.45)
1.25 (1.5)
0.83
1.14
1.04, 1.34
a From1. b Lifetimes measured in deaerated acetonitrile at 298 K (lifetimes for protonated species are sub-nanosecond and were not measured). c All
redox potentials are vs. SCE in 0.1 M TEAP–acetonitrile, 100 mV sϪ1 scan rate. d Value obtained from ref. 19.
contravention of the energy gap law22) is a result of the reduc-
tion in the 3MLCT Ϫ 3MC gap, which facilitates fast thermally
activated radiationless deactivation of the emissive 3MLCT
3
excited state via population of the MC state.21 The acid–base
properties of 1 and 2 (pKa = 4.0 and 4.1, respectively) have been
determined using absorption spectroscopy by titration both in
Britton–Robinson buffer and in acetonitrile and are similar to
those reported previously for 3 (pKa = 4.07).21 In the case of
complex 2, a two-step protonation process is possible however
only a single protonation step is observed in aqueous media in
the range of pH 1.5 to 10 (Fig. 5). Titration in acetonitrile
Fig. 6 Cyclic voltammogram of 1 showing reversibility of RuII/RuIII
redox couple and subsequent irreversible ligand oxidation couples (in
0.1 M TEAP–acetonitrile, 100 mV sϪ1 scan rate, ref. SCE).
higher oxidation potential for both 1 and 2, which arises
from the reduced σ-donor capacity of the protonated triazole
moieties. Comparison of the first oxidation potential of 2 with
those of 1, 3 and 4, indicates that Ru(bipy)2(1,2,4-triazole)-
units of the dimer behave independently and have properties
similar to monomeric species and are quite different from those
observed for the dinuclear complex 4.
The extent of interaction between metal centres in poly-
nuclear complexes has been classified by Robin and Day as
Type I, II or III.23 In dinuclear metal complexes, the metal
redox properties are useful in investigating electronic inter-
action between metal subunits. In principle, because two
redox-active sites are present (the two metal centres), two metal-
centred oxidation processes are possible, and their separation
(∆E) is related to the stability of the mixed-valence species by
the comproportionation constant, Kc, as defined in eqn. 1. The
Fig. 5 Titration of 2 in Britton–Robinson buffer pH (a) 1.84; (b) 2.50;
(c) 2.84; (d) 3.16; (e) 3.62; (f ) 3.92; (g) 4.29; (h) 4.79; (i) 10.06 (inset; pH
titration plot monitored at 415 nm).
shows complete protonation with two equivalents of trifluoro-
acetic acid, indicating that in aqueous media both protonation
steps occur at essentially the same pH.
Kc = exp(∆E/25.69)
(1)
Electrochemical and spectroelectrochemical properties
absence of a measurable difference (∆E < 40 mV) in the two
metal-based oxidations of 2 indicates that little or no com-
munication between the metal centres exists, and the value of Kc
for 2 is therefore reported to be less than 5.24
Redox properties of 1 and 2 are presented in Table 2. E1/2 values
were determined by cyclic voltammetry (CV). For each of the
deprotonated complexes a single reversible redox wave was
observed at low oxidation potentials followed by two quasi-
reversible oxidation waves at higher potentials. The reversible
peaks at 0.8 (1) and 0.82 V (2) (vs. SCE) are attributed to oxid-
ation of the metal centre (vide infra), while the quasi-reversible
peaks at 1.2 (1) and 1.26 V (2) are assigned to the first oxidation
of the dimethoxyphenyl grouping (Fig. 6). No clear electro-
chemistry was obtained for the free dimethoxy ligand, most
likely because of adsorption on the electrode surface. The metal
oxidation potentials are considerably lower (∼300 mV) than for
[Ru(bipy)3]2ϩ, as is expected for the stronger σ-donating 1,2,4-
triazoles. The larger peak-to-peak separation (Ep) in the first
oxidation wave of 2 compared with 1 reflects the bi-electronic
process involved in the metal oxidation (RuIIRuII to RuIIIRuIII).
For the protonated complexes all redox waves are observed at a
However, since fast electron transfer between metal centres
can occur with an electronic coupling as low as tens of wave-
numbers,10 the absence of two resolvable redox waves does not
automatically indicate that the coupling between the metal
centres is absent. The most direct measure of the metal–metal
electronic coupling in a mixed valence system, RuIIRuIII, can be
obtained from intervalence transitions (IT).25,26 For this reason
the spectroelectrochemical investigations have been carried out.
The electronic properties of partially and completely oxidised
complexes, 1 and 2, together with data from model systems are
presented in Table 3.
Applying a potential of 0.8 V (vs. pseudo Ag/AgCl refer-
ence electrode) to a basic acetonitrile solution of 1 results in
1744
J. Chem. Soc., Dalton Trans., 2002, 1740–1746

View Article Online
Table 3 Absorption bands observed for complexes 1, 2 and 4 in RuII, RuIIRuIII (mixed valence) and RuIII oxidation states
Complex
RuII MLCT bands/nm (cmϪ1
)
RuIIRuIII IT bands/nm (cmϪ1
)
RuIII LMCT bands/nm (cmϪ1
)
a[Ru(bipy)3]2ϩ
452 (22075)
485 (20620)
440 (22725)
482 (20745)
412 (24270)
453 (22075)
—
—
—
675 (14810)
1075 (9300)
825 (12120)
1220 (8220)
840 (11930)
753 (13280)
1 [Ru(bipy)2(L1)]ϩ
[Ru(bipy)2(HL1)]2ϩ
2[(Ru(bipy)2)2(L2)]2ϩ
[(Ru(bipy)2)2(H2L2)]2ϩ
4 [(Ru(bipy)2)2(bpt)]3ϩ
1540 (6470)
—
1800 (5556)
a Value obtained from ref. 27
depletion in the MLCT band at 485 nm, coupled with the
formation of a new band at 1075 nm. This band is comparable
to the ligand to metal charge transfer bands (LMCT) observed
for other [Ru(LL)3]3ϩ complexes (where LL = substituted-2,2-
bipyridine).27 Upon increasing the oxidation potential to 1.3 V,
the depletion of the MLCT band continues, while the band at
1075 nm also decreases in intensity. The depletion of the MLCT
bands at 485 nm and the appearance of the LMCT band at
0.8 V is consistent with the oxidation of the metal centre. Fur-
ther increase of the oxidation potential to 1.3 V is irreversible
and results in decomposition of the complex in agreement with
data obtained from cyclic voltammetry (vide supra). The revers-
ibility of the initial oxidation process has been confirmed by
reformation of the initial spectrum upon bulk electrolysis at
0.3 V, subsequent to electrolysis at 0.8 V.
energy, indicating irreversible ligand oxidation of 2. The nature
of the product formed was not further investigated.
Under acidic conditions, at which the triazole ring is proton-
ated, no IT band is observed for 2. The LMCT bands observed
are shifted to higher energy compared with the LMCT bands
observed for the deprotonated complexes, from 1075 nm to 825
nm for 1 and from 1220 nm to 840 nm for 2. The intensities of
these bands relative to the LMCT bands of the deprotonated
complexes is also much reduced. These observations are not
unexpected and can be rationalised in terms of the relative
energies of the donor (ligand based) and acceptor (metal based)
orbitals. The effect of protonation and the subsequent decrease
in the σ-donor properties of the bridging ligand is to increase
the energy of both the donor and acceptor orbitals leading
to the observed shift in the LMCT transition. In addition
upon protonation there is a noticeable decrease in the intensity
of the LMCT band, which reflects the decreased electron
density of the triazole ligands. These results are consistent with
the findings of Nazeeruddin et al.,27 who have found that the
intensity of the LMCT transition increases with increasing
electron-donating capacity of the donor ligand.
For 2 the situation is similar except that for this compound
an additional broad feature at 1545 nm is formed at 0.75 V
prior to formation of the expected LMCT absorption band at
1216 nm (Fig. 7). Further oxidation at higher potentials results
Estimation of electronic coupling (Hab)
Hush theory may be applied to multinuclear systems exhibiting
IT bands in their mixed valence states25 to quantify the level of
interaction between the metal centres of the dinuclear complex.
A measure of the interaction between the two metal centres in
the mixed valence state can be obtained via the determination
of the resonance exchange integral Hab, as shown in eqn. 2.26
2 1/2
Hab = [α2ؒEop
]
(2)
where α2 is a measure of the extent of electron delocalisation
and can be obtained from the intervalence band using eqn. 3;
Fig. 7 Near-IR absorption spectra of the fully reduced (RuIIRuII)
mixed valence (RuIIRuIII) (electrolysis at 0.75 V vs. pseudo Ag/AgCl)
and full oxidised (RuIIIRuIII) (electrolysis at 0.85 V vs. pseudo Ag/AgCl)
complex 2.
(3)
in depletion of all spectroscopic bands as was observed for 1.
The band initially formed at 1545 nm is assigned as an IT band
on the basis of its absence in the spectra of 1 and its energy and
bandwidth which are comparable to those observed for
[(Ru(bipy)2)2bpt]4ϩ.19 The depletion of the 1545 nm band and
the formation of the LMCT band at higher energy occur with
the presence of a sharp isosbestic point at 1490 nm. The situ-
ation is similar to that of Bonvoisin et al.28 for the formation of
an IT band of an organic aromatic polyamine system, which
exhibited a single reversible three electron oxidation wave.
Emission-spectroelectrochemistry has also been used to investi-
gate the species, which are present during formation of both IT
and LMCT bands of the partially and fully oxidised species,
respectively. Oxidation of 2 up to 0.85 V (vs. pseudo Ag/AgCl)
results in depletion of the emission at 683 nm. The initial emis-
sion spectrum completely recovered upon returning the poten-
tial to 0.3 V. In contrast when 2 is oxidised at 1.3 V and the
potential returned to 0.3 V the original emission spectrum is
not reformed rather a new emission is observed at higher
where εmax is the extinction coefficient of the IT band (MϪ1
cmϪ1), ∆ν1/2 is the peak width at half maximum, d is the
estimated internuclear distance in Å, Eop is the maximum of
the IT band expressed in cmϪ1.
As noted above, the absence of two resolvable metal oxida-
tion waves in the CV of 2 indicates that any interaction if pres-
ent is weak. As a result, it is also expected that Kc will be very
small and hence at equilibrium only a small proportion of the
total concentration of 2 in solution comprises of the mixed
valence species. Since the concentration of the RuIIRuIII species
is difficult to accurately calculate, the evaluation of εmax (and
thence α2) can only be approximate. Based on the crystal
structure of 1, the internuclear distance of 2 can be estimated
as 12 Å. Eop is taken directly from the spectra of the mixed
valence species (6470 cmϪ1) and ∆ν1/2 is taken as double the
width at half maximum of the high energy side of the IT band
(5100 cmϪ1). Assuming a maximum extinction coefficient of
2400 MϪ1 cmϪ1 (that of the IT band of 4) the upper limit of the
J. Chem. Soc., Dalton Trans., 2002, 1740–1746
1745

View Article Online
interaction parameter α2 for the system is found to be 0.0055
References
cmϪ1 (Hab = 480 cmϪ1), compared with 0.016 for 4 (Hab
=
1 A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser and
A. von Zelewsky, Coord. Chem. Rev., 1988, 84, 85.
2 K. Kalyanasundaram, Coord. Chem. Rev., 1982, 46, 159.
3 E. A. Seddon and K. R. Seddon, The Chemistry of Ruthenium,
Elsevier, Amsterdam, 1984, ch. 15.
4 V. Balzani, F. Scandola, Supramolecular Photochemistry, Ellis
Horwood, Chichester, UK, 1991.
700 cmϪ1). This confirms that although an interaction is present
it is at most very weak and hence the system is best described as
type II. For the protonated complex the absence of an IT band
and the single bielectronic redox wave confirms it to be type I
i.e. no interaction.
5 Supramolecular Photochemistry, ed. V. Balzani, Reidel, Dordrecht,
1997.
6 P. R. Rich, Faraday Discuss. Chem. Soc., 1982, 75, 349.
7 K. S. Schanze and K. Sauer, J. Am. Chem. Soc., 1998, 110,
1180.
8 V. Goulle, A. Harriman and J.-M. Lehn, J. Chem. Soc., Chem.
Commun., 1993, 1034.
9 K. A. Opperman, S. L. Mecklenburg and T. J. Meyer, Inorg. Chem.,
1994, 33, 5295.
10 V. Balzani, S. Campagna, G. Denti, A. Juris and M. Venturi, Coord.
Chem. Rev., 1994, 132, 1.
11 S. Fanni, T. E. Keyes, S. Campagna and J. G. Vos, Inorg. Chem.,
1998, 37, 5933.
12 T. E. Keyes, P. M. Jayaweera, J. J. McGarvey and J. G. Vos, J. Chem.
Soc., Dalton Trans., 1997, 1627.
13 R. Wang, T. E. Keyes, R. Hage, R. H. Schmehl and J. G. Vos,
J. Chem. Soc., Chem. Commun., 1993, 1652.
Concluding remarks
Complexes 1 and 2 are synthetic intermediates in the prepar-
ation of hydroquinone and quinone containing complexes. The
present study forms therefore a basis for the investigation
of quinone complexes, which will be reported in a later publi-
cation. X-Ray and molecular structures of HL1 and H2L2 and
of 1 facilitate detailed studies of these and similar compounds
with respect to the distance dependence of electron and energy
transfer processes in the quinone/hydroquinone target com-
pounds. The electrochemical, acid–base, spectroscopic and
photophysical results reported in this study for 1 and 2, indicate
that for the dinuclear species the individual metal centres
behave independently and in an almost identical fashion to the
mononuclear complex 1. Importantly they show only a single
two electron metal-based oxidation. This indicates that the
interaction between the two metal centres is weak. In the
deprotonated complex 2 the presence of an intervalence band in
the spectroelectrochemical data confirm the presence of a weak
interaction. The results obtained show that the electronic inter-
action (Hab) in 2 is considerable less than observed in 4. This is
not unexpected on the basis of the increased distance (d )
between the metal centres and the presence of the ‘phenyl
spacer’, which has been found in previous studies to provide
poor electronic communication.19 This study allows for more
successful prediction of the supramolecular aspects of analo-
gous systems containing hydroquinone and quinone moieties
where direct electrochemical and spectroelectrochemical investi-
gations are not possible due to the presence of ligand based
oxidations at potentials lower than that of the metal centres.12
The present study is aimed at providing a base for the com-
parison of the electrochemical and photophysical properties of
such compounds and assesses the potential of the hydroquin-
one/quinone redox couple to act as an electrochemical switch.
14 B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978, 17,
3334.
15 R. Wang, J. G. Vos, R. H. Schmehl and R. Hage, J. Am. Chem. Soc.,
1992, 114, 1964.
16 J. Fait, XSCANS, Program for Data Collection and Processing,
1993, Bruker, Madison, WI.
17 G. M. Sheldrick, SHELXTL version 5.10, University of Gottingen,
Germany, 1997.
18 F. Barigelletti, L. De Cola, V. Balzani, R. Hage, J. G. Haasnoot,
J. Reedijk and J. G. Vos, Inorg. Chem., 1989, 28, 4344.
19 R. Hage, J. G. Haasnoot, J. Reedijk, R. Wang, E. M. Ryan, J. G. Vos,
A. L. Spek and A. J. M. Duisenberg, Inorg. Chim. Acta, 1990, 174,
77.
20 D. P. Rillema, D. S. Jones, C. Woods and H. A. Levy, Inorg. Chem.,
1992, 31, 2935.
21 R. Hage, A. H. J. Dijkhuis, J. G. Haasnoot, R. Prins, J. Reedijk,
B. E. Buchanan and J. G. Vos, Inorg. Chem., 1988, 27, 2185.
22 T. J. Meyer, Pure Appl. Chem., 1984, 630.
23 M. P. Robin and P. Day, Adv. Inorg. Chem. Radiochem., 1967, 10,
247.
24 D. E. Richardson and H. Taube, Inorg. Chem., 1981, 20, 1278.
25 (a) N. S. Hush, Prog. Inorg. Chem., 1967, 8, 391; (b) N. S. Hush,
Electrochim. Acta, 1968, 13, 1005.
26 C. Creutz, O. N. Marshall and N. Sutin, J. Photochem. Photobiol.,
A: Chem., 1994, 82, 47.
27 M. K. Nazeeruddin, S. M. Zakeeruddin and K. Kalyanasundaram,
J. Phys. Chem., 1993, 97, 9607.
28 J. Bonvision, J.-P Launay, M Van der Auweraer and F. C. De
Schryver, J. Phys. Chem., 1994, 98, 5052.
Acknowledgements
The authors thank Enterprise Ireland and the EU TMR Grant
no. CT96076 for financial assistance.
1746
J. Chem. Soc., Dalton Trans., 2002, 1740–1746