Electrochemical probing of ground state electronic interactions in polynuclear complexes of a new heteroditopic ligand

Article abstract of DOI:10.1039/b404602a

The synthesis and electronic properties of dinuclear ([(bipy) ₂Ru(I)M(terpy)][PF_6]4 (bipy = 2,2′-bipyridine, terpy = 2,2′:6′,2″-terpyridine; M = Ru, Os)) and trinuclear ([{(bipy)₂Ru(I)}₂M][PF_6]6 M = Ru, Os, Fe, Co) complexes bridged by 4′-(2,2′-bipyridin-4-yl)-2,2′:6′, 2″-terpyridine (I) have been investigated and are compared with those of mononuclear model complexes. The electrochemical analysis using cyclic voltammetry and differential pulse voltammetry reveals that there are no interactions in the ground state between adjacent metal centres. However, there is strong electronic communication between the 2,2′-bipyridine and 2,2′:6′,2″-terpyridine components of the bridging ligand. This conclusion is supported by a step-by-step reduction of the dinuclear and trinuclear complexes and the assignment of each electrochemical process to localised ligand sites within the didentate and terdentate domains. The investigation of the electronic absorption and emission spectra reveals an energy transfer in the excited state from the terminating bipy-bound metal centres to the central terpy-bound metal centre. This indicates that the bridge is able to facilitate energy transfer in the excited state between the metal centres despite the lack of interactions in the ground state.

Full text of DOI:10.1039/b404602a

View Article Online / Journal Homepage / Table of Contents for this issue
Electrochemical probing of ground state electronic interactions in
polynuclear complexes of a new heteroditopic ligand
Edwin C. Constable,*a Egbert Figgemeier,*b Catherine E. Housecroft,a Jerry Olssona
and Yves C. Zimmermanna
a Department of Chemistry, University of Basel, Spitalstrasse 51, CH-4056 Basel, Switzerland.
E-mail: edwin.constable@unibas.ch; Fax: ϩ41 61 2671005; Tel: ϩ41 61 2671001
b Department of Physical Chemistry, University of Uppsala, P.O. Box 579, 75123 Uppsala,
Sweden. E-mail: egbert.figgemeier@fki.uu.se; Fax: ϩ46 18 4713654; Tel: 46 18 4713661
Received 26th March 2004, Accepted 17th May 2004
First published as an Advance Article on the web 27th May 2004
The synthesis and electronic properties of dinuclear ([(bipy)2Ru(I)M(terpy)][PF6]4 (bipy = 2,2Ј-bipyridine, terpy =
2,2Ј:6Ј,2Љ-terpyridine; M = Ru, Os)) and trinuclear ([{(bipy)2Ru(I)}2M][PF6]6 M = Ru, Os, Fe, Co) complexes
bridged by 4Ј-(2,2Ј-bipyridin-4-yl)-2,2Ј:6Ј,2Љ-terpyridine (I) have been investigated and are compared with those
of mononuclear model complexes. The electrochemical analysis using cyclic voltammetry and differential pulse
voltammetry reveals that there are no interactions in the ground state between adjacent metal centres. However, there
is strong electronic communication between the 2,2Ј-bipyridine and 2,2Ј:6Ј,2Љ-terpyridine components of the bridging
ligand. This conclusion is supported by a step-by-step reduction of the dinuclear and trinuclear complexes and the
assignment of each electrochemical process to localised ligand sites within the didentate and terdentate domains. The
investigation of the electronic absorption and emission spectra reveals an energy transfer in the excited state from the
terminating bipy-bound metal centres to the central terpy-bound metal centre. This indicates that the bridge is able to
facilitate energy transfer in the excited state between the metal centres despite the lack of interactions in the ground
state.
of intra-molecular interactions.40–44 Whereas these studies
Introduction
focused mainly on investigations of the interactions using
Understanding the electronic communication within multi-
absorption spectroscopy and, to a smaller extent, electro-
nuclear metal complexes is still a challenge despite decades of
chemical techniques, we have focussed our attention on the use
research. In particular, polypyridine complexes of Ru and Os
of electrochemistry to analyse intra-bridge and inter-metal
are currently being investigated and interactions in the excited
interactions within the multinuclear complexes. Energy transfer
state have been examined in detail for a number of polypyridine
processes within the assemblies in the excited state have been
complexes with a range of bridging ligands.1,2 There is now
measured and the results are compared with those obtained
active interest in the development of integrated chemical
from the electrochemical investigations.
systems in which metal complexes play a key role as light-
harvesting components, e.g. in photoelectrochemical solar
cells or models for photosynthetic centres.3,4 A great number
of contributions dealing with intra-molecular energy transfer
in the excited state by time resolved laser spectroscopy has
been published.5–32 Interaction parameters in the ground state
probed by metal-centred oxidation processes have also attracted
significant interest.2,33,34 On the other hand, there are relatively
few electrochemical studies investigating intramolecular
interactions of the reduced ground state in multinuclear
complexes.35–38
Nevertheless, in the light of the application of multinuclear
complexes in electroluminescent devices the role of the reduced
state of bridging ligands for intramolecular interactions is of
great importance.39 In this context we present here the synthesis,
electrochemical properties and fluorescence of di- and tri-
nuclear metal complexes based on 4Ј-(2,2Ј-bipyridin-4-yl)-
2,2Ј:6Ј,2Љ-terpyridine (I) (Scheme 1) as the bridging ligand.
Similar bridges have been synthesised and investigated in terms
Experimental
General
Commercially available chemicals were reagent grade and were
used without further purification. H and 13C NMR spectra
1
were recorded on Bruker AM 250 or 400 and Gemini Varian
300 MHz spectrometers; δ is relative to TMS. IR spectra were
recorded on a Mattson Genesis FT spectrophotometer with
samples in compressed KBr discs. UV/VIS measurements were
performed using a Perkin-Elmer Lambda 19 spectrophoto-
meter and were recorded in MeCN or CHCl3. Fluorescence
experiments used a Perkin-Elmer luminescence spectrometer
LS 50 and were measured in MeCN or CHCl3. FAB and EI
mass spectra were recorded on a VG70–250 or a Finnigan
MAT312 instrument. Electrospray mass spectra (ES-MS)
were recorded using a Bruker FTMS 4.7T BioAPEX II or a
Finnigan MAT LCQ workstation. The microanalyses were
performed with a Leco CHN-900. Melting points were meas-
ured by a electrothermal digital melting point apparatus. All
electrochemical data was measured with a BAS 100 B/W
electrochemical workstation and a three-electrode cell, consist-
ing of a silver wire as a pseudo-reference, a glassy carbon disk
as the working and a platinum wire as counter electrode, was
used. After gaining a full set of voltammograms, ferrocene was
added in order to determine the exact position of the signals
within the potential window. Tetra-n-butylammonium hexa-
fluorophosphate (TBAPF6, 0.1 M) acted as the electrolyte. All
solutions were degassed thoroughly for at least 15 min with
Scheme 1
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
1918

View Article Online
Scheme 2 Reaction scheme for the formation of ligand I.
argon and an inert gas blanket was maintained over the solu-
solvent was evaporated in vacuo and CH2Cl2 (50 ml) was added.
tion during the measurements. The preparations of complexes
The organic layer was washed with saturated aqueous NaHCO3
(3 × 20 ml), dried over magnesium sulfate and concentrated
in vacuo to give VI (0.226 g, 82%) as a cream coloured solid;
containing ligand II are described in a related paper.45
1
mp 85–86 ЊC (lit. 84.8–86 ЊC47); H NMR (CDCl3, 400 MHz):
4-Methyl-2,2Ј-bipyridine (IV). A solution of [III]ϩIϪ (Scheme
2) (15 g, 46 mmol), ammonium acetate (15 g, 195 mmol) and
crotonaldehyde (3.8 ml, 46 mmol) in MeOH (150 ml) was
heated to 65 ЊC overnight. The solvent was then evaporated
in vacuo and the black oil extracted with hexane (6 × 50 ml).
The combined organic layers were washed with saturated brine
and dried over magnesium sulfate. The product was purified by
column chromatography (Al2O3/hexane–acetone, 9 : 1) followed
by recrystallisation via shock freezing from light petroleum
(bp 40–60 ЊC) to give IV (3.1 g, 40%) as a white powder; mp 62–
64 ЊC; 1H NMR (CDCl3, 250 MHz): δ 8.71 (1H, m, H6B), 8.57
(1H, d, J = 4.8 Hz, H6A), 8.43 (1H, dm, J = 7.8 Hz, H3B), 8.27
(1H, s, H3A), 7.84 (1H, dt, J = 1.7, 7.8 Hz, H4B), 7.33 (1H, m,
H5B), 7.17 (1H, m, H5A), 2.47 (3H, s, Me); 13C NMR (CDCl3,
100 MHz): δ 156.6, 156.2, 149.5, 149.3, 148.7, 137.4, 125.1,
124.1, 122.4, 121.7, 21.6; EIMS: m/z 170 (Mϩ, 100%); elemental
analysis: calc. (%) for C11H10N2: C 77.6, H 5.9, N 16.4; found:
C 77.4, H 6.0, N 16.3.
δ 10.19 (1H, s, CHO), 8.90 (1H, d, J = 4.7 Hz, H6A), 8.84 (1H, s,
H3A), 8.72 (1H, dm, J = 4.7 Hz, H6B), 8.45 (1H, d, J = 7.3 Hz,
H3B), 7.86 (1H, dt, J = 1.7, 7.3 Hz, H4B), 7.73 (1H, dm, J = 4.7
Hz, H5A), 7.37 (1H, dd, J = 1.8, 4.7 Hz, H5B); 13C NMR
(CDCl3, 100 MHz): δ 192.0, 175.2, 150.9, 149.2, 143.2, 139.6,
138.2, 125.0, 122.0, 121.9, 121.3. IR: ν(CO) 1704s cmϪ1; EIMS:
m/z 184 (Mϩ, 100%); elemental analysis: calc. (%) for
C11H8N2O: C 71.7, H 4.4, N 15.2, O 8.7; found: C 71.9, H 4.7, N
14.8, O 8.6.
4Ј-(2,2Ј-Bipyridin-4-yl)-2,2Ј:6Ј,2Љ-terpyridine (I). A solution
of 4-formyl-2,2Ј-bipyridine (0.250 g, 1.36 mmol) and NaOH
(1.0 g, 25 mmol) in MeOH (50 ml) was stirred for 5 min at room
temperature. 2-Acetylpyridine (0.430 ml, 3.00 mmol) was added
and the mixture was stirred for 1 h. Water (100 ml) was added
and the mixture was then extracted with CH2Cl2 (3 × 50 ml).
The organic layers were dried over MgSO4 and concentrated
in vacuo. The solid residue and ammonium acetate (3.0 g,
39 mmol) in EtOH (100 ml) were heated to reflux overnight.
The solution was then cooled, reduced in volume, and water
(100 ml) was added. The mixture was extracted with CH2Cl2
(3 × 50 ml), and the organic layers were dried over MgSO4 and
concentrated in vacuo. Purification by column chromatography
(Al2O3/toluene–diethylamine, 95:5) gave 4Ј-(2,2Ј-bipyridin-4-
yl)-2,2Ј:6Ј,2Љ-terpyridine (0.286 g, 55%) as an off-white solid. 1H
NMR (250 MHz, CDCl3, see Scheme 1 for ring labelling):
δ 8.93 (1H, d, J 2.0 Hz, F3), 8.88 (2 H, s, G3), 8.84 (1H, dd, J 2.0,
5.0 Hz, F6), 8.75 (3H, m, H6, E6), 8.70 (2 H, d, J 7.7 Hz, H3),
8.49 (1H, d, J 7.7 Hz, E3), 7.91 (2H, dt, J 1.7, 7.7 Hz, H4), 7.88
(1H, m, E4), 7.83 (1H, m, F5), 7.38 (3H, m, H5, E5). 13C NMR
(75 MHz, CDCl3): δ 156.2, 155.7, 149.9, 149.2, 149.1, 147.9,
147.3, 137.1, 124.1 (2C), 124.0 (2C), 121.9, 121.5 (2C), 121.4,
119.3, 119.1. IR (KBr, cmϪ1): 3045m, 3002w, 1583s, 1566s,
1536s, 1465s, 1440m, 1402m, 1383s, 1358m, 1268w, 1039m,
784s, 732s, 668m, 626s. EI-MS: m/z 387 (Mϩ, 100%). Anal.
Calc. for C25H17N5 (387.5): C 77.5, H 4.4, N 18.1; found: C 77.6,
H 4.7, N 17.7%; mp 159–160 ЊC.
[2-(N,N-Dimethylamino)vinyl]-2,2Ј-bipyridine (V). A solution
of IV (0.282 g, 1.66 mmol) and tert-butoxybis(N,N-dimethyl-
amino)methane (1.0 ml, 4.8 mmol) in dry DMF (5 ml) was
degassed with argon for 15 min and then heated to 140 ЊC
under argon for 18 h. The reaction mixture was cooled to room
temperature and water (50 ml) was added. The mixture was
then extracted with CH2Cl2 (5 × 20 ml) and the combined
organic layers dried over magnesium sulfate and concentrated
in vacuo. Purification by column chromatography (Al2O3/
toluene–diethylamine, 40 : 1) gave V (0.350 g, 94%) as a yellow
1
oil. H NMR (CDCl3, 400 MHz): δ 8.66 (1H, m, H3A), 8.42
(1H, d, J = 7.8 Hz, H6A), 8.34 (1H, d, J = 5.3 Hz, H6B), 8.13
(1H, d, J = 2.0 Hz, H3B), 7.80 (1H, dt, J = 2.0, 7.6 Hz, H4B),
7.28 (1H, ddd, J 1.2, 5.3, 7.6 Hz, H5B), 7.24 (1H, d, J 13.5 Hz,
NCH᎐), 6.99 (1H, dd, J = 2.0, 7.8 Hz, H5A), 5.11 (1H, d,
᎐
J = 13.5 Hz, ᎐CHpy), 2.91 (6H, s, NMe ). 13C NMR (CDCl3,
᎐
2
100 MHz): δ 156.9, 155.5, 149.6, 149.3, 148.9, 144.1, 137.3,
123.8, 121.7, 118.5, 115.2, 94.9, 41.0 (2C); EIMS: m/z 225 (Mϩ,
100%); elemental analysis: calc. (%) for C14H15N3: C 74.6,
H 6.8, N 18.6; found: C 74.3, H 6.9, N 18.8.
[(bipy)2Ru(I)][PF6]2. A solution of I (0.030 g, 0.077 mmol)
and cis-[RuCl2(bipy)2]ؒ2H2O (0.040 g, 0.77 mmol) in EtOH
(25 ml) was heated to reflux for 6 h.46 The solvent was evap-
orated in vacuo and the residue was purified by column
chromatography (SiO2 eluting with MeCN–saturated aqueous
KNO3–H2O, 7 : 2 : 2). An excess of NH4PF6 was added to the
4-Formyl-2,2Ј-bipyridine (VI). NaIO4 (1.2 g, 5.6 mmol) was
added to a solution of V (0.338 g, 1.50 mmol) in THF (50 ml)
and water (50 ml) The yellow solution was stirred at room
temperature for 4 h after which the insoluble products were
removed by filtration and washed with tetrahydrofuran. The
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1919

View Article Online
major red–orange fraction and the solution was reduced in
volume. The precipitate was collected by filtration over Celite,
dissolved in MeCN, and evaporated to dryness to give [(bipy)2-
N-ethylmorpholine was heated to reflux in a microwave (800 W)
for 10 min. A saturated aqueous solution of NH4PF6 (25 ml)
was added to the brown solution, the precipitate was collected
by filtration over Celite and dissolved in MeCN. The solvent
was evaporated in vacuo and the residue was purified by
preparative thin-layer chromatography (SiO2 eluting with
MeCN–saturated aqueous KNO3–H2O 7 : 2 : 2). The major
brown fraction was dissolved in MeCN, an excess of NH4PF6
was added, and the solution was reduced in volume. The
precipitate was collected by filtration over Celite, dissolved in
MeCN and evaporated to dryness to afford [(bipy)2Ru(µ-I)-
1
Ru(I)][PF6]2 (0.080 g, 95%) as a red–orange powder. H NMR
(400 MHz, CD3CN, see Scheme 3 for ring labelling): δ 9.12 (1H,
d, J 2.4 Hz, F3), 9.06 (2H, s, G3), 8.86 (1H, d, J 7.5 Hz, E3), 8.72
(2H, d, J 7.5 Hz, H3), 8.56–8.54 (4H, m, A3, B3, C3, D3), 8.19
(2H, m, H4), 8.18 (1H, m, E4), 8.14–8.11 (4H, m, A4, B4, C4, D4),
8.10 (1H, m, F6), 8.06 (1H, m, F5), 7.85–7.76 (7H, overlapping
m, A6, B6, C6, D6, E6, H6), 7.52–7.40 (7H, overlapping m, A5, B5,
C5, D5, E5, H5). IR (KBr, cmϪ1): 3430m, 1617m, 1475w, 1446w,
1423w, 842s, 793w, 765m, 558s. UV/VIS (CH3CN): λmax/nm (ε/
dm3 molϪ1 cmϪ1) 242.7 (59400), 286.7 (82500), 461.0 (14500).
Fluorescence (CH3CN): λmax/nm (λex/nm) 627.6 (460). ES-MS:
m/z 946.0 ([M Ϫ PF6]ϩ), 400.6 ([M Ϫ 2PF6]2ϩ). Anal. Calc. for
C45H33N9RuP2F12ؒ3H2O (1144.8): C 47.2, H 3.4, N 11.0; found:
C 47.2, H 3.5, N 11.2%. EЊ/V vs. Fc: (Ru2ϩ/Ru3ϩ) 0.93.
1
Os(terpy)][PF6]4 (0.007 g, 43%) as a dark-brown powder. H
NMR (400 MHz, CD3CN, see Scheme 4 for ring labelling):
δ 9.26 (1H, m, F3), 9.17 (2H, s, G3), 9.04 (1H, d, J 8.0 Hz, E3),
8.82 (2H, d, J 8.2 Hz, K3), 8.74 (2H, d, J 8.2 Hz, H3), 8.63–8.58
(4H, m, A3, B3, C3, D3), 8.51 (2 H, d, J 8.3 Hz, J3), 8.23 (1H, dt,
J 2.2, 8.0 Hz, E4), 8.19–8.06 (6H, m, F5, F6, A4, B4, C4, D4), 8.03
(1H, t, J 8.2 Hz, K4), 7.96 (1H, m, E6), 7.90–7.78 (8H, m, H4, J4,
A6, B6, C6, D6), 7.53–7.48 (5H, m, E5, A5, B5, C5, D5), 7.33–7.25
(4H, m, H6, J6), 7.21–7.08 (4H, m, H5, J5). IR (KBr, cmϪ1):
3430m, 2924w, 1638w, 1618w, 1466w, 1449m, 1422m, 1384s,
1042w, 836s, 786m, 730w, 558s. UV/VIS (CH3CN): λmax/nm
(ε/dm3 molϪ1 cmϪ1) 286.8 (66900), 310.8 (46300), 497.9
(22600), 669.9 (4000). Fluorescence (CH3CN): λmax/nm (λex/nm)
792.8 (498), 782.1 (671). ES-MS: m/z 1659.0 ([M Ϫ PF6]ϩ),
757.0 ([M Ϫ 2PF6]2ϩ). Anal. Calc. for C60H44N12RuOsP4F24ؒ
4H2O (1876.2): C 38.4, H 2.8, N 8.9; found: C 38.1, H 2.6,
N 8.9%. EЊ/V vs. Fc: (Ru2ϩ/Ru3ϩ) 0.94 (Os2ϩ/Os3ϩ) 0.58.
Scheme 3 Ring labelling in [(bipy)2Ru(I)]2ϩ. This labelling is also used
in the complexes [{(bipy)2Ru(µ-I)}2M]6ϩ (M = Fe, Ru, Os and Co).
[{(bipy)2Ru(ꢀ-I)}2Fe][PF6]6. A solution of [(bipy)2Ru(I)][PF6]2
(0.025 g, 0.023 mmol) and [Fe(H2O)6][BF4]2 (0.057 g, 0.017
mmol) in EtOH (50 ml) was heated to reflux for 1 h. The solvent
was then evaporated in vacuo and the residue purified by
column chromatography (SiO2 eluting with MeCN–saturated
aqueous KNO3–H2O 7 : 1.5 : 0.5). An excess of NH4PF6 was
added to the major green–brown fraction and the solution was
reduced in volume. The precipitate was collected by filtration
through Celite, dissolved in MeCN and evaporated to dryness
to yield [{(bipy)2Ru(µ-I)}2Fe][PF6]6 (0.023 g, 80%) as a red-
brown powder. 1H NMR (250 MHz, CD3CN, see Scheme 4 for
ring labelling): δ 9.34 (2H, d, J 2.4 Hz, F3), 9.32 (4H, s, G3), 8.98
(2H, d, J 7.9 Hz, E3), 8.68 (4H, d, J 7.9 Hz, H3), 8.60–8.54 (8H,
m, A3, B3, C3, D3), 8.24–8.10 (14H, m, F5, F6, E4, A4, B4, C4,
D4), 7.96–7.90 (6H, m, H4, E6), 7.86–7.79 (8H, m, A6, B6, C6,
D6), 7.55–7.42 (10H, m, E5, A5, B5, C5, D5), 7.18–7.08 (8H, m,
H5, H6). IR (KBr, cmϪ1): 3650w, 3450m, 3088w, 2924w, 1720w,
1616m, 1606m, 1533w, 1466m, 1446m, 1423m, 1401m, 1243w,
1027w, 839s, 788m, 762m, 732w, 558s. UV/VIS (CH3CN):
λmax/nm (ε/dm3 molϪ1 cmϪ1) 243.8 (78800), 286.6 (155700),
462.2 (27000), 591.0 (34700). Fluorescence (CH3CN): λmax/nm
(λex/nm) 628.0 (460). ES-MS: m/z 1119.1 ([M Ϫ 2PF6]2ϩ), 698.8
([M Ϫ 3PF6]3ϩ), 487.1 ([M Ϫ 4PF6]4ϩ), 276.4 ([M Ϫ 6PF6]6ϩ).
Anal. Calc. for C90H66N18FeRu2P6F36ؒ6H2O (2635.4): C 40.9,
[(bipy)2Ru(ꢀ-I)Ru(terpy)][PF6]4. A solution of [(bipy)2Ru(I)]-
[PF6]2 (0.030 g, 0.027 mmol) and [Ru(terpy)Cl3]48 (0.012 g,
0.027 mmol) in EtOH (50 ml) with two drops of N-ethylmor-
pholine was heated to reflux for 2 h. The solvent was evaporated
in vacuo and the residue was purified by column chromato-
graphy (silica/acetonitrile–saturated aqueous potassium
nitrate–water 7 : 1.5 : 0.5). To the major red–orange fraction
an excess of ammonium hexafluorophosphate was added and
the solution reduced in volume. The precipitate was collected by
filtration through Celite, dissolved in MeCN and evaporated to
dryness to afford [(bipy)2Ru(µ-I)Ru(terpy)][PF6]4 (0.030 g, 64%)
1
as a red–orange powder. H NMR (250 MHz, CD3CN, see
Scheme 4 for ring labelling): δ 9.18 (1H, m, F3), 9.10 (2H, s, G3),
8.92 (1H, d, J 8.1 Hz, E3), 8.76 (2H, d, J 8.0 Hz, K3), 8.67 (2H,
d, J 7.9 Hz, H3), 8.58–8.53 (4H, m, A3, B3, C3, D3), 8.49 (2H, d,
J 8.4 Hz, J3), 8.44 (1H, t, J 8.0 Hz, K4), 8.19 (1H, dt, J 2.2, 8.0
Hz, E4), 8.14–8.04 (6H, m, F5, F6, A4, B4, C4, D4), 7.99–7.88
(4H, m, H4, J4), 7.85–7.70 (5H, m, E6, A6, B6, C6, D6), 7.53–7.41
(5H, m, E5, A5, B5, C5, D5), 7.39–7.33 (4H, m, H6, J6), 7.23–7.11
(4H, m, H5, J5). IR (KBr, cmϪ1): 3443m, 3060w, 1605m, 1466m,
1448m, 1421w, 1388m, 1243w, 1029w, 840s, 788m, 765m, 732w,
558s. UV/VIS (CH3CN): λmax/nm (ε/dm3 molϪ1 cmϪ1) 241.7
(54900), 286.9 (104200), 460.0 (22500), 500.3 (37400). Fluor-
escence (CH3CN): λmax/nm (λex/nm) 688.8 (460), 691.0 (500).
ES-MS: m/z 1570.0 ([M Ϫ PF6]ϩ), 712.5 ([M Ϫ 2PF6]2ϩ). Anal.
Calc. for C60H44N12Ru2P4F24ؒ4H2O (1787.1): C 40.3, H 3.0, N
9.4; found: C 40.3, H 3.2, N 9.4%. EЊ/V vs. Fc: (Ru2ϩ/Ru3ϩ) 0.96.
H 3.0, N 9.6; found: C 40.3, H 3.3, N 9.4%. EЊ/V vs. Fc: (Ru2ϩ
/
Ru3ϩ) 0.94, (Fe2ϩ/Fe3ϩ) 0.79.
[{(bipy)2Ru(ꢀ-I)}2Co][PF6]6. A solution of [(bipy)2Ru(I)]-
[PF6]2 (0.016 g, 0.015 mmol) and [Co(H2O)6][BF4]2 (27 mg,
0.0080 mmol) in ethanol (10 ml) was stirred at room temper-
ature for 1 h. The solvent was then evaporated in vacuo and the
residue was purified by column chromatography (SiO2 eluting
with MeCN–saturated aqueous KNO3–H2O 7 : 1.5 : 0.5). An
excess of NH4PF6 was added to the major orange fraction and
the solution reduced in volume. The precipitate was collected by
filtration through Celite, dissolved in MeCN and evaporated to
dryness to give [{(bipy)2Ru(µ-I)}2Co][PF6]6 (0.015 g, 81%) as an
Scheme 4 Ring labelling in [(bipy)2Ru(µ-I)M(terpy)]4ϩ (M = Ru, Os).
1
orange powder. H NMR (250 MHz, CD3CN): δ 91.5 (4H, br,
[(bipy)2Ru(ꢀ-I)Os(terpy)][PF6]4. A solution of [(bipy)2Ru(I)]-
[PF6]2 (0.010 g, 0.009 mmol) and [Os(terpy)Cl3]49,50 (0.006 g,
0.027 mmol) in ethylene glycol (4 ml) with two drops of
H6), 52.5 (4H, s, H3), 36.9 (4H, s, G3), 32.9 (4 H, s, H5), 13.9
(2H, s, Hbipy), 12.7 (2H, s, Hbipy), 10.2 (2H, s, Hbipy), 9.8 (4H, s,
H4), 9.2–7.7 (40H, m, Hbipy, A3,4,5,6, B3,4,5,6, C3,4,5,6, D3,4,5,6). IR
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1920

View Article Online
(KBr, cmϪ1): 3446m, 2924w, 1618m, 1604m, 1561w, 1534w,
1509w, 1467m, 1446m, 1422m, 1384s, 1274w, 1245w, 1028w,
1016w, 839s, 790m, 763m, 731w, 662w, 557s. UV/VIS (CH3CN):
λmax/nm (ε/dm3 molϪ1 cmϪ1) 243.2 (87900), 286.6 (162000),
469.5 (30500). ES-MS: m/z 1120.1 ([M Ϫ 2PF6]2ϩ), 698.4 ([M Ϫ
3PF6]3ϩ), 487.6 ([M Ϫ 4PF6]4ϩ), 361.3 ([M Ϫ 5PF6]5ϩ), 276.7
([M Ϫ 6PF6]6ϩ). Anal. Calc. for C90H66N18CoRu2P6F36ؒ10H2O
(2710.4): C 39.8, H 3.2, N 9.3; found: C 39.8, H 3.5, N 9.4%.
EЊ/V vs. Fc: (Ru2ϩ/Ru3ϩ) 0.95, (Co2ϩ/Co3ϩ) Ϫ0.07.
Results and discussion
Ligand synthesis
The target ligand I contains bipy and terpy metal-binding
domains directly linked by a C–C bond between the 4- and
4Ј-positions, respectively. The key synthetic step was to be the
generation of an intermediate 1,5-dicarbonyl compound by
condensation of 2-acetylpyridine with 2,2Ј-bipyridine-4-
carbaldehyde and subsequent cyclisation to build the central
ring of the terpy domain (Scheme 2). The starting bipy
derivative IV was prepared in reasonable yield by a Kröhnke47
reaction of the activated pyridinium salt [III]I with crotonalde-
hyde in the presence of ammonium acetate,51 a method that we
find superior in terms of accessibility of starting materials to
other published routes.52–54
We have found the oxidation of IV to the aldehyde VI with
selenium dioxide55 gives variable yields; in contrast, our
two-step procedure gives excellent and reproducible yields. The
reaction of IV with Bredereck’s reagent tBuOCH(NMe2)2 56 gave
the N,N-dimethylaminovinyl derivative V in near-quantitative
yield. Subsequent oxidation with periodate57,58 gave VI in high
yield. The aldehyde is the key intermediate for the preparation
of the ditopic ligand I containing both the terpy and bipy
metal-binding domains. We considered stepwise Kröhnke-type
synthesis47 of I, but found that the most reliable method
for the preparation of I involved the reaction of VI with
2-acetylpyridine in the presence of base to give 4-(2,2Ј-bi-
pyridin-4Ј-yl)-1,5-bis(2-pyridyl)pentane-1,5-dione. This was
not isolated but was reacted in situ with ammonium acetate
under aerobic conditions to give I in 55% isolated yield. The
compound was fully characterised by conventional methods. A
parent ion was observed at m/z 387 in the EI mass spectrum.
The 1H NMR spectrum was well-resolved with a characteristic
singlet assigned to proton 3Ј of the terpy domain appearing at
δ 8.88.
[{(bipy)2Ru(ꢀ-I)}2Ru][PF6]6. A solution of [(bipy)2Ru(I)]-
[PF6]2 (0.010 g, 0.009 mmol) and RuCl3ؒ3H2O (0.001 g, 0.004
mmol) in ethylene glycol (4 ml) together with two drops of
N-ethylmorpholine was heated to reflux in a microwave oven
(800 W) for 6 min. A solution of saturated aqueous NH4PF6 (25
ml) was added to the orange solution, and the precipitate was
collected by filtration through Celite and dissolved in MeCN.
The solvent was evaporated in vacuo and the residue was
purified by column chromatography (SiO2 eluting with MeCN–
saturated aqueous KNO3–H2O 7 : 1.5 : 0.5). An excess of
NH4PF6 was added to the major red fraction and the solution
reduced in volume. The precipitate was collected by filtration
through Celite, washed with MeCN and evaporated to dryness
to yield [{(bipy)2Ru(µ-I)}2Ru][PF6]6 (0.006 g, 51%) as a red–
1
orange powder. H NMR (250 MHz, CD3CN): δ 9.25 (2H, d,
J 2.4 Hz, F3), 9.17 (4H, s, G3), 8.98 (2H, d, J 7.9 Hz, E3), 8.73
(4H, d, J 7.9 Hz, H3), 8.59–8.54 (8H, m, A3, B3, C3, D3), 8.23–
8.08 (14H, m, F5, F6, E4, A4, B4, C4, D4), 8.01–7.92 (6H, m, H4,
E6), 7.86–7.78 (8H, m, A6, B6, C6, D6), 7.53–7.40 (14H, m, E5,
H6, A5, B5, C5, D5), 7.25–7.18 (4H, m, H5). IR (KBr, cmϪ1):
3650w, 3443m, 3091w, 2924w, 1617m, 1607m, 1466m, 1447m,
1421m, 1383m, 1244w, 1027w, 837s, 787m, 763m, 731w, 558s.
UV/VIS (CH3CN): λmax/nm (ε/dm3 molϪ1 cmϪ1) 242.4 (51100),
286.4 (118300), 460.0 (17600), 512.1 (38700). Fluorescence
(CH3CN): λmax/nm (λex/nm) 670.4 (460), 680.0 (500). ES-MS:
m/z 1141.6 ([M Ϫ 2PF6]2ϩ), 713.0 ([M Ϫ 3PF6]3ϩ), 498.3 ([M Ϫ
4PF6]4ϩ), 369.9 ([M Ϫ 5PF6]5ϩ), 283.9 ([M Ϫ 6PF6]6ϩ). Anal.
Calc. for C90H66N18Ru3P6F36ؒ6H2O (2680.6): C 40.3, H 2.9,
Preparation and characterisation of complexes
N 9.4; found: C 40.0, H 3.2, N 9.2. EЊ/V vs. Fc: (Ru2ϩ/Ru3ϩ
)
The reaction of ligand I with cis-[RuCl2(bipy)2]ؒ2H2O in EtOH
at reflux followed by purification and anion exchange, gave
[(bipy)2Ru(I)][PF6]2 in 95% yield and in which the bipy domain
of I is selectively coordinated. The ES-MS of the complex
showed major peaks at m/z 946.0 and 400.6 assigned to [M Ϫ
PF6]ϩ and [M Ϫ 2PF6]2ϩ, respectively; isotope patterns matched
0.91.
[{(bipy)2Ru(ꢀ-I)}2Os][PF6]6. A solution of [(bipy)2Ru(I)]-
[PF6]2 (0.015 g, 0.013 mmol) and [NH4]2[OsCl6] (0.003 g, 0.007
mmol) in ethylene glycol (4 ml) together with two drops of
N-ethylmorpholine was heated to reflux in a microwave oven
(800 W) for 6 min. A solution of saturated aqueous NH4PF6
(25 ml) was then added to the dark-brown solution, the precipi-
tate was collected by filtration through Celite and washed with
MeCN. The solvent was evaporated in vacuo and the residue
was purified by column chromatography (SiO2 eluting with
MeCN–saturated aqueous KNO3–H2O 7 : 2 : 2). An excess of
NH4PF6 was added to the major black fraction and the solution
reduced in volume. The precipitate was collected by filtration
through Celite, washed with MeCN and evaporated to dryness
to yield [{(bipy)2Ru(µ-I)}2Os][PF6]6 (0.012 g, 67%) as a dark-
1
those calculated. In the H NMR spectrum of [(bipy)2Ru(I)]-
[PF6]2, a characteristic singlet was observed at δ 8.89, assigned
to proton G3 (see Scheme 2). A COSY experiment allowed the
1
full assignment of the H NMR spectrum in which there was
significant overlapping of signals arising from corresponding
protons on bipy rings A–F.
The preparations of the dinuclear and trinuclear complexes
used the ‘complexes as ligands’ approach.59 Reaction of
[Ru(terpy)Cl3] with [(bipy)2Ru(I)][PF6]2 in the presence
of N-ethylmorpholine resulted, after work up, in the isolation
of [(bipy)2Ru(µ-I)Ru(terpy)][PF6]4 in 64% yield. The dominant
peaks in the ES-MS were at m/z = 1570.0 and 712.5, consistent
with the ions [M Ϫ PF6]ϩ and [M Ϫ 2PF6]2ϩ; isotope patterns
were in accord with those simulated for these ions and con-
1
brown powder. H NMR (250 MHz, CD3CN): δ 9.25 (2 H, d,
J 2.4 Hz, F3), 9.19 (4 H, s, G3), 9.04 (2 H, d, J 8.0 Hz, E3), 8.74
(4 H, d, J 7.9 Hz, H3), 8.59–8.53 (8 H, m, A3, B3, C3, D3), 8.23–
8.05 (14 H, m, F5, F6, E4, A4, B4, C4, D4), 7.95–7.89 (14 H, m,
H4, E6, A6, B6, C6, D6), 7.55–7.42 (10 H, m, E5, A5, B5, C5, D5),
7.29 (4 H, d, J 5.2 Hz, H6), 7.16–7.11 (4 H, m, H5). IR (KBr,
cmϪ1): 3450m, 2923w, 1617m, 1605m, 1465m, 1445m, 1422m,
1395m, 1353w, 1027w, 837s, 786m, 762m, 730w, 557s. UV/VIS
(CH3CN): λmax/nm (ε/dm3 molϪ1 cmϪ1) 241.3 (53400), 287.1
(112400), 460.0 (21600), 510.3 (37700), 685.7 (8300). Fluor-
escence (CH3CN): λmax/nm (λex/nm) 767.2 (460), 767.4 (500),
776.6 (690). ES-MS: m/z 1186.5 ([M Ϫ 2PF6]2ϩ), 742.3 ([M Ϫ
3PF6]3ϩ). Calc. for C90H66N18Ru2OsP6F36ؒ4H2O (2733.7): C
39.5, H 2.7, N 9.2; found: C 39.2, H 2.8, N 9.4. EЊ/V vs. Fc:
(Ru2ϩ/Ru3ϩ) 0.92, (Os2ϩ/Os3ϩ) 0.60.
1
firmed the presence of two Ru atoms. The H NMR spectrum
of [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ was assigned by standard COSY
techniques. With the exception of new signals assigned to
the terminal terpy domain, on going from [(bipy)2Ru(I)]2ϩ to
[(bipy)2Ru(µ-I)Ru(terpy)]4ϩ, the 1H NMR spectrum showed
only significant changes in chemical shift for protons H6, H5
and H4. The complex [(bipy)2Ru(µ-I)Os(terpy)]4ϩ was prepared
from [(bipy)2Ru(I)]2ϩ and [Os(terpy)Cl3] under reducing con-
ditions (N-ethylmorpholine) in a microwave oven, and was
isolated as the [PF6]Ϫ salt. In the ES-MS, the highest mass peak
at m/z corresponded to [M Ϫ PF6]ϩ. The 1H NMR spectrum of
[(bipy)2Ru(µ-I)Os(terpy)]4ϩ was assigned by routine methods,
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1921

View Article Online
Ϫ
Table 1 Redox potentials of mono-, di- and trinuclear complexes (as PF6 salts) in MeCN with [nBuN4][PF6] as supporting electrolyte (V, vs.
internal Fc/Fcϩ reference, error: 20 mV)
Complex
RuII–RuIII
MII–MIII
Reductions
Ref.
[Ru(bipy)3]2ϩ
0.88
0.88
Ϫ1.72
Ϫ1.65
Ϫ1.50
Ϫ1.54
Ϫ1.47
Ϫ1.61
Ϫ1.34
Ϫ1.37
Ϫ1.32
Ϫ1.37
Ϫ1.31
Ϫ1.33
Ϫ1.30
Ϫ1.43
Ϫ1.96
Ϫ1.90
Ϫ1.82
Ϫ1.80
Ϫ1.77
Ϫ1.87
Ϫ1.68
Ϫ1.68
Ϫ1.62
Ϫ1.65
Ϫ1.43
Ϫ1.52
Ϫ1.41
Ϫ1.60
Ϫ2.17
62
62
63
63
63
[Ru(terpy)2]2ϩ
[Fe(II)2]2ϩ
0.80
0.62
[Ru(II)2]2ϩ
0.95
[Os(II)2]2ϩ
[(bipy)2Ru(I)]2ϩ
0.93
0.90
0.94
0.97
0.94
0.90
0.91
0.92
0.95
Ϫ2.10
Ϫ1.84
Ϫ1.83
Ϫ1.78
Ϫ1.81
Ϫ1.73
Ϫ1.77
Ϫ1.73
Ϫ1.83
Ϫ2.57
Ϫ1.90
Ϫ1.92
Ϫ1.88
Ϫ1.93
Ϫ1.85
Ϫ1.89
Ϫ1.88
Ϫ1.91
Ϫ2.82
Ϫ2.23
Ϫ2.22
Ϫ2.21
Ϫ2.25
Ϫ2.23
Ϫ2.31
Ϫ2.22
Ϫ2.27
[(bipy)2Ru(µ-I)Ru(terpy)]4ϩ
[(bipy)2Ru(µ-I)Os(terpy)2]4ϩ
[(bipy)2Ru(µ-I)Ru(II)]4ϩ
[(bipy)2Ru(µ-I)Os(II)]4ϩ
[{(bipy)2Ru(µ-I)}2Ru]6ϩ
[{(bipy)2Ru(µ-I)}2Os]6ϩ
[{(bipy)2Ru(µ-I)}2Fe]6ϩ
[{(bipy)2Ru(µ-I)}2Co]6ϩ
0.96
0.58
0.97
0.62
0.98
0.59
0.78
Ϫ0.07
and signals for the bridging ligand I were at shifts close to those
for the same ligand in [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ
.
The trinuclear complexes [{(bipy)2Ru(µ-I)}2M]6ϩ (M = Fe,
Ru, Os and Co) were assembled by reactions of [(bipy)2-
Ru(µ-I)]2ϩ with [Fe(H2O)6][BF4]2, RuCl3ؒ3H2O under reducing
conditions, [NH4]2[OsCl6] and [Co(H2O)6][BF4]2, respectively.
Each was isolated as a [PF6]Ϫ salt. The ES-MS of each complex
gave highest mass peaks corresponding to the parent ion with
loss of two [PF6]Ϫ. Observed isotope patterns were in accord
with those calculated, giving evidence for the presence of
Ru2Fe, Ru3, Ru2Os and Ru2Co cores, respectively. The 1H NMR
spectra of the diamagnetic complexes [{(bipy)2Ru(µ-I)}2M]6ϩ
(M = Fe, Ru, Os) were assigned by routine methods. Signals
assigned to bridging ligand I were at chemical shifts little
changed from those in [(bipy)2Ru(µ-I)M(terpy)]4ϩ (M = Ru,
Fig. 1 Cyclic voltammogram (A) and differential pulse voltammo-
gram (B) of the compound [{(bipy)2Ru(I)}2Fe][PF6]6 in acetonitrile
containing 0.1 M [nBuN4][PF6] showing the positive part of the
potential window. The potential scale refers to the internal Fc/Fcϩ
redox couple, a glassy carbon disk electrode acted as the working
electrode and a Pt wire as the counter electrode. The scan rate was
1
Os). In the H NMR spectrum of [{(bipy)2Ru(µ-I)}2Co]6ϩ, the
chemical shifts of the signals assigned to the terminal bipy
ligands were, as expected, affected the least by the presence
of the paramagnetic Co2ϩ centre. All other protons were
significantly paramagnetically shifted. The signals in the terpy
domain of ligand I could be assigned by comparison with those
of [Co(terpy)2]2ϩ and related complexes for which COSY
spectra have been obtained.60 Although all signals were
broadened, only the signal for protons H6 (the closest to the
Co2ϩ centre) was so greatly broadened as to prohibit inte-
gration. The relative integrals of the signals at δ 52.5, 36.9, 32.9,
9.8, 13.9, 12.7 and 10.2 were 2 : 2 : 2 : 2 : 1 : 1 : 1 confirming
these as belonging to terpy (relative integral 2) or bipy (relative
integral 1) domains of the bridging ligand.
0.1 V sϪ1
.
the compounds.2 In terms of electrochemistry, the coupling
of adjacent metal centres in identical chemical environments
results in a peak splitting of the redox-processes, or, more
generally, in a shift of the redox potential relative to the
mononuclear fragments. The first and most representative
example for this behavior is the well known Creutz–Taube ion.64
In order to investigate the electronic coupling mediated by
the bridging ligand I, we have studied the electrochemical
behavior of [(bipy)2Ru(µ-I)Ru(terpy)][PF6]4, [(bipy)2Ru(µ-I)-
Os(terpy)][PF6]4, [(bipy)2Ru(µ-I)Ru(II)][PF6]4, [(bipy)2Ru(µ-I)-
Os(II)][PF6]4 [(bipy)2Ru(µ-I)}2Fe][PF6]6, [(bipy)2Ru(µ-I)}2Ru]-
[PF6]6, [(bipy)2Ru(µ-I)}2Os][PF6]6 and [{(bipy)2Ru(µ-I)}2Co]6ϩ
and redox potentials are listed in Table 1. Table 1 also
contains electrochemical data for the parent [M(bipy)3]2ϩ and
[M(terpy)2]2ϩ complexes together with the potentials of
[M(II)2][PF6]2 (M = Fe, Ru, Os).63,65
Electrochemical characterisation
All of the mono-, di- and trinuclear complexes were electro-
chemically active in acetonitrile solution. All oxidation and
reduction processes observed were quasi-reversible and the
potentials for the complexes and for a variety of model
substances are presented in Table 1.61
Cyclic (A) and differential pulse voltammograms (DPV) (B)
for the representative trinuclear complex [{(bipy)2Ru(I)}2-
Fe][PF6]6 are presented in Fig. 1 as examples of the electro-
chemical response in the oxidative region of all the trinuclear
complexes that we have investigated. Two well-defined peaks
are observed by both methods and were assigned to the redox-
processes of RuII/RuIII at ϩ0.92 V and FeII/FeIII at ϩ0.78 V,
respectively. The CV indicates quasi-reversible processes while
the DPV reveals an expected ratio of the integrated charge for
the redox-active species of 2 : 1 within an error of 5%. In the
case of the complex [{(bipy)2Ru(I)}2Ru][PF6]6, only a single
Ru-centred process is observed.
A comparison of the RuII/RuIII redox potential of the
complex [Ru(bipy)2(I)][PF6]2 with those of the ruthenium-
containing trinuclear complexes shows them to be identical
within experimental error indicating that there is no electronic
coupling mediated by ligand I. This conclusion is supported by
comparisons of the oxidation waves between other pairs of
mononuclear and multinuclear complexes. The electrochemical
responses of the central metal ions of [{(bipy)2Ru(µ-I)}2Fe]-
[PF6]6, [{(bipy)2Ru(µ-I)}2Ru][PF6]6, [{(bipy)2Ru(µ-I)}2Os][PF6]6
and [{(bipy)2Ru(µ-I)}2Fe][PF6]6 are within 10 mV of those of
the mononuclear complexes [M(II)2][PF6]2 (M = Fe, Ru, Os).63,66
Furthermore, the Ru- and Os-centred processes of the
complexes [{(bipy)2Ru(µ-I)Ru(terpy)][PF6]4, [{(bipy)2Ru(µ-I)-
Os(terpy)][PF6]4 do not differ from those of the model
mononuclear complexes.67,68
In multinuclear complexes, the electronic interactions
between the metal centres are of major importance in deter-
mining the physical, photophysical and chemical behaviour of
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1922

View Article Online
Fig. 2 Cyclic voltammogram (A) and differential pulse voltammogram (B) of the compound [{(bipy)2Ru(µ-I)}2Fe][PF6]6 in acetonitrile showing the
negative part of the potential window. The potential scale refers to the Fc/Fcϩ redox couple, a glassy carbon disk electrode acted as the working
electrode and a platinum wire as the counter electrode. The numbers given in (B) indicate the integrated area of the fitted peaks and the arrow
indicates the scan direction. The scan rate for the CV was 0.1 V sϪ1
.
Multinuclear complexes that are most similar to those con-
sidered in this paper are dinuclear complexes bridged by the
‘back-to-back’ ligands VII and VIII (Scheme 5) which are
analogues of bipy and terpy, respectively.40,43 Whereas for
[(bipy)2Ru(µ-VII)Ru(bipy)2]4ϩ a weak electronic interaction is
reported,43 in [(terpy)Ru(µ-VIII)Ru(terpy)]4ϩ there is neither
splitting of the RuII/RuIII process nor shifting of the peak with
For the starting material, [Ru(bipy)2(I)]2ϩ, five processes were
observed (Fig. 3(A)). Three of these peaks are assigned to the
step-wise reduction of the three bipyridine ligands, including
the bipyridine part of ligand I. By a comparison with the elec-
trochemical signature of terpyridine (dashed line, Fig. 3(A)),
the peak at lowest potential can be assigned to the terpyridine
part of ligand I. The signal at Ϫ2.30 V could not be assigned to
a specific reduction site and is probably caused by the proton-
ation of the non-coordinated terpy part of [Ru(bipy)2(I)]2ϩ. In
spite of this uncertainty, the areas of the four dominant peaks
are equal within an error of <10% and therefore it can be
concluded that each is a product of a one-electron reduction.
During the stepwise addition of Fe(), the peak at most
negative potential disappears and two new peaks at Ϫ1.30 and
Ϫ1.40 V evolve (Figs. 3(B) and (C)). At a 2 : 1 ratio of the
monomeric complex and Fe(), the characteristic pattern
for the trinuclear complexes could be observed (Fig. 3(D),
comparison with the dashed line).
respect to [Ru(terpy)2]2ϩ 40
behaves more like VIII than VII.
.
Our current results indicate that I
From this experiment, together with the result of the
integration of the five peaks in Fig. 2 (1 : 1 : 2 : 2 : 2.5), we can
conclude that the two terpyridine ligands of the two bridges
in the trinuclear complexes are reduced first. This is also in
agreement with the fact that the reduction potentials in
Scheme 5
[Ru(terpy)2]2ϩ are less negative than those in [Ru(bipy)3]2ϩ
,
Discussion of ligand reduction
which implies that in a multinuclear complex with mixed
ligands, the terpyridine ligands are reduced before the
bipyridine ligands.
The three remaining peaks at most negative potentials
(see Fig. 3(B) and (D)) can be assigned to the three different
bipyridine domains of the terminal Ru coordination centres.
The integration of 2 : 2 : 2 in relation to the first two peaks
means that equivalent bipy ligands on both sides of the tri-
nuclear complex are reduced at the same potentials. This indi-
cates that these ligands are not communicating in the reduced
A nearly ideal behavior was observed on the reductive side of
the potential window, and this is demonstrated by the plots
shown in Fig. 2. The CV is characterised by five quasi-reversible
peaks, which also show up in the DPV experiment. The final
and sharp peak in the CV at around Ϫ2.3 V is irreversible. The
shape and the irreversibility is most likely due to a adsorption
phenomenon of the highly reduced complexes.69 Assuming that
the LUMOs in the polypyridine complexes are located on the
ligands, each reduction can be assigned to specific ligands
rather than to a delocalisation on the entire complex.66 In order
to assign these peaks to the ligands, an experiment was
performed in which a solution of the monomeric compound
[Ru(bipy)2(I)]2ϩ was doped with drops of an 20 mM acetonitrile
solution of Fe(BF4)2 until the Fe2ϩ concentration reached half
of the concentration of [Ru(bipy)2(I)]2ϩ. This titration was
monitored by DPV and the result is shown in Fig. 3.
state, in contrast to the situation in [Ru(bipy)3]2ϩ 62
. We are left
with the problem of assigning the three peaks to the terminal
Ru(bipy)2 ligands and to the bipy domain of ligand I.
Fig. 4 shows the positions of the reductions of the trinuclear,
dinuclear and some model monomeric complexes. The solid
lines connect equivalent ligands in related complexes, whereas
the red dotted arrows indicate the titration experiment. Moving
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1923

View Article Online
Fig. 3 The titration of [Ru(bipy)2(I)]ϩ with Fe(BF4)2 in MeCN monitored by DPV ((A) = no Fe(BF4)2 added; (D) = Fe(BF4)2 concentration is half of
[Ru(bipy)2(I)]ϩ concentration; (B) and (C) = intermediate concentrations). The scale is standardised to Fc/Fcϩ and the electrode configuration is
identical to that described earlier. For comparison, the dashed line in plot A indicates the DPV of terpy in MeCN, and the dashed line in plot D
represents the voltammogram of the synthetically isolated [{(bipy)2Ru(µ-I)}2Fe]; these were measured separately.
conclude that the latter are reduced before the bipy unit in
ligand I. It follows that the peak at Ϫ2.23 V in [(bipy)2Ru-
(µ-I)Ru(terpy)]4ϩ arises from the reduction of the bipy part
of bridging ligand I. This means that the order of reduction of
the bipy domains changes on going from [Ru(bipy)2(I)]2ϩ to
[(bipy)2Ru(µ-I)Ru(terpy)]4ϩ
.
Following from these results, the following picture emerges
for [{(bipy)2Ru(µ-I)}2Fe]6ϩ, [{(bipy)2Ru(µ-I)}2Ru]6ϩ, [{(bipy)2-
Ru(µ-I)}2Os]6ϩ and [{(bipy)2Ru(µ-I)}2Co]6ϩ. The first two peaks
of equal integrated area arise from the reduction of the two
terpy ligands. The next two peaks arise from the reduction of
the bipy ligands not involved in the bridging ligand. The last
reduction is located on the bipy part of bridging ligand I. This
is illustrated in Fig. 2 in which the different colors code the
reduction processes to particular ligand domains. The integral
of the last signal in the reductive region was larger than
expected. What other process is involved here could not be
resolved in detail, but a further reduction of the bridging ligand
is likely. This is underlined by the fact that a terpy without any
coordination can already be reduced twice.70
Although there are no metal–metal interactions in the
trinuclear complexes, there is evidence for electronic com-
munication between the ‘pillars’ of the bridging ligand. In
[{(bipy)2Ru(µ-I)}2Fe]6ϩ, for example, the reduction of the bipy
part of bridging ligand I occurs at Ϫ2.22 V, i.e. a 0.34 V separ-
ation from the reductions of the terminal bipy ligands. This is
0.11 V more than the separation between the second and third
bipy-based reductions in [Ru(bipy)2(I)]2ϩ. This indicates that
the bipy part of ligand I in the trinuclear complexes ‘feels’ the
reduction of the terpy domain and therefore the enhanced
electron density of the coordinated terpy domain. Assuming
a communication between the terpy and the bipy part of the
bridging ligand also means that after the reduction of the terpy,
part of the additional negative charge can be off-loaded to the
bipy. This ability to stabilise the electron after reduction is
reflected by the lowered difference between the first and second
reductions of the two terpy ligands (∆E = 0.11 V) of the Fe
centre in [{(bipy)2Ru(µ-I)}2Fe]6ϩ compared to the model
compound [Fe(II)2]2ϩ (∆E = 0.15 V). Moreover, the two first
reductions of the multinuclear complexes are taking place at
less negative potentials compared with the mononuclear models
(see Table 1 and Fig. 4). This suggests a lowering of the LUMO
by the coordination of the bridging ligand, whereas the HOMO
(represented by the first oxidation) stays in place whether there
is a second or third metal coordinated or not. The general
Fig. 4 Correlation of reduction potentials of the trinuclear and
dinuclear complexes with model complexes.
from the bottom of the diagram upwards the following picture
appears. Modifying a polypyridine ligand by pyridine substit-
uents, e.g. [Ru(terpy)2]2ϩ to [Ru(II)2]2ϩ, results in a shift to less
negative potential. The complex [Ru(bipy)2(I)]2ϩ also follows
this trend in relation to [Ru(bipy)3]2ϩ. Assuming that the
presence of the terpy substituent in [Ru(bipy)2(I)]2ϩ affects the
bipy domain in I more than the other two bipy ligands in
the complex, we can conclude that the first reduction of
[Ru(bipy)2(I)]2ϩ is centred on the bipy part of ligand I. Co-
ordination of the available terpy ligand in [Ru(bipy)2(I)]2ϩ with
an {Ru(terpy)}2ϩ fragment leads to [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ
,
the first reduction of which is at Ϫ1.34 V. We assign this peak to
a reduction of the bridging terpy domain. The basis of this
assignment is a comparison between the potential of the first
reduction in the trinuclear complexes, which was defined by the
titration experiment described above. The reduction at Ϫ1.84 V
in [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ is assigned to the terminal
terpy ligand by comparison with [Ru(terpy)2]2ϩ (Fig. 4). The
relationship between the three remaining peaks and the bipy
ligands in [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ is more difficult to estab-
lish. In comparison to [Ru(bipy)2(I)]2ϩ, the three reductions of
the bipy ligands in [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ are shifted to
more negative potentials. A reasonable explanation for this is
that the first reduction of the terpy ligand results in an increase
in negative charge density. Assuming that this affects the bipy
domain in I more than the terminal bipy ligands, one can
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1924

View Article Online
Table 2 Electronic spectroscopic data for the trinuclear and dinuclear complexes and model mononuclear compounds
Complex
λmax/nm (10Ϫ3 ε/MϪ1 cmϪ1
)
Ref.
[Ru(bipy)3]2ϩ
452 (13.0)
475 (11.6)
569 (24.5)
488 (30.9)
62
62
63
63
[Ru(terpy)2]2ϩ
307 (52.4)
324 (51.8)
312 (61.6)
315 (64.4)
270 (31.6)
276 (64.1)
273 (78.4)
275 (71.8)
[Fe(II)2]2ϩ
284 (72.0)
245 (39.4)
[Ru(II)2]2ϩ
238 (43.5)
238 (43.6)
243 (59.4)
242 (54.9)
[Os(II)2]2ϩ
668 (7.3)
486 (29.4)
[Ru(bipy)2(I)]2ϩ
461 (14.5)
287 (82.5)
287 (104.2)
287 (100)
285 (89)
285 (114)
286 (118)
287 (112.4)
287 (155.7)
287 (162.0)
[(bipy)2Ru(µ-I)Ru(terpy)]4ϩ
[(bipy)2Ru(µ-I)Os(terpy)]4ϩ
[(bipy)2Ru(µ-I)Ru(II)]4ϩ
[(bipy)2Ru(µ-I)Os(II)]4ϩ
[{(bipy)2Ru(µ-I)}2Ru]6ϩ
[{(bipy)2Ru(µ-I)}2Os]6ϩ
[{(bipy)2Ru(µ-I)}2Fe]6ϩ
[{(bipy)2Ru(µ-I)}2Co]6ϩ
500 (37.4)
498 (34)
502 (36)
670 (6)
678 (9)
686 (8.3)
444.0 (17.6) (sh)
440.0 (18.0) (sh)
460 (17.6)
460 (21.6)
462 (27.0)
242 (45)
235 (67)
242 (51.1)
241 (43.6)
244 (78.8)
243 (87.9)
501 (41)
512 (38.7)
510 (37.7)
591 (34.7)
470 (30.5)
properties described above for [{(bipy)2Ru(µ-I)}2Fe]6ϩ were also
observed for [{(bipy)2Ru(µ-I)}2Ru]6ϩ and [{(bipy)2Ru(µ-I)}2-
Os]6ϩ with respect to the shape and the position of the peaks.
For [{(bipy)2Ru(µ-I)}2Co]6ϩ, another reduction process was
found at Ϫ0.93 V which is assigned to the Co()/Co() reduc-
tion. The lower charge density on the Co centre is presumably
the reason for the less negative reduction potentials of the
ligands in [{(bipy)2Ru(µ-I)}2Co]6ϩ compared to the Fe, Ru and
Os-containing trinuclear counterparts (see Table 1).
UV-Vis and fluorescence spectroscopy
The electronic spectra of the model complexes [Ru(bipy)3]2ϩ
,
[Ru(terpy)2]2ϩ, [Fe(II)2][BF4]2, [Ru(II)2][BF4]2, [Os(II)2][BF4]2,
[Ru(bipy)2(I)]2ϩ, [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ, [(bipy)2Ru(µ-I)-
Fig.
5
Absorption spectra of [{(bipy)2Ru(µ-I)}2Ru][PF6]6 in
Os(terpy)]4ϩ, [{(bipy)2Ru(µ-I)}2Fe]6ϩ, [{(bipy)2Ru(µ-I)}2Ru]6ϩ
,
acetonitrile.
[{(bipy)2Ru(µ-I)}2Os]6ϩ and [{(bipy)2Ru(µ-I)}2Co]6ϩ were
measured in MeCN in the range of 200–800 nm. The absorp-
tion maxima are listed in Table 2.
on going to the trinuclear complexes. The electrochemical
results were consistent with the metal centred HOMO being
unaffected by the complexation of the terpy ligand, and,
combined with the electronic spectroscopic data, we can con-
clude that the lowest lying ligand orbital for the terminating Ru
groups is also not shifted due to the addition of metal ions.
The luminescence data (Table 3) were recorded in MeCN and
at room temperature. In order to probe the luminescence
properties selectively for the different metals involved, the
excitation wavelength was varied according to the absorption
properties. Exciting [Ru(bipy)2(I)]2ϩ at 454 nm results in an
emission (λem) at 628 nm, which is reasonably close to that
The mononuclear complexes are characterised by two spec-
tral regions. Between 200 and 400 nm, a number of peaks and
shoulders are present which we assign to ligand-centred π*
π
and π* n transitions. This region is followed at lower energy
by one broad absorption around 500 nm which represents the
metal-to-ligand charge transition (MLCT). Both types of trans-
itions are typical of polypyridine complexes.65 On going from
[Ru(bipy)2(I)]2ϩ to the dinuclear complexes [(bipy)2Ru(µ-I)-
Ru(terpy)]4ϩ and [(bipy)2Ru(µ-I)Os(terpy)]4ϩ, the electronic
spectra exhibit an additional shoulder for [(bipy)2Ru(µ-I)-
Ru(terpy)]4ϩ and
a
new absorption for [(bipy)2Ru(µ-I)-
observed for [Ru(bipy)3]2ϩ 62 On going from [Ru(bipy)2(I)]2ϩ to
.
Os(terpy)]4ϩ. These observations can be attributed to the addi-
tional MLCT taking place on the terpy-coordinated metal
centre. The MLCT band for [Ru(bipy)2(I)]2ϩ is red shifted on
going to the dinuclear complexes. This move to lower energy
is in qualitative agreement with the findings of the electro-
chemistry, i.e. the lowering of the LUMO in conjunction with
an unchanged energy of the HOMO. This assumes that the
redox orbitals are equivalent to the spectroscopic orbitals,
and that the electrochemical reductions are taking place
sequentially on the ligands.66,71
[(bipy)2Ru(µ-I)Ru(terpy)]4ϩ, one observes a significant red shift
of the emission band. Similar values for the emission were
also observed for [(bipy)2Ru(µ-I)Ru(II)]4ϩ and [{(bipy)2Ru-
(µ-I)}2Ru]6ϩ. Assuming that the emission takes place from a
3MLCT state, the observed shift of the emission band to lower
energy is in agreement with the shift of the electrochemically
observed LUMO to less negative potentials.62
The mixed metal di- and trinuclear complexes exhibit a
somewhat different picture to that of the homometallic
ruthenium complexes. When excited at around 500 nm,
[{(bipy)2Ru(µ-I)}2Os]6ϩ and [(bipy)2Ru(µ-I)Os(terpy)]4ϩ show
luminescence bands only at 767 and 793 nm, respectively. From
a comparison with the mononuclear Os complexes and based
on the electrochemical data, this points to an Os-centred
emission from the 3MLCT state. A surprising feature is that the
same emission was detected when [{(bipy)2Ru(µ-I)}2Os]6ϩ and
[(bipy)2Ru(µ-I)Os(terpy)]4ϩ were excited at around 450 nm, the
wavelength at which both the Ru- and Os-based MLCT transi-
tion should occur. This suggests that an efficient energy transfer
takes place between the Ru–bipy and the Os–terpy centres,
facilitated by the bridging ligand. This feature was observed for
all dinuclear and trinuclear complexes involving Os. For the
multinuclear, ruthenium-only complexes, only one emission
In the trinuclear complexes, two distinguishable absorptions
are observed in the MLCT region; a representative spectrum is
given in Fig. 5. The additional band can be attributed to a
second MLCT transition between the central metal (Fe, Ru or
Os) and one of the coordinated terpy ligands. With this
assumption and by comparison with the mononuclear model
complexes, it shows that in the trinuclear complexes the corre-
sponding absorption bands are shifted to lower energies. This
observation is also in agreement with the electrochemically
observed shift of the first reduction (representing the LUMO)
to less negative potentials as discussed above for the dinuclear
complexes. An interesting feature is that, in comparison to
[Ru(bipy)2(I)]2ϩ, the MLCT transition at 460 nm is not shifted
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1925

View Article Online
Table 3 Luminescent data of trinuclear, dinuclear and model mononuclear compounds at different excitation wavelengths (λex)
Complex
λem/nm (λex ≈ 450 nm (1MLCT))a
λem/nm (λex ≈ 500 nm (1MLCT))a
Ref.
62
65,74
65,74
[Ru(bipy)3]2ϩ
615
615
ca. 640 (480)
628 (454)
689 (460)
[Os(terpy)2]2ϩ
718
[Ru(terpy)2]2ϩ
[Ru(bipy)2(I)]2ϩ
[(bipy)2Ru(µ-I)Ru(terpy)]4ϩ
[(bipy)2Ru(µ-I)Os(terpy)]4ϩ
[(bipy)2Ru(µ-I)Ru(II)]4ϩ
[(bipy)2Ru(µ-I)Os(II)]4ϩ
[{(bipy)2Ru(µ-I)}2Ru]6ϩ
[{(bipy)2Ru(µ-I)}2Os]6ϩ
691 (500)
793 (498)
686 (503)
776 (501)
681 (513)
767 (510)
670 (460)
767 (446)
a Room temperature, acetonitrile solvent.
band was observed, independent of the excitation wavelength.
From electrochemical results, the LUMO of [{(bipy)2Ru-
(µ-I)}2Ru]6ϩ is localised on the {Ru(terpy)2}-fragment and it
could be suggested that the emission stems also from there,
implying an energy transfer from the bipy-coordinated to the
terpy-coordinated Ru. Similar results are obtained for [(bipy)2-
Ru(µ-I)Ru(terpy)]4ϩ. This is a somewhat surprising feature,
since complexes containing Ru(terpy) units are normally not
luminescent at room temperature due to fast deactivation
processes involving the 3MC state.42 Nevertheless, a few
examples of Ru(terpy) complexes with long-lived excited state
properties have recently been reported.72,73
In terms of supramolecular chemistry, this means that in the
ground state, the metal centres are independent entities and
therefore the dinuclear and trinuclear complexes are supra-
molecular species, whereas in the excited state, there is a strong
interaction and energy transfer occurs neglecting the independ-
ence of the parts of the entire complex.
Acknowledgements
We thank the Schweizerischer Nationalfonds zur Förderung
der wissenschaftlichen Forschung and the University of Basel
for the financial support of the work.
Electronic interactions
References
The difference between the interactions between the metal
centres and between the two sides of the bridging ligand is
remarkable and indicates that different orbitals are involved in
the two interactions. An interaction between the metal centres
is mediated by the overlap of the d(π)-orbitals with the
π-acceptor ligand, and by the extension of the π-system of the
ligand. In our case, the π-system between the terpy and bipy
domains of ligand I is obviously interrupted, and this is
consistent with the well-established non-planar conformation
observed for adjacent phenyl or pyridine rings.75 Our results
also indicate that purely electrostatic interactions do not
influence the position of the oxidation of the complexes con-
sidered in this paper. On the other hand, in the bridging ligand,
a lowering of the LUMO and an efficient distribution of the
negative charge seems to take place despite the interruption of
the conjugation between the terpy and bipy domains of the
bridge. The σ-bond skeleton of the polypyridine ligand and
field effects are the mediators of these effects. Altnernatively,
it can be speculated that the negative charge on the bridging
ligand leads to a planar conformation due to the gain of energy
by delocalising the additional charge over an extended con-
jugated system. This was suggested for related dinuclear
complexes the mechanism leading to a strong communication
between the metal centres.39 The shift of λem from [Ru(bipy)2-
(I)]2ϩ to [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ and [{(bipy)2Ru(µ-I)}2-
Ru]6ϩ, respectively, also shows that the excited electron is
already delocalised onto the terpy part of the bridging ligand.
1 F. Barigelletti and L. Flamigni, Chem. Soc. Rev., 2000, 29, 1.
2 M. D. Ward, Chem. Soc. Rev., 1995, 24, 121.
3 C. A. Bignozzi, R. Argazzi and C. J. Kleverlaan, Chem. Soc. Rev.,
2000, 29, 87.
4 L. C. Sun, L. Hammarstrom, B. Akermark and S. Styring, Chem.
Soc. Rev., 2001, 30, 36.
5 V. Balzani, P. Ceroni, A. Juris, M. Venturi, S. Campagna,
F. Puntoriero and S. Serroni, Coord. Chem. Rev., 2001, 219, 545.
6 A. C. Benniston, V. Grosshenny, A. Harriman and R. Ziessel,
Angew. Chem., Int. Ed. Engl., 1994, 33, 1884.
7 C. A. Bignozzi, R. Argazzi, J. R. Schoonover, K. C. Gordon,
R. B. Dyer and F. Scandola, Inorg. Chem., 1992, 31, 5260.
8 C. A. Bignozzi, R. Argazzi, C. Chiorboli, F. Scandola, R. B. Dyer,
J. R. Schoonover and T. J. Meyer, Inorg. Chem., 1994, 33, 1652.
9 C. A. Bignozzi, J. R. Schoonover and F. Scandola,
A
supramolecular approach to light harvesting and sensitisation of
wide-bandgap semiconductors: Antenna effects and charge
separation,in Molecular Level Artificial Photosynthetic Materials,
John Wiley & Sons Inc., New York, 1997, vol. 44, pp. 1–95.
10 A. K. Bilakhiya, B. Tyagi and P. Paul, Polyhedron, 2000, 19, 1233.
11 S. Campagna, S. Serroni, F. Puntoriero, F. Loiseau, L. De Cola,
C. L. Kleverlaan, J. Becher, A. P. Sorensen, P. Hascoat and
N. Thorup, Chem.–Eur. J., 2002, 8, 4461.
12 C. Chiorboli, M. A. J. Rodgers and F. Scandola, J. Am. Chem. Soc.,
2003, 125, 483.
13 R. L. Cleary, K. J. Byrom, D. A. Bardwell, J. C. Jeffery, M. D. Ward,
G. Calogero, N. Armaroli, L. Flamigni and F. Barigelletti, Inorg.
Chem., 1997, 36, 2601.
14 N. H. Damrauer and J. K. McCusker, J. Phys. Chem. A, 1999, 103,
8440.
15 L. De Cola and P. Belser, Coord. Chem. Rev., 1998, 177, 301.
16 L. De Cola, V. Balzani, F. Barigelletti, L. Flamigni,
P. Belser, A. von Zelewsky, M. Frank and F. Vogtle, Mol. Cryst. Liq.
Cryst. Sci. Technol. Sect. A: Mol. Cryst. Liq. Cryst., 1994, 252, 97.
17 H. Durr and S. Bossmann, Acc. Chem. Res., 2001, 34, 905.
18 M. Furue, K. Maruyama, Y. Kanematsu, T. Kushida and
M. Kamachi, Coord. Chem. Rev., 1994, 132, 201.
19 M. Furue, M. M. Ishibashi, A. Satoh, T. Oguni, K. Maruyama,
K. Sumi and M. Kamachi, Coord. Chem. Rev., 2000, 208, 103.
20 V. Grosshenny, A. Harriman, M. Hissler and R. Ziessel, J. Chem.
Soc., Faraday Trans., 1996, 92, 2223.
21 M. T. Indelli, F. Scandola, J. P. Collin, J. P. Sauvage and A. Sour,
Inorg. Chem., 1996, 35, 303.
22 O. Johansson, M. Borgstrom, R. Lomoth, M. Palmblad,
J. Bergquist, L. Hammarstrom, L. C. Sun and B. Akermark, Inorg.
Chem., 2003, 42, 2908.
Conclusions
We have shown that the metal centres of the di- and trinuclear
complexes bridged by the ‘back-to-back’ terpy-bipy ligand I do
not interact with each other in the ground state, whereas there
is significant communication between the two sides of the
bridging ligand upon reduction. Analysis of the absorption and
emission spectra indicate that there is energy transfer and,
therefore, an interaction between the different coordination
sites. This stresses the fact that one has to distinguish clearly
between interaction forces in the ground state (probed electro-
chemically) and in the excited state (probed spectroscopically).
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1926

View Article Online
23 A. Juris, L. Prodi, A. Harriman, R. Ziessel, M. Hissler, A. El-
ghayoury, F. Y. Wu, E. C. Riesgo and R. P. Thummel, Inorg. Chem.,
2000, 39, 3590.
24 A. F. Morales, G. Accorsi, N. Armaroli, F. Barigelletti, S. J. Pope
and M. D. Ward, Inorg. Chem., 2002, 41, 6711.
52 K. T. Potts and P. A. Winslow, J. Org. Chem., 1985, 50, 5405.
53 S. A. Savage, A. P. Smith and C. L. Fraser, J. Org. Chem., 1998, 63,
10048.
54 B. M. Kellybasetti, I. Krodkiewska, W. H. F. Sasse, G. P. Savage and
G. W. Simpson, Tetrahedron Lett., 1995, 36, 327.
25 T. Ohno, K. Nozaki, N. Ikeda and M. Haga, Adv. Chem. Ser., 1991,
215.
26 T. Ohno, K. Nozaki and M. Haga, Inorg. Chem., 1992, 31, 548.
27 E. A. Plummer, J. W. Hofstraat and L. De Cola, Dalton Trans., 2003,
2080.
55 B. Imperiali, T. J. Prins and S. L. Fisher, J. Org. Chem., 1993, 58,
1613.
56 H. Brederec, G. Simchen and R. Wahl, Chem. Ber./Recl., 1968, 101,
4048.
57 M. G. Vetelino and J. W. Coe, Tetrahedron Lett., 1994, 35, 219.
58 P. Dupau, T. Renouard and H. LeBozec, Tetrahedron Lett., 1996, 37,
7503.
59 G. Denti, S. Serroni, S. Campagna, A. Juris, M. Ciano and
V. Balzani, in Perspectives in Coordination Chemistry, ed.
A. F. Williams, C. Floriani and A. E. Merbach, VHCA and VCH,
Basel and Weinheim, 1992, pp. 153–164.
28 F. Scandola, R. Argazzi, C. A. Bignozzi, C. Chiorboli, M. Indelli
and M. A. Rampi, Coord. Chem. Rev., 1993, 125, 283.
29 F. Scandola, R. Argazzi, C. A. Bignozzi and M. T. Indelli,
J. Photochem. Photobiol. A: Chem., 1994, 82, 191.
30 M. Staffilani, P. Belser, F. Hartl, C. J. Kleverlaan and L. De Cola,
J. Phys. Chem. A, 2002, 106, 9242.
31 A. Vlcek, Coord. Chem. Rev., 1998, 177, 219.
60 E. C. Constable, C. E. Housecroft and Y. Tao, to be published.
61 All values listed for the trimeric and dimeric compounds were
determined by differential pulse voltammetry (DPV). These values
were compared with numbers determined by cyclic voltammetry
(CV), but did not differ significantly (<20 mV).
32 M. D. Ward, C. M. White, F. Barigelletti, N. Armaroli, G. Calogero
and L. Flamigni, Coord. Chem. Rev., 1998, 171, 481.
33 S. Fanni, C. Di Pietro, S. Serroni, S. Campagna and J G. Vos, Inorg.
Chem. Commun., 2000, 3, 42.
34 D. M. Guldi, M. Maggini, E. Menna, G. Scorrano, P. Ceroni,
M. Marcaccio, F. Paolucci and S. Roffia, Chem.–Eur. J., 2001, 7,
1597.
35 M. Marcaccio, F. Paolucci, C. Paradisi, M. Carano, S. Roffia,
C. Fontanesi, L. J. Yellowlees, S. Serroni, S. Campagna and
V. Balzani, J. Electroanal. Chem., 2002, 532, 99.
36 M. Marcaccio, F. Paolucci, C. Paradisi, S. Roffia, C. Fontanesi,
L. J. Yellowlees, S. Serroni, S. Campagna, C. Denti and V. Balzani,
J. Am. Chem. Soc., 1999, 121, 10081.
37 M. Krejcik and A. Vlcek, Inorg. Chem., 1992, 31, 2390.
38 M. Carano, P. Ceroni, C. Fontanesi, M. Marcaccio, F. Paolucci,
C. Paradisi and S. Roffia, Electrochim. Acta, 2001, 46, 3199.
39 S. Welter, K. Brunner, J. W. Hofstraat and L. De Cola, Nature, 2003,
421, 54.
62 For better comparison, the redox potentials of [Ru(bipy)3]2ϩ and
[Ru(terpy)2]2ϩ for measured under the same conditions and also with
Fc/Fcϩ as internal reference. The results are in accordance with the
values taken from: A. Juris, V. Balzani, F. Barigelletti, S. Campagna,
P. Belser and A. von Zelewsky, Coord. Chem. Rev., 1988, 84, 85–277.
63 E. C. Constable and A. Thompson, J. Chem. Soc., Dalton Trans.,
1994, 1409.
64 C. Creutz and H. Taube, J. Am. Chem. Soc., 1973, 95, 1086.
65 J. P. Sauvage, J. P. Collin, J. C. Chambron, S. Guillerez, C. Coudret,
V. Balzani, F. Barigelletti, L. De Cola and L. Flamigni, Chem. Rev.,
1994, 94, 993.
66 V. Balzani, A. Juris, M. Venturi, S. Campagna and S. Serroni, Chem.
Rev., 1996, 96, 759.
67 The DPV of [(bipy)2Ru(µ-I)Ru(terpy)]4ϩ was dominated by one
signal. In order to extract the redox potentials, the peak was fitted by
two gaussian curves assuming two one-electron processes lying
beneath.
40 E. C. Constable and M. D. Ward, J. Chem. Soc., Dalton Trans., 1990,
1405.
41 J. P. Collin, P. Laine, J. P. Launay, J. P. Sauvage and A. Sour, J. Chem.
Soc., Chem. Commun., 1993, 434.
68 P. Ceroni, A. Credi, V. Balzani, S. Campagna, G. S. Hanan,
C. R. Arana and J. M. Lehn, Eur. J. Inorg. Chem., 1999, 1409.
69 A. J. Bard and L. R. Forster, Electrochemical Methods,
Fundamentals and Applications, John Wiley & Sons, New York, 2nd
edn., 2001.
70 Two reductions for the terpy were observed, although in Fig. 3(A) –
dashed line – only one peak is shown.
71 It should also be mentioned that latest studies indicate that this
assumption is not entirely correct for related compounds such as Ru
and Os phenolate complexes: T. E. Keyes, D. Leane, R. J. Forster,
C. G. Coates, J. J. McGarvey, N. Nieuwenhuyzen, E. Figgemeier and
J. G. Vos, Inorg. Chem., 2002, 41, 5721.
72 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.
42 L. Hammarstrom, F. Barigelletti, L. Flamigni, M. T. Indelli,
N. Armaroli, G. Calogero, M. Guardigli, A. Sour, J. P. Collin and
J. P. Sauvage, J. Phys. Chem. A, 1997, 101, 9061.
43 A. J. Downard, G. E. Honey, L. F. Phillips and P. Steel, Inorg. Chem.,
1991, 30, 2259.
44 M. Beley, S. Chodorowskikimmes, J. P. Collin, P. Laine, J. P. Launay
and J. P. Sauvage, Angew. Chem., Int. Ed. Engl., 1994, 33, 1775.
45 E. Figgemeier, E. C. Constable, C. E. Housecroft and
Y. C. Zimmermann, 2004, submitted for publication.
46 B. P. Sullivan, D. J. Salmon and T. J. Meyer, Inorg. Chem., 1978, 17,
3334.
47 F. Kröhnke, Synthesis, 1976, 1.
48 E. C. Constable, A. Thompson, D. A. Tocher and M. A. M. Daniels,
New J. Chem., 1992, 16, 855.
49 D. A. Buckingham, A. M. Sargeson and F. P. Dwyer, Aust. J. Chem.,
1964, 17, 622.
73 M. Duati, S. Fanni and J. G. Vos, Inorg. Chem. Commun., 2000, 3,
68.
50 K. D. Demadis, E. S. El-Samanody, T. J. Meyer and P. S. White,
Polyhedron, 1999, 18, 1587.
51 T. L. J. Huang and D. J. Brewer, Can. J. Chem., 1981, 59, 1689.
74 J. R. Winkler, T. L. Netzel, C. Creutz and N. Sutin, J. Am. Chem.
Soc., 1987, 109, 2381.
75 Y. Kim and C. M. Lieber, Inorg. Chem., 1989, 28, 3990.
D a l t o n T r a n s . , 2 0 0 4 , 1 9 1 8 – 1 9 2 7
1927