Divinylphenylene-bridged diruthenium complexes bearing Ru(CO)CI(P iPr3)2 entities

Article abstract of DOI:10.1021/om0602660

The divinylphenylene-bridged diruthenium complexes (E,E)-[{(P ⁱPr₃)₂(CO)ClRu}₂(μ-HC=CHC ₆H₄CH=CH-1,3)] (m-2) and (E,E)-[{(PⁱPr ₃)₂(CO)ClRu}₂(μ-HC=CHC₆H ₄CH=CH-1,4)] (p-2) have been prepared and compared to their PPh ₃-containing analogues m-1 and p-1. The higher electron density at the metal atoms increases the contribution of the metal end groups to the bridge-dominated occupied frontier orbitals and stabilizes the various oxidized forms with respect to those of m-1 and p-1. This has been confirmed and quantified electrochemically, because the two reversible oxidation waves were observed at considerably lower potentials than for the PPh₃ complexes. Owing to their greater stability, the one- and two-electron-oxidized forms m-2ⁿ⁺ and p-2ⁿ⁺ of both complexes could be generated and spectroscopically characterized inside an optically transparent thin layer electrolysis cell. UV/vis/near-IR and ESR spectroelectrochemistry indicates that the oxidation processes are centered at the organic bridging ligand. σ-Bonded divinylphenylenes thus constitute an unusual class of noninnocent ligands for organometallic compounds. Electronic transitions observed for the mono- and dioxidized forms closely resemble those of donor-substituted phenylenevinylene compounds, including oligo(phenylenevinylenes) (OPVs) and poly-(phenylenevinylene) (PPV) in the respective oxidation states. Strong ESR signals and nearly isotropic g tensors are observed for the monocations in fluid and frozen solutions. The metal contribution to the redox orbitals is illustrated by a shift of the CO stretching bands to notably higher energies upon stepwise oxidation. The shifts strongly exceed those observed for the PPh₃ containing, six-coordinated species (E,E)-[{(PPh₃)₂(CO)Cl(L)Ru} ₂(μ-HC=CHC₆H₄CH=CH)]ⁿ⁺ (L = substituted pyridine). IR spectroelectrochemistry reveals the presence of two electronically different transition-metal moieties in m-2⁺, while they resemble each other more closely in p-2⁺. Differences in electronic coupling are illustrated by the charge distribution parameters calculated from the spectra. Bulk electrolysis experiments confirm the results from the in situ spectroelectrochemistry and the overall stoichiometry of the redox processes. Quantum-chemical calculations were performed in order to provide insight into the nature and composition of the frontier orbitals. The electronic transitions observed for the neutral forms were assigned by TD DFT. IR frequencies calculated for m-2 and p-2 in their various oxidation states retrace the experimental observations. They fail, however, in the case of m-2⁺, where a symmetrical structure is calculated, as opposed to the distinctly asymmetric electron distribution observed by IR spectroscopy. Geometry-optimized structures were calculated for all accessible oxidation states. The structural changes following stepwise oxidation agree well with the experimental findings: e.g., a successive low-energy shift of the C=C stretching vibration of the bridge. The radical cation m-2⁺ displays a broad composite electronic absorption band at low energy that extends into the mid-IR region.

Full text of DOI:10.1021/om0602660

Organometallics 2006, 25, 3701-3712
3701
Divinylphenylene-Bridged Diruthenium Complexes Bearing
Ru(CO)Cl(PiPr3)2 Entities†
Jo¨rg Maurer,‡,§ Biprajit Sarkar,§ Brigitte Schwederski,§ Wolfgang Kaim,§
Rainer F. Winter,*,‡ and Stanislav Za´lisˇ|
Institut fu¨r Anorganische Chemie der UniVersita¨t Regensburg, UniVersita¨tsstrasse 31,
D-93040 Regensburg, Germany, Institut fu¨r Anorganische Chemie der UniVersita¨t Stuttgart,
Pfaffenwaldring 55, D-70569 Stuttgart, Germany, and J. HeyroVsky´ Institute of Physical Chemistry,
Academy of Sciences of the Czech Republic, DolejsˇkoVa 3, Prague, Czech Republic
ReceiVed March 24, 2006
The divinylphenylene-bridged diruthenium complexes (E,E)-[{(PiPr3)2(CO)ClRu}2(µ-HCdCHC6H4CHd
CH-1,3)] (m-2) and (E,E)-[{(PiPr3)2(CO)ClRu}2(µ-HCdCHC6H4CHdCH-1,4)] (p-2) have been prepared
and compared to their PPh3-containing analogues m-1 and p-1. The higher electron density at the metal
atoms increases the contribution of the metal end groups to the bridge-dominated occupied frontier orbitals
and stabilizes the various oxidized forms with respect to those of m-1 and p-1. This has been confirmed
and quantified electrochemically, because the two reversible oxidation waves were observed at considerably
lower potentials than for the PPh3 complexes. Owing to their greater stability, the one- and two-electron-
oxidized forms m-2n+ and p-2n+ of both complexes could be generated and spectroscopically characterized
inside an optically transparent thin layer electrolysis cell. UV/vis/near-IR and ESR spectroelectrochemistry
indicates that the oxidation processes are centered at the organic bridging ligand. σ-Bonded divinyl-
phenylenes thus constitute an unusual class of “noninnocent” ligands for organometallic compounds.
Electronic transitions observed for the mono- and dioxidized forms closely resemble those of donor-
substituted phenylenevinylene compounds, including oligo(phenylenevinylenes) (OPVs) and poly-
(phenylenevinylene) (PPV) in the respective oxidation states. Strong ESR signals and nearly isotropic g
tensors are observed for the monocations in fluid and frozen solutions. The metal contribution to the
redox orbitals is illustrated by a shift of the CO stretching bands to notably higher energies upon stepwise
oxidation. The shifts strongly exceed those observed for the PPh3 containing, six-coordinated species
(E,E)-[{(PPh3)2(CO)Cl(L)Ru}2(µ-HCdCHC6H4CHdCH)]n+ (L ) substituted pyridine). IR spectroelec-
trochemistry reveals the presence of two electronically different transition-metal moieties in m-2+, while
they resemble each other more closely in p-2+. Differences in electronic coupling are illustrated by the
charge distribution parameters calculated from the spectra. Bulk electrolysis experiments confirm the
results from the in situ spectroelectrochemistry and the overall stoichiometry of the redox processes.
Quantum-chemical calculations were performed in order to provide insight into the nature and composition
of the frontier orbitals. The electronic transitions observed for the neutral forms were assigned by TD
DFT. IR frequencies calculated for m-2 and p-2 in their various oxidation states retrace the experimental
observations. They fail, however, in the case of m-2+, where a symmetrical structure is calculated, as
opposed to the distinctly asymmetric electron distribution observed by IR spectroscopy. Geometry-
optimized structures were calculated for all accessible oxidation states. The structural changes following
stepwise oxidation agree well with the experimental findings: e.g., a successive low-energy shift of the
CdC stretching vibration of the bridge. The radical cation m-2+ displays a broad composite electronic
absorption band at low energy that extends into the mid-IR region.
drain, while the bridge serves as the conduit. The voltage
gradient is then established by oxidizing or reducing one of the
terminal redox sites, thus generating a mixed-valent state.1 Here
it is possible to probe the degree of electron delocalization and
the rate at which the odd electron and the charge generated
during the redox process are transmitted between the end groups.
This can be done by analyzing ESR hyperfine splitting involving
active nuclei in the bridge or at the metal centers, by investigat-
ing the redox-induced changes in the shifts and band patterns
of charge-sensitive infrared- or Raman-active functions which
accompany electron transfer in vibrational spectroscopy, or by
analyzing the typical intervalence charge transfer (IVCT) bands
Introduction
Complexes in which two redox-active transition-metal moi-
eties are linked together by an unsaturated conjugated bridging
ligand have received great interest, first as a testing ground for
all aspects related to intramolecular electron transfer and then
as models for and principal constituents of possibly conducting
molecule-based “wires”. In adhering to this picture, the redox-
active end groups are mostly viewed as the voltage source or
* To whom correspondence should be addressed. E-mail: rainer.winter@
chemie.uni-regensburg.de. Fax: +49 941 9434488.
† Dedicated to Prof. Piero Zanello on the occasion of his 65th birthday.
‡ Universita¨t Regensburg.
§ Universita¨t Stuttgart.
| Academy of Sciences of the Czech Republic.
(1) Launay, J.-P. Chem. Soc. ReV. 2001, 30, 386.
10.1021/om0602660 CCC: $33.50 © 2006 American Chemical Society
Publication on Web 06/16/2006

3702 Organometallics, Vol. 25, No. 15, 2006
Maurer et al.
which often appear in the low-energy end of the visible region
or the near-IR region.
bridge, the redox processes in these systems are clearly
dominated by the metal.
Highly unsaturated hydrocarbon bridges offer an extended
π-system and have been recognized as particularly effective
coupling units. A vast number of oligoynediyl- and diethynyl-
arylene-bridged dimetal complexes have thus been synthesized
and investigated.2-24 Oligoenediyl bridges, though less well
precedented, are also efficient electronic coupling units with
sometimes even better performances than their oligoynediyl
counterparts.25,26 Efficient electronic coupling usually requires
that the unsaturated bridge contribute to the singly occupied
MO (SOMO). The bridge thus constitutes an integral part of
the entire redox system. C4-bridged dimetal complexes provide
particularly instructive examples, and redox-induced inter-
conversions between dimetal butadiynediyl and butatrienylidene
or between butatrienylidene and ethynylbis(carbyne) binding
motifs are known, depending on the electron count of the
metal.2,3,5,7,20,22,27-32 However, despite large contributions of the
We have recently observed that the divinylphenylene-bridged
diruthenium complex (E,E)-[{(PPh3)2(CO)Cl(4-EtOOCpy)Ru}2-
(µ-HCdCHC6H4CHdCH-1,3)] and its mononuclear styryl
derivative constitute examples where the redox processes are
dominated by the unsaturated organic ligand. Further investiga-
tions have shown that this holds irrespective of the complexes’
topology: i.e., for both the 1,3- and the 1,4-isomers.33 As a
consequence of the largely ligand-centered oxidations, none of
these complexes exhibited any IVCT band in its monooxidized,
formally mixed-valent state. Quantum-chemical calculations
suggest that the occupied frontier orbitals correspond to anti-
bonding interactions between higher lying, nearly degenerate
π levels of the bridge and lower lying metal dπ orbitals. It thus
seemed logical to increase the energies of the d energy levels
of the metal in order to seek for a more balanced electron
distribution within these systems. We herein show that changing
the {Ru(PPh3)2L} moieties to {Ru(PiPr3)2} induces a higher
metal character of the frontier orbitals. The Ru(PiPr3)2(CO)Cl
group stabilizes the oxidized forms to such a degree that even
the dications could be conveniently studied by vibrational and
electronic spectroscopy.
(2) Bruce, M. I.; Low, P. J.; Costuas, K.; Halet, J.-F.; Best, S. P.; Heath,
G. A. J. Am. Chem. Soc. 2000, 122, 1949.
(3) Bruce, M. I.; Ellis, B. G.; Low, P. J.; Skelton, B. W.; White, A. H.
Organometallics 2003, 22, 3184.
(4) Bruce, M. I.; Kostuas, K.; Davin, T.; Ellis, B. G.; Halet, J.-F.; Lapinte,
C.; Low, P. J.; Smith, M. E.; Skelton, B. W.; Toupet, L.; White, A. H.
Organometallics 2005, 24, 3864.
(5) Le Narvor, N.; Toupet, L.; Lapinte, C. J. Am. Chem. Soc. 1995, 117,
7129.
Experimental Section
(6) Coat, F.; Guillevic, M.-A.; Toupet, L.; Paul, F.; Lapinte, C.
Organometallics 1997, 16, 5988.
(7) Guillemot, M.; Toupet, L.; Lapinte, C. Organometallics 1998, 17,
1928.
(8) Paul, F.; Meyer, W. E.; Toupet, L.; Jiao, H.; Gladysz, J. A.; Lapinte,
C. J. Am. Chem. Soc. 2000, 122, 9405.
(9) Paul, F.; Costuas, K.; Ledoux, I.; Deveau, S.; Zyss, J.; Halet, J.-F.;
Lapinte, C. Organometallics 2002, 21, 5229.
(10) Coat, F.; Paul, F.; Lapinte, C.; Toupet, L.; Costuas, K.; Halet, J.-F.
J. Organomet. Chem. 2003, 683, 368.
(11) Jiao, H.; Costuas, K.; Gladysz, J. A.; Halet, J.-F.; Guillemot, M.;
Toupet, L.; Paul, F.; Lapinte, C. J. Am. Chem. Soc. 2003, 125, 9511.
(12) Weyland, T.; Lapinte, C.; Frapper, G.; Calhorda, M. J.; Halet, J.-
F.; Toupet, L. Organometallics 1997, 16, 2024.
(13) Weyland, T.; Costuas, K.; Mari, A.; Halet, J.-F.; Lapinte, C.
Organometallics 1998, 17, 5569.
(14) Weyland, T.; Costuas, K.; Toupet, L.; Halet, J.-F.; Lapinte, C.
Organometallics 2000, 19, 4228.
(15) Le Stang, S.; Paul, F.; Lapinte, C. Organometallics 2000, 19, 1035.
(16) Roue´, S.; Lapinte, C. J. Organomet. Chem. 2005, 690, 594.
(17) de Montigny, F.; Argouarch, G.; Costuas, K.; Halet, J.-F.; Roisnel,
T.; Toupet, L.; Lapinte, C. Organometallics 2005, 24, 4558.
(18) Dembinski, R.; Bartik, T.; Bartik, B.; Jaeger, M.; Gladysz, J. A. J.
Am. Chem. Soc. 2000, 122, 810.
(19) Szafert, S.; Gladysz, J. A. Chem. ReV. 2003, 103, 4175.
(20) Kheradmandan, S.; Heinze, K.; Schmalle, H. W.; Berke, H. Angew.
Chem. 1999, 111, 2412.
(21) Venkatesan, K.; Ferna´ndez, F. J.; Blacque, O.; Fox, T.; Alfonso,
M.; Schmalle, H. W.; Berke, H. Chem. Commun. 2003, 2006.
(22) Ferna´ndez, F. J.; Venkatesan, K.; Blacque, O.; Alfonso, M.;
Schmalle, H. W.; Berke, H. Chem. Eur. J. 2003, 9, 6192.
(23) Kheradmandan, S.; Venkatesan, K.; Blacque, O.; Schmalle, H. W.;
Berke, H. Chem. Eur. J. 2004, 10, 4872.
(24) Venkatesan, K.; Fox, T.; Schmalle, H. W.; Berke, H. Organo-
metallics 2005, 24, 2834.
(25) Sponsler, M. B. Organometallics 1995, 14, 1920.
(26) Chung, M.-C.; Gu, X.; Etzenhouser, B. A.; Spuches, A. M.; Rye,
P. T.; Seetharaman, S. K.; Rose, D. J.; Zubieta, J.; Sponsler, M. B.
Organometallics 2003, 22, 3485.
(27) Ferna´ndez, F. J.; Blacque, O.; Alfonso, M.; Berke, H. Chem.
Commun. 2001, 1266.
(28) Zhou, Y.; Seyler, J. W.; Weng, W.; Arif, A. M.; Gladysz, J. A. J.
Am. Chem. Soc. 1993, 115, 8509.
All manipulations were performed by standard Schlenk tech-
niques under an argon atmosphere. Solvents were dried by standard
procedures and degassed by saturation with argon prior to use. Ether
employed in the washing process was saturated with argon but
otherwise used as received. Infrared spectra were obtained on a
Perkin-Elmer Paragon 1000 PC FT-IR instrument. 1H (250.13
MHz), 13C (62.90 MHz), and 31P NMR spectra (101.26 MHz) were
recorded on a Bruker AC 250 spectrometer as CDCl3 solutions at
303 K or in the solvent indicated. The spectra were referenced to
the residual protonated solvent (1H), the solvent signal itself (13C),
or external H3PO4 (31P). If necessary, the assignment of 13C NMR
spectra was aided by DEPT-135 experiments. UV/vis spectra were
obtained on an Omega 10 spectrometer from Bruins Instruments
in HELMA quartz cuvettes whith 1 cm optical path lengths or on
a J&M TIDAS diode array spectrometer. The ESR equipment
consisted of a Bruker ESP 3000 spectrometer equipped with an
HP 5350 B frequency counter, a Bruker ER 035 M NMR
gaussmeter, and an ESR 900 continuous flow cryostat from Oxford
Instruments for low-temperature work. Elemental analyses (C, H,
N) were performed at in-house facilities. The equipment for
voltammetric and spectroelectrochemical studies and the conditions
employed in this work were as described elsewhere.34 The cell used
in quantitative coulometry experiments was a home-built three
compartment H-cell with a 3.5 cm diameter Pt/Ir alloy basket as
the working electrode and a smaller 1 cm diameter basket as the
counter electrode. A silver plate was used as the reference electrode.
The electrodes are welded to platinum wires (Supratronic), and the
platinum wires are welded to wires manufactured of Vakon alloy;
these in turn were sealed into 10 mm diameter glass tubes. The
electrodes were inserted into the cell via standard joints and fixed
by using Quickfit seals. The cell can be attached to a Schlenk line
through a joint attached to the central working electrode compart-
ment and charged via inlets on top of each compartment. Individual
compartments are separated via medium-porosity frits.
Molecular Orbital Calculations. Quantum-chemical studies
were performed without any symmetry constraints on PMe3-
(29) Brady, M.; Weng, W.; Zhou, Y.; Seyler, J. W.; Amoroso, A. J.;
Arif, A. M.; Bo¨hme, M.; Frenking, G.; Gladysz, J. A. J. Am. Chem. Soc.
1997, 119, 775.
(30) Woodworth, B. E.; White, P. S.; Templeton, J. L. J. Am. Chem.
Soc. 1997, 119, 828.
(31) Le Narvor, N.; Lapinte, C. J. Chem. Soc., Chem. Commun. 1993,
357.
(32) Coat, F.; Lapinte, C. Organometallics 1996, 15, 477.
(33) Maurer, J.; Winter, R.; Sarkar, B. Manuscript in preparation.
(34) Winter, R. F.; Klinkhammer, K.-W.; Za´lis, S. Organometallics 2001,
20, 1317.

DiVinylphenylene-Bridged Diruthenium Complexes
Organometallics, Vol. 25, No. 15, 2006 3703
substituted models m-2Me n+ and p-2Me n+ (n ) 0-2) instead of
the PiPr3 complexes. The ground-state electronic structure was
calculated by density functional theory (DFT) methods using the
ADF2005.0135,36 and Gaussian 0337 program packages. Within
ADF, Slater type orbital (STO) basis sets of triple-ú quality with
polarization functions were employed, with the exception of the
CH3 substituents on the P atoms, which were described by using a
double-ú basis. The inner shells were represented by a frozen-core
approximation: i.e., 1s for C and N, 1s-2p for P and Cl, and 1s-
3d for Ru were kept frozen. The calculations were done with a
functional including Becke’s gradient correction38 to the local
exchange expression in conjunction with Perdew’s gradient cor-
rection39 to the local correlation (ADF/BP). The following density
functionals were used within ADF: a local density approximation
(LDA) with VWN parametrization of electron gas data and a
functional including Becke’s gradient correction38 to the local
exchange expression in conjunction with Perdew’s gradient cor-
rection39 to the LDA expression (ADF/BP). The scalar relativistic
(SR) zero-order regular approximation (ZORA)40 was used within
this study. The g tensor was obtained from a spin-nonpolarized
wave function after incorporating the spin-orbit (SO) coupling by
first-order perturbation theory from the ZORA Hamiltonian in the
presence of a time-independent magnetic field.41,42
Within G03 calculations the quasirelativistic effective core
pseudopotentials and the corresponding optimized set of basis
functions for Ru43 and 6-31G* polarized double-ú basis sets44 for
the remaining atoms were employed together with the B3LYP45 or
BPW9138 functional.
Synthesis of [{Ru(PiPr3)2(CO)Cl}2(µ-CHdCHC6H4CHd
CH-1,3)] (m-2). To a stirred solution of [RuHCl(CO)(PiPr3)2] (0.27
g, 0.555 mmol) in CH2Cl2 (10 mL) was slowly added a solution of
1,3-diethynylbenzene (0.035 g, 0.277 mmol) in CH2Cl2 (10 mL).
After the reaction mixture was stirred for 30 min, the solvent was
removed under reduced pressure. The red product was washed twice
with n-hexane, filtered off, and dried under vacuum. Yield: 195
mg (64%). Anal. Calcd (found): C, 52.5 (51.31); H, 8.44 (8.47).
1H NMR (200 MHz, CDCl3, 293 K): δ 8.43 (dt, 2H, 3JH-H ) 13.27
Hz, 4JP-H ) 0.8 Hz, RuCdCH), 7.24 (s, 1H, phenyl (bridge)), 6.98
Chart 1
(t, 1H, 3JH-H ) 7.75 Hz, phenyl (bridge)), 6.68 (virtual t, 2H, JH-H
) 6.63 Hz, phenyl (bridge)), 5.89 (dt, 2H, 3JH-H ) 13.27 Hz, 3JP-H
) 1.9 Hz, RuHCdC), 2.83-2.26 (m, 12H, PiPr3), 1.52-1.16 (m,
72H, PiPr3). 31P NMR (80 MHz, CDCl3, 20 °C): δ 38.4 (s). IR
(KBr, ν in cm-1): 1910 (CO), 1555, 1470 (CdC, aryl, vinyl).
Synthesis of [{Ru(PiPr3)2(CO)Cl}2(µ-CHdCHC6H4CHdCH-
1,4)] (p-2). A solution of 1,4-diethynylbenzene (0,029 g, 0.299
mmol) in CH2Cl2 (10 mL) was slowly added to a stirred solution
of [RuHCl(CO)(PiPr3)2] (0.212 g, 0.435 mmol) in CH2Cl2 (10 mL).
The reaction mixture was stirred for 30 min. The solvent was
evaporated under reduced pressure, and the red product was washed
twice with n-hexane, filtered off, and dried under vacuum. Yield:
0.166 g (66%). Anal. Calcd (found): C, 52.50 (52.45); H, 8.44
(8.19). 1H NMR (250 MHz, CDCl3, 20 °C): δ 8.27 (dt, 2H, 3JH-H
) 13.2 Hz, 4JH-P ) 0.7 Hz, RuCdCH), 6.82 (virtual t, 2H, 4JH-H
) 6.6 Hz, phenyl (bridge)), 5.87 (dt, 2H, 3JH-H ) 13.16 Hz, 3JH-P
) 2.4 Hz, RuHCdC), 2.84-2.60 (m, 12H, PiPr3), 1.39-1.15 (m,
72H, PiPr3). 31P NMR (100 MHz, CDCl3, 20 °C): δ 38.2 (s). IR
(KBr, ν in cm-1): 1910 (CO), 1565, 1450 (CdC aryl, vinyl).
Results and Discussion
(35) te Velde, G.; Bickelhaupt, F. M.; van Gisbergen, S. J. A.; Fonseca
Guerra, C.; Baerends, E. J.; Snijders, J. G.; Ziegler, T. J. Comput. Chem.
2001, 22, 931.
Synthesis and Characterization. The dinuclear divinyl-
phenylene-bridged complex {(PiPr3)2(CO)ClRu}(µ-CHdCHC6-
H4HCdCH-1,3) (m-2) and its 1,4-isomer p-2 (see Chart 1) were
readily prepared by reacting the hydride HRu(CO)Cl(PiPr3)2 with
the corresponding diethynylbenzene in CH2Cl2 in a stoichio-
metric ratio of slightly larger than 2:1. Upon addition of the
alkyne the dark yellow solution of the hydride rapidly turned
red, indicating formation of the vinyl complexes, and the
reactions were complete within a few minutes. This short time
is despite the rather intricate reaction mechanism, which,
according to experimental and quantum-chemical studies con-
ducted for the osmium analogue, includes adduct formation and
the nonproductive formation of vinylidene complexes.46 The
pure products were obtained as intense red microcrystalline
solids by evaporating the solvent and washing the residue with
cold hexane to remove small quantities of excess hydride.
Product identity was confirmed by 31P NMR spectroscopy and
the characteristic NMR resonance signals of the vinyl protons.
These gave well-resolved doublet-of-triplet patterns with a large
3JHH coupling constant of around 13 Hz, indicating formal cis
addition of the Ru-H bond to the alkyne and smaller 3J and 4J
couplings to the phosphine P atoms. Unlike their PPh3-
substituted counterparts, complexes 2, despite being coordina-
(36) SCM, T. C. ADF2005.01; Vrije Universiteit, Amsterdam, Amster-
dam, The Netherlands, 2005.
(37) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K.
N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li,
X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A.
D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A.
G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian
03, revision B.2; Gaussian, Inc.: Wallingford, CT, 2004.
(38) Becke, A. D. Phys. ReV. A 1988, 38, 3098.
(39) Perdew, J. P. Phys. ReV. A 1986, 33, 8822.
(40) van Lenthe, E.; Ehlers, A. E.; Baerends, E. J. J. Chem. Phys. 1999,
110, 8943.
(41) van Lenthe, E.; van der Avoird, A.; Wormer, P. E. S. J. Chem.
Phys. 1997, 107, 2488.
(42) van Lenthe, E.; van der Avoird, A.; Wormer, P. E. S. J. Chem.
Phys. 1998, 108, 4783.
(43) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor.
Chim. Acta 1990, 77, 123.
(44) Hariharan, P. H.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213.
(45) Stephens, P. J.; Devlin, F. J.; Cabalowski, C. F.; Frisch, M. J. J.
Phys. Chem. 1994, 98, 11623.
(46) Marchenko, A. V.; Ge´rard, H.; Eisenstein, O.; Caulton, K. G. New
J. Chem. 2001, 25, 1244.

3704 Organometallics, Vol. 25, No. 15, 2006
Maurer et al.
For complexes of the general architecture {M}-bridge-{M}
with redox-active transition-metal-based end groups {M}, the
differences in half-wave potentials E1/2, ∆E1/2, are usually
considered as reflecting a kind of electronic interaction (or
communication) between the remote redox sites, with larger
splittings as the intermetal interactions increase. The closely
related isomeric diethynylphenylene-bridged diruthenium com-
plexes {Cl(dppm)2Ru}2(µ-CtCC6H4-CtC-1,3 or -1,4) display
∆E1/2 values of 190 and 300 mV, respectively.48 Further
instructive examples of such behavior are known from di-
ethynylphenylene-bridged dinuclear complexes of iron12,49 and
osmium.48 π-Conjugation is much more efficient for the para
isomer, and the potential splitting for the latter is hence
substantially larger than that for the meta form. A mutual relation
between differences in E1/2 and the degree of the electronic
interaction between the metal sites has been corroborated by
the determination of electronic coupling, as provided by the
position and the shape of the intervalence charge transfer (IVCT)
bands in the near-IR region (vide infra).12,48
Electrochemistry makes the different characteristics of the
divinylphenylene-bridged complexes, as opposed to the bis-
(ethynyl)phenylene-bridged counterparts, immediately apparent.
In the former, the ∆E1/2 value of the meta isomer slightly
exceeds that of the para isomer (270 vs 250 mV). This
counterintuitive result shows that a metal-based description as
above does not apply to the divinylphenylene systems presented
here. In complexes such as m-2 and p-2, the occupied frontier
levels (redox orbitals) are dominated by the unsaturated
ligands.47 In this respect, it is quite revealing to compare the
voltammetric responses of complexes 2 to those of purely
organic phenylenevinylene compounds. Under the appropriate
conditions, the latter are oxidized in sequential one-electron steps
with potential separations of ca. 90-180 mV that resemble those
observed for complexes 2. Oxidation potentials are lowered as
the number of phenylenevinylene repeat units increases or as
donor substituents are added.50-52 The redox potentials of even
heavily donor substituted phenylenevinylenes are, however,
substantially more anodic (>0.6 V vs Fc/Fc+) than those
observed for complexes 2. When they are viewed in this manner,
the {Ru(PR3)2(CO)Cl(L)} entities act as powerful electron-donor
substituents that even surpass common organic donor substit-
uents such as dimethylamino or alkoxy groups. This view also
agrees with the finding that the oxidation potentials of p-2,
where the {(PiPr3)2(CO)ClRu} moieties are in direct conjugation
with the divinylphenylene ligand, are distinctly lower than for
m-2, where such conjugation is expected to be less efficient.
The metal entities probably have higher contributions to the
redox orbitals than more conventional organic donor substitu-
ents. The site of attachment, i.e., the vinyl rather than the phenyl
groups, may also play a role.
Figure 1. Cyclic voltammograms of m-2 (upper curve) and p-2
(lower curve) at room temperature in CH2Cl2/0.2 M NBu4PF6 (V
) 0.1 V/s). Potentials are calibrated against the ferrocene/
ferrocenium standard.
Table 1. Redox Properties of the Five-Coordinate
Complexes {(PR3)2Ru(CO)Cl}nL
compd
E1/20/+, V
E1/2+/2+, V
∆E1/2, mV
m-2
p-2
m-1
p-1
0.190
-0.075
0.310a
0.460
0.175
0.490a
270
250
180
0.180a,b
a Half-wave potentials of chemically partially reversible processes at -78
°C. b Composite wave.
tively unsaturated, have only little propensity toward the addition
of a sixth ligand. This is certainly due to the superior donor
properties of the PiPr3 ligand as compared to those of PPh3 and
to the steric crowding. Higher electron density at the ruthenium
despite coordinative unsaturation is evident from the position
of the CO stretching band in the IR spectra. That band appears
at ca. 1910 cm-1 for complexes m-2 and p-2, ca. 15-20 cm-1
lower than for the six-coordinated bis(triphenylphosphine) 4-
picoline or bis(triphenylphosphine) 4-isonicotinate congeners.33
Electrochemistry. In CH2Cl2/TBAPF6 m-2 and p-2 undergo
two consecutive, chemically and electrochemically reversible
one-electron-oxidation processes (Figure 1, Table 1). This is in
stark contrast to the case for five-coordinated PPh3-containing
analogues, which give irreversible responses at ambient tem-
perature and only limited chemical reversibility as the temper-
ature is lowered to 195 K. Likewise, the half-wave potentials
of m-2 and p-2 are shifted by 120 and 255 mV, respectively,
toward more negative (cathodic) potentials in comparison to
the half-wave potentials (195 K) of the PPh3-substituted species.
The half-wave potentials of m-2 and p-2 are even somewhat
lower than those of coordinatively saturated pyridine adducts
with PPh3 ligands.47 Considering that the oxidation processes
mainly involve the bridging ligand (vide infra), the redox
potentials are, however, only an indirect probe of the electron
densities at the metal atoms; rather, they track the electron-
releasing properties of the {RuL2(CO)Cl} entities toward the
divinylphenylene ligands.
UV/Vis/Near-IR Spectroelectrochemistry. Verification of
bridge-dominated oxidation in the divinylphenylene-bridged
dinuclear complexes stems from the changes in their electronic
spectra upon oxidation. For metal-centered oxidation processes
one expects a lowering of the intensities of metal-to-ligand
(48) Colbert, M. C. B.; Lewis, J.; Long, N. J.; Raithby, P. R.; Younus,
M.; White, A. J. P.; Williams, D. J.; Payne, N. N.; Yellowlees, L.; Beljonne,
D.; Chawdhury, N.; Friend, R. H. Organometallics 1998, 17, 3034.
(49) Le Narvor, N.; Lapinte, C. Organometallics 1995, 14, 634.
(50) Kla¨rner, G.; Former, C.; Yan, X.; Richert, R.; Mu¨llen, K. AdV. Mater.
1996, 8, 932.
(51) Heinze, J.; Mortensen, J.; Mu¨llen, K.; Schenk, R. J. Chem. Soc.,
Chem. Commun. 1987, 701.
(52) Ndayikengurukiye, H.; Jacobs, S.; Tachelet, W.; Van der Looy, J.;
Pollaris, A.; Geise, H. J.; Claeys, M.; Kauffmann, J. M.; Janietz, S.
Tetrahedron 1997, 53, 13811.
(47) Maurer, J.; Winter, R. F.; Sarkar, B.; Fiedler, J.; Za´lisˇ, S. Chem.
Commun. 2004, 1900.

DiVinylphenylene-Bridged Diruthenium Complexes
Organometallics, Vol. 25, No. 15, 2006 3705
of ca. 1600 cm-1 for the higher energy band and of 1420 and
1165 cm-1 for the lower energy band have been observed for
m-2+ and p-2+ (see footnotes b and e of Table 2), and these
are again similar to those observed for OPV radical cations and
anions.55,56 It is interesting to note that these bands appear at
energies somewhat higher (higher energy band) or very similar
to (lower energy band) those for the radical anion derived from
diphenylethene (Eop ) 15 727 and 8307 cm-1). The similarities
in both band shape and positions argue for a predominantly
ligand-centered oxidation in the case of the PiPr3-containing
diruthenium complexes m-2 and p-2.
Further conversion from the mono- to the dioxidized forms
also led to distinct changes in the electronic spectra. As is
depicted in Figure 3, the vibrationally structured bands in the
visible and near-IR regions disappear while new, unstructured
bands at 396 and 626 nm (m-22+) or at 430 and 624 nm (p-
2
2+) grow in. There are again clean isosbestic points at 347,
1006, and 1393 cm-1 (m-2+f m-22+), and at 318, 383, 466,
and 860 nm (p-2+f p-22+). The band around 625 nm is similar
for both isomers in terms of position and intensity; it accounts
for the intense blue color of the dications. In contrast, there are
profound differences between the two isomers with regard to
the band at higher energy. This band becomes the dominant
absorption in the visible region for m-22+ but is only of low
intensity in the case of p-22+. Again we note similarities of the
optical spectra from the dications m-22+ and p-22+ with those
of the purely organic counterparts. Thus, the dioxidized form
of the 2,5-distyrylthiophene dication absorbs at 511 nm, while
maxima at 512 and 700 nm in heavily doped poly(phenylene-
vinylene) serve as a benchmark for highly delocalized divinyl-
phenylene- or distyrylbenzene-derived dications.55,56
IR Spectroelectrochemistry. We next sought to probe the
extent of the metal contribution to the SOMO by means of IR
spectroelectrochemistry. Owing to the prototypical π-acid
properties of the carbonyl ligand, the redox-induced CO
stretching band shifts faithfully mirror the electron density
changes at the metal. As an additional feature, IR spectroscopy
of dinuclear carbonyl complexes also provides a convenient
means to probe the intrinsic charge and valence delocalization:
i.e., identity (symmetry) or nonidentity (asymmetry) of the
bridged metal sites. The stronger the electronic coupling between
the carbonyl bearing metal entities, the more the CO band
positions of each subunit in the mixed-valent state will deviate
from the position in a related mononuclear complex (in the
respective redox state) and the more closely the CO energies
of the individual subunits will resemble each other.
Figure 2. UV/Vis/near-IR spectra recorded during the first
oxidations of m-2 (upper graph) and p-2 (lower graph). Spikes
marked by asterisks denote instrumental artifacts.
charge transfer (MLCT) bands and the appearance of new
absorptions corresponding to the transfer of charge from any
donor ligand(s) to the now electron-deficient metal center
(LMCT). If, however, the oxidation involves the bridging ligand,
any new absorption generated in the oxidation process should
be more of a π f π* parentage and closely resemble the
absorptions observed for similar, purely organic congeners. This
is indeed the case here as follows from UV/vis/near-IR
spectroelectrochemistry (see Figure 2). The orange-red color
of the neutral starting compounds arises from transitions at 492
(p-2) or 513 nm (m-2) that involve an excitation mainly from
the bridging ligand to orbitals that are delocalized over the metal
and the phosphine ligands and are thus assigned as mixed LLCT/
LMCT transitions (vide infra). Upon sequential oxidation inside
an optically transparent thin layer electrolysis (OTTLE) cell,53
this band is gradually replaced by a structured, more intense
feature with the main peak at a slightly higher energy (see Figure
2). Another band with easily discernible vibrational progression
grows in at ca. 1260 nm (see footnotes b and e of Table 2). In
both cases there are two isosbestic points for the conversions
of the neutral to the monocationic forms at 302 and 414 nm for
p-2 and at 295 and 398 nm for the oxidation of the meta isomer
m-2. While the overall spectral patterns are very similar,
irrespective of the complexes’ topology, we note the signifi-
cantly larger band intensities for the para form as opposed to
those for the meta isomer. The new absorptions in the optical
and the near-infrared (near-IR) spectra resemble the lower
energy absorptions of the radical anions (or cations) of 1,2-
diphenylethylene and 1,4-distyrylbenzene, which are commonly
referred to as the A1 and A2 or C1 and C2 bands, respectively.
These organic radical cations and anions have been thoroughly
investigated as models of the polarons that exist in the charged
states of conducting polymers based on oligo- and poly-
(phenylenevinylene) (OPV, PPV).51,54-58 Vibrational spacings
Sequential oxidation of m-2 in an OTTLE cell first led to
the replacement of the single band of the neutral form at 1910
cm-1 by two equally intense features at 1915 and 1971 cm-1
(Figure 4a). At the same time, the HCdCH stretching bands of
the divinylphenylene bridge shift from 1577 and 1554 cm-1 to
1523 cm-1, along with a significant gain in intensity. Upon the
second oxidation, the two CO bands of m-2+ merge into a single
band at 1983 cm-1, again with a clean isosbestic point (Figure
4b).
When p-2 was oxidized to the monocation, the original CO
stretching band at 1910 cm-1 was gradually replaced by a
structured band with an apparent maximum at 1932 cm-1 and
(55) Deussen, M.; Ba¨ssler, H. Chem. Phys. 1992, 164, 247.
(56) Ba¨ssler, H.; Deussen, M.; Heun, S.; Lemmer, U.; Mahrt, R. F. Z.
Phys. Chem. 1994, 184, 233.
(53) Krejcik, M.; Danek, M.; Hartl, F. J. Electroanal. Chem. Interfacial
Electrochem. 1991, 317, 179.
(54) Heidenreich, A.; Mu¨nzel, N.; Schweig, A. Z. Naturforsch. 1986,
41a, 1415.
(57) Sakamoto, A.; Furukawa, Y.; Tasumi, M. J. Phys. Chem. 1994, 98,
4635.
(58) Sakamoto, A.; Furukuwa, Y.; Tasumi, M. J. Phys. Chem. 1992,
96, 3870.

3706 Organometallics, Vol. 25, No. 15, 2006
Table 2. Spectroscopic Parameters of m-2 and p-2 in Various Oxidation States
Maurer et al.
IR ν, cm-1
UV/vis λmax, nm
(1480-2200 cm-1
)
(ꢀmax, 10-3 M-1 cm-1
)
ESR A, G ()10-4 T)
m-2
1910, 1577, 1554
1971, 1915, 1524
308 (16.6), 377 (sh, 4.7), 513 (0.87)
308 (15.3), 386 (sh, 3.56), 531 (1.84),a
1265 (1.80),b 2330 (0.51)c
m-2+
T ) 230 K: giso ) 2.0274
A(1H) ) 10 (2 H), 8 (2 H); A(99/101Ru) ) 6
T ) 110 K: g ) 2.0113
m-22+
p-2
1983
1910, 1573, 1561
1942, 1932, 1915, 1519, 1503, 1481
309 (15.7), 396 (12.3), 626 (6.03)
353 (10.3), 405 (2.63), 503 (1.33)
346 (5.82), 585 (4.27),d 1255 (4.11)e
p-2+
T ) 293 K: giso ) 2.0278
A(1H) ) 9.5 (2 H), 8 (2 H); A(99/101Ru) ) 4.5
T ) 110 K: g ) 2.0221
p-22+
1991
266 (9.06), 430 (3.23), 624 (5.36)
a Data for main peak; resolved progressions at 489 and 581 nm. b Data for main peak; resolved progressions at 928 and 1078 nm. c Data for main peak
from deconvolution (see text). d Data for main peak; deconvolution gives additional progressions at 492 and 535 nm. e Data for main peak; deconvolution
gives additional progressions at 949 and 1067 nm.
Figure 3. UV/Vis/near-IR spectroelectrochemical traces recorded
Figure 4. IR spectroelectrochemistry of m-2 in DCE/NBu4PF6 at
during the second oxidation of m-2 (upper graph) and p-2 (lower
295 K: (upper graph) first oxidation; (lower graph) second
graph).
oxidation.
additional high- and low-energy shoulders (Figure 5a). Spectral
minor isomer.60 As the oxidation proceeds, the weak HCdCH
deconvolution placed the individual peaks at 1915, 1932, and
bands of p-2 at 1573 and 1561 cm-1 give way to much stronger
1942 cm-1, while indicating that the shoulders are associated
absorptions at 1519, 1503, and 1481 cm-1 of p-2+. During the
with band half-widths and areas larger than for the 1932 cm-1
second anodic step the band(s) of the monocation are replaced
feature. Solutions prepared by quantitative coulometry in a
by just one CO band for the fully oxidized dication p-22+, which
conventional bulk electrolysis cell displayed an identical band
is now located at 1991 cm-1 (Figure 5b). The energy of the
pattern (Figue S2, vide infra). The overall band shape thus points
HCdCH/PhCdC bands decreased further such that they were
to the presence of two coexisting species in solution. As a
possible explanation, we invoke two distinct rotamers that differ
with respect to the orientation of the {Ru(CO)Cl} subunits: i.e.,
a cisoid or transoid arrangement of the carbonyl groups (see
+
shifted to a range where the NBu4 cation of the supporting
electrolyte absorbs strongly. It was thus impossible to determine
the peak position of these bands for p-22+ from IR spectro-
electrochemistry.
Chart 2). DFT calculations indicate that both these rotamers
correspond to minima on the potential energy surface with only
(60) One should note that any dinuclear complex featuring two mono-
a small energy difference between them.59 The three-band
carbonylmetal entities should give rise to two nondegenerate CO bands
representing the symmetrical and antisymmetrical combinations of the
individual M-CO stretches. Calculated IR spectra for the two rotamers of
m-2 and p-2 in any of its oxidation states consequently give two closely
spaced bands with separations in the range of some 4-10 cm-1 for each
pair of bands. Only one band is, however, experimentally observed for both
neutral and dioxidized forms. Obviously broadening due to solute/solvent
interactions and rotational averaging reduces the band splitting beyond
recognition. The calculated larger CO band energy difference for m-22+
nevertheless agrees nicely with the notably large CO bandwidth of this
dication observed in our experiments.
pattern may then arise form an overlap between two pairs of
bands for the different rotamers or from that of a two-band
pattern for the major isomer and a single-band pattern for the
(59) The calculated energies represent gas-phase values. Relative energies
of both rotamers will be different in solution, owing to the differences in
the dipole moments (none for the transoid and ca. 3.19 D for the cisoid
conformation of m-2). A polar environment is thus expected to even decrease
the energy difference between these rotamers.

DiVinylphenylene-Bridged Diruthenium Complexes
Organometallics, Vol. 25, No. 15, 2006 3707
overall oxidation state, especially for the para isomer. In fact,
DFT calculations indicate that the metal contribution to the redox
orbital increases from 25% to 31% when going from p-2 to
p-22+
.
(iii) Two widely spaced CO stretching bands are present for
the monocation m-2+ (∆ν ) 56 cm-1), the one at lower energy
being only slightly shifted from the position in m-2 and the
one at higher energy being moderately shifted from that of
m-22+, and two more closely spaced absorptions for the major
form of p-2+ (∆ν ) 27 cm-1), indicating that these isomers
differ significantly with respect to the charge distribution across
the metal sites. We note here that the relative shifts of the CO
bands in the mixed-valent form in comparison to the neighboring
homovalent states provide the charge distribution parameter ∆F,
which constitutes a quantitative measure for the extent of valence
delocalization.61,62 The latter is defined as
∆F ) (∆νox - ∆νred)/2[ν′(ox) - ν′(red)]
(1)
where ν′(ox) and ν′(red) denote the CO stretching wavenumbers
of the fully oxidized and the fully reduced forms while ∆νox
and ∆νred are the energy differences between the band positions
of the fully oxidized species and the higher energy band of the
monooxidized form and between the lower energy band of the
intermediate and that of the fully reduced form, respectively.61,62
The ∆F parameter is expected to increase with increasing
metal-metal interactions in the mixed-valent state, approaching
values of 0.5 in the completely delocalized case (class III) and
a value of 0.0 in the completely localized limit (class I). This
model works under the premise that the metal contribution to
the corresponding frontier orbitals remains constant throughout
the oxidation processes under consideration, which may not
necessarily be the case here. A ∆F value of 0.12 was calculated
for m-2+, whereas that of 0.31 for p-2+ is significantly larger.
The latter value was derived under the assumption that the
outermost features of the composite CO band belong to the same
species. In other words, the charge is more evenly spread across
the entire dimetal-divinylphenylene unit in p-2+ but is distinctly
localized on half of the molecule in m-2+. This finding is
consistent with what is known from phenylenevinylene-derived
polarons. While in the para-linked oligomers charge and spin
are delocalized over rather long distances,56,63,64 the meta isomers
tend to have localized charge and spin on just one styryl unit.64
ESR Spectroscopy. ESR spectroscopy constitutes a highly
sensitive method in order to elucidate the metal versus (organic)
ligand contributions to the SOMO of paramagnetic species. Such
information can be obtained from the isotropic g values, from
the g-tensor anisotropy measured in glassy frozen solutions or
solid samples, and from the hyperfine coupling, either to the
metal atom(s) or to other ESR-active nuclei of coordinated
ligands. With ruthenium as the metal, the situation is far from
ideal: rapid relaxation often renders Ru(III) species ESR silent
in fluid solution, and hyperfine splitting to 101Ru or 99Ru (17.0
and 12.7%, respectively; I ) 5/2) may be too small to be
observed in the frozen state. The intense ESR signals of m-2+
and p-2+ around giso ≈ 2.028 in fluid solution make therefore
a strong case for ligand-dominated oxidation underpinning the
results from UV/vis and IR spectroelectrochemistry. Interest-
Figure 5. IR spectroelectrochemistry of p-2 in DCE/NBu4PF6 at
295 K: (upper graph) first oxidation; (lower graph) second
oxidation.
Chart 2
Three points are of special interest.
(i) The overall CO band shift upon twofold oxidation of
complexes m-2 (73 cm-1) and p-2 (71 cm-1) is consistently
larger than that of the {(PPh3)2(CO)Cl(pyR)Ru}2(µ-HCd
CHC6H4CHdCH) counterparts (ca. 45 cm-1). This attests to
higher contribution from the ruthenium entities to the redox
orbitals of m-2n+ and p-2n+, owing to the higher d-orbital
energies in the more electron rich PiPr3-containing complexes.
The larger CO band shift upon oxidation may, however, also
rely on the absence of the pyridine ligand, which, as the only
weak π-acceptor, can serve to buffer the loss of electron density
at the metal in the six-coordinated congeners. There are indeed
small, yet clearly notable shifts of the ester bands of the
isonicotinate ligands (R ) COOEt) for the PPh3-derived
complexes.
(ii) The average shifts of the CO stretching bands for the
first oxidation step are consistently smaller than those for the
second oxidation process. With an average value of 33 cm-1
for the first and of 40 cm-1 for the second oxidation step, the
difference is only moderate for m-2. For p-2, however, the
average shift difference is much more pronounced at ca. 20 cm-1
as opposed to 60 cm-1. This may indicate that the metal
contribution to the respective frontier orbital increases with the
(61) Stoll, M. E.; Lovelace, S. R.; Geiger, W. E.; Schimanke, H.; Hyla-
Kryspin, I.; Gleiter, R. J. Am. Chem. Soc. 1999, 121, 9343.
(62) Atwood, C. G.; Geiger, W. E. J. Am. Chem. Soc. 2000, 122, 5477.
(63) Rauscher, U.; Ba¨ssler, H.; Bradley, D. D. C.; Hennecke, M. Phys.
ReV. B 1990, 42, 9830.
(64) Karabunarliev, S.; Baumgarten, M.; Tyutyulkov, N.; Mu¨llen, K. J.
Phys. Chem. 1994, 98, 11892.

3708 Organometallics, Vol. 25, No. 15, 2006
Maurer et al.
S1 of the Supporting Information) and p-2+ are nearly super-
imposable, with only slightly different hyperfine coupling
constants. This parallels the phenomenon of time-dependent
valence detrapping, where the assignment of a mixed-valent
system as either fully delocalized class III or localized class II
depends on the time scale of the experiment.62,66,67 In mixed-
valent chemistry, such behavior is typical of systems that are
close to the interesting borderline between these two regimes.68,69
In the case of m-2+, however, it may indicate a time-dependent
localization of charge and spin on predominantly one styryl
subunit or delocalization over both styryl subunits.
Quantitative Coulometry. Spectroelectrochemistry allows
us to simultaneously generate and spectroscopically investigate
oxidized or reduced forms associated with a redox-active
compound. It is an important merit of this in situ approach that
it may extend the detection limits to quite reactive oxidized or
reduced species that might otherwise escape detection under
the conditions of conventional bulk electrolysis.70,71 There is,
however, the possibility that chemically reversible behavior in
spectroelectrochemical experiments conceals some intricate
chemistry following electron transfer. As long as the overall
chemical reversibility is maintained upon completing a full
oxidation/reduction cycle and redox-induced interconversions
are fast, chemical processes such as redox-induced isomeriza-
tions72 may well go unnoticed under the conditions of spectro-
electrochemistry.
To confirm the results obtained for m-2 and p-2, we
supplemented the spectroelectrochemical studies conducted
within the OTTLE cell (UV/vis/near-IR, IR) or via in situ ESR
spectroscopy by bulk electrolyses inside a conventional elec-
trolysis cell. In each case, the working potential was adjusted
to about 150 mV positive to the half-wave potential of the first
anodic wave, and electrolysis was continued until the current
had dropped to about 2% of the initial value. At this stage close
to 1 faraday/mol of charge had been transferred, in accordance
with the proposed one-electron nature of each redox wave.
During electrolysis the color of the originally orange-red
solutions intensified to deep green (m-2+) or intense purple (p-
2+). Voltammograms recorded after the first oxidation step are
complementary to those of the nonoxidized forms, except for
the observation that the wave at lower potential now appeared
as a reduction wave. Close comparison revealed the presence
of an additional couple at a potential higher than that of any of
the original analyte waves. The associated peak currents are,
however, small and increase just slightly upon repetition of a
full reduction/reoxidation cycle. This indicates that the singly
oxidized forms are rather stable in solution at ambient temper-
ature, even on the extended time scale (ca. 30 min) of bulk
electrolyses. Samples were removed from the working electrode
compartment after monooxidation was completed and their IR,
UV/vis/near-IR, and ESR spectra recorded. The obtained spectra
Figure 6. ESR spectra of electrogenerated p-2+ in CH2Cl2/
NBu4PF6: (a) in fluid solution at 293 K, experimental spectrum
(upper trace) and simulated spectrum (lower trace, see text); (b)
experimental spectrum at 110 K.
ingly, both m-2+ and p-2+ exhibit resolved hyperfine splitting
to the vinyl protons and to the ruthenium nuclei. Experimental
spectra were well reproduced by simulations when three
different coupling parameters were applied: one for the vinylic
protons at the metal-bonded carbon atom, one for the protons
at the vinylic carbon atom which connects to the central
phenylene unit, and one for two equivalent ruthenium nuclei.
Figure 6a compares the experimental and simulated spectra of
p-2+. Spectra recorded in glassy frozen solution display only a
single broad resonance line without discernible g anisotropy at
X-band frequency (9.5 GHz). A certain contribution from the
ruthenium atoms to the SOMO is nevertheless indicated by the
relatively high isotropic g values in fluid solution and in frozen
samples (see Table 2). Organic ligand-centered radicals are
characterized by g values close to that of the free electron at
2.0023. The deviations observed for m-2+ and p-2+ from this
value are therefore indicative of nonnegligible metal contribution
to the SOMO. We also note that giso values of m-2+ and p-2+
are slightly larger than the value of 2.020 measured for (E,E)-
[{(PPh3)(CO)(Cl)(4-EtOOCpy)Ru}2(µ-CHdCH-C6H4CHdCH-
1,3)], again in line with some larger metal contribution for the
PiPr3-substituted congeners. Our results thus point to an es-
sentially ligand-centered oxidation and a +II oxidation state for
the ruthenium atoms for m-2+ and p-2+. Apart from being ESR
silent in fluid solution, organometallic paramagnetic Ru(III)
species usually display much larger g value anisotropies in their
frozen-solution spectra and substantially larger deviations of the
isotropic g value from that of the free electron. Typical examples
are the alkynyl complexes [(Me2bipy)(PPh3)2ClRu(CtCR)]+ 65
and allenylidene complexes [Cl(dppm)2Ru{dCdCdC(NRR′)-
(66) Atwood, C. G.; Geiger, W. E.; Rheingold, A. L. J. Am. Chem. Soc.
1993, 115, 5310.
(67) Ito, T.; Hamaguchi, T.; Nagino, H.; Yamaguchi, T.; Kido, H.;
Zavarine, I. S.; Richmond, T.; Washington, J.; Kubiak, C. P. J. Am. Chem.
Soc. 1999, 121, 4625.
(68) Lambert, C.; No¨ll, G. J. Am. Chem. Soc. 1999, 121, 8434.
(69) Nelsen, S. F. Chem. Eur. J. 2000, 6, 581.
(70) McCreery, R. L. Spectroelectrochemistry. In Electrochemical
Methods; Rossiter, B. W., Hamilton, J. F., Eds.; Wiley: New York, 1986;
Vol. II, p 591.
(71) Alessio, E.; Daff, S.; Elliot, M.; Iengo, E.; Jack, L. A.; Macnamara,
K. G.; Pratt, J. M.; Yellowlees, L. J. Spectroelectrochemical techniques. In
Trends in Molecular Electrochemistry; Pombeiro, A. J. L., Amatore, C.,
Eds.; Marcel Dekker: Lausanne, New York, 2004; p 339.
(72) Pombeiro, A. J. L.; Guedes da Silva, M. F. M. C.; Lemos, M. A.
N. D. A., Coord. Chem. ReV. 2001, 219-221, 53.
(but-3-enyl)]2+ 34
, where the giso values range from 2.127 to 2.170
with g value anisotropies larger than 0.5.
One finding deserves further mention here. While the radical
cation m-2+ of the meta isomer is clearly unsymmetrical on
the vibrational time scale of 1 × 10-11 to 1 × 10-12 s, it appears
to be symmetrical on the slower ESR time scale of 1 × 10-9 to
1 × 10-8 s. In fact, solution ESR spectra of m-2+ (see Figure
(65) Adams, C. J.; Pope, S. J. A. Inorg. Chem. 2004, 43, 3492.

DiVinylphenylene-Bridged Diruthenium Complexes
Organometallics, Vol. 25, No. 15, 2006 3709
Figure 7. Qualitative MO scheme of m-2Me. Arrows indicate the main contributions to the lowest allowed TD DFT calculated transitions.
The letters s and a label the symmetric and antisymmetric combinations of 4d Ru orbitals, respectively.
were essentially identical with those observed under the
conditions of spectroelectrochemistry (see Figures S2 and S3
in the Supporting Information). When we then performed the
second oxidation, the color of the solution rapidly changed to
deep blue and optical spectra recorded at this point showed the
625 nm absorption peak of the dioxidized forms. When
electrolyses were continued, the color gradually faded and
decomposition was indicated by the disappearance of the initial
redox waves. The fully oxidized solutions displayed a single
anodic wave which is identical with that observed as a minor
constituent after the first oxidation step. Samples taken from
the decomposed oxidized solution did not show any significant
absorption in the carbonyl stretching region of the IR spectrum.
This is highly suggestive of a decarbonylation or demetalation
process underlying the decomposition. While our bulk elec-
trolysis studies thus confirm the one-electron nature of each
voltammetric wave and the results obtained in the spectroelec-
trochemical investigations, they also demonstrate the superior
performance of the in situ approach when reactive redox
congeners are involved.
Quantum-Chemical Calculations. Quantum-chemical cal-
culations based on density functional (DFT) methods (see
Experimental Section) were performed in order to address the
issues of (i) comparing the electronic structures of the meta and
para isomers, (ii) assessing the geometrical structures and the
degree of torsion around the metal-vinyl and central phenylene
units, (iii) probing the effect of coordinative unsaturation of the
metal as opposed to the case for the six-coordinate pyridine
adducts, especially with respect to the LUMO and LUMO+1
levels, (iv) comparing the calculated and experimental spectro-
scopic parameters for these derivatives at their various oxidation
levels, and (v) probing the oxidation-state-dependent metal
contribution to the frontier orbitals, as was suggested by the
varying CO band shifts in IR spectroelectrochemistry. For the
sake of reducing the computational effort, the actual systems
were simplified by replacing the PiPr3 ligands by PMe3. These
model systems are denoted as m-2Me and p-2Me, respectively.
analysis are given in Tables S1 and S2 in the Supporting
Information. As is evident from these figures and tables, the
isomers closely resemble each other with respect to the
composition of the frontier orbitals that are relevant for their
optical and electrochemical properties. The HOMO and
HOMO-1 levels have large contributions from the π3 levels of
the divinylphenylene ligand interacting with the appropriate
combination of d orbitals at the metal atoms. In the case of
m-2Me, the HOMO and HOMO-1 orbitals are nearly degenerate,
with an energy gap of just 0.22 eV. In contrast, the separation
of these levels is 0.97 eV for p-2Me. Below is an again nearly
isoenergetic set derived from the symmetrical and antisym-
metrical combinations of {Ru(CO)Cl}-based orbitals with only
very little contribution from the bridge. Despite the vacancy of
the sixth coordination site, the unoccupied frontier orbitals,
LUMO and LUMO+1, are still very much centered on the metal
end groups and are of a mainly Ru/phosphine parentage. The
LUMO+2 essentially is a π-level of the divinylphenylene bridge
with only minor contributions from the ruthenium moieties.
Geometry optimizations of the m-2Me and p-2Me model
complexes in their various oxidation states reveal the impact
of the redox processes on the structures. Table 3 gives the
important bond parameters calculated for neutral systems.
Related five-coordinated mononuclear styryl or vinyl com-
plexes73-78 may serve as points of comparison for the neutral
forms. The overall agreement between the calculated and
experimental values for the vinyl complexes containing
{M(CO)Cl(PR3)2} is quite good. The corresponding parameters
of the meta divinylphenylene ligand are best compared with
those of the trivinylphenylene-bridged diruthenium complex
[{(Ph3P)2(py)(CO)ClRu}3(µ-HCdCH)3C6H3-1,3,5] reported by
(73) Alcock, N. W.; Cartwright, J.; Hill, A. F.; Marcellin, M.; Rawles,
H. M. J. Chem. Soc., Chem. Commun. 1995, 369.
(74) Torres, M. R.; Vegas, A.; Santos, A. J. Organomet. Chem. 1986,
309, 169.
(75) Werner, H.; Esteruelas, M. A.; Otto, H. Organometallics 1986, 5,
2295.
(76) Wakatsuki, Y.; Yamazaki, H.; Kumegawa, N.; Satoh, T.; Satoh, J.
Y. Organometallics 1991, 113, 9604.
(77) Maruyama, Y.; Yamamura, K.; Sagawa, T.; Katayama, H.; Ozawa,
F. Organometallics 2000, 19, 1308.
Figures 7 and 8 provide the MO schemes of m-2Me and p-2Me
in the frontier orbital region along with contour plots of the
most important orbitals, while the contributions of the individual
constituents to the respective orbitals as based on a Mulliken
(78) Jung, S.; Ilg, K.; Brandt, C. D.; Wolf, J.; Werner, H. Eur. J. Inorg.
Chem. 2004, 469.

3710 Organometallics, Vol. 25, No. 15, 2006
Maurer et al.
Figure 8. Qualitative MO scheme of p-2Me. Arrows indicate the main contributions to the lowest allowed TD DFT calculated transitions.
The letters s and a label the symmetric and antisymmetric combinations of 4d Ru orbitals, respectively.
Table 3. G03/B3LYP Calculated Symmetry-Averaged Bond
Lengths (Å) and Angles (deg) for the Model Compounds
m-2Me and p-2Me in Comparison with Experimental Dataa
reduction have very similar effects on the structural and
spectroscopic properties.55,57,58,63,82 Other structural changes
of the Ru(CO)Cl(PMe3)2 moiety upon oxidation consist of a
slight lengthening of the Ru-P bonds and a shortening of the
Ru-Cl bonds, indicating the loss of some electron density from
the metal. Alterations of these parameters are, however,
considerably smaller than one would expect for the loss of a
full unit charge from the metal.83-85 All of these data, along
with the overall composition of the HOMO and HOMO-1
orbitals, point to a strong involvement of the divinylphenylene
bridge in the oxidation processes. Electron density difference
plots for the 0/+ states (+/2+ density difference plots are
similar), as are depicted in Figure S4 in the Supporting
Information, nicely confirm these findings and are strongly
supportive of our experimental results. They show that the loss
of electron density from the organic π-system (40.7% for the
first and 28.0% for the second oxidation of m-2, 38.2% for the
first and 27.6% for the second oxidation of p-2) largely
outweighs that of the ruthenium centers (13.3 and 9.5% for m-2,
11.3 and 10.6% for p-2) for both oxidation steps, regardless of
the topology of the bridge. We also note that the DFT
calculations predict that the doubly oxidized para isomer
possesses a singlet ground state with the triplet state lying ca.
0.5 eV higher in energy. In contrast to this, the meta isomer
has a triplet ground state with an energy difference to the higher
lying singlet state of 0.15 eV.
m-2Me
p-2Me
calcd
calcd
exptla
Ru-C(CO)
Ru-Cl
Ru-C1
Ru-P
C1-C2
C2-C3
C3-C4
C4-C5
C-O
1.822
2.457
2.037
2.388
1.349
1.479
1.404
1.393
1.170
1.810
2.467
2.044
2.396
1.323
1.472
1.389
1.393
1.139
1.823
2.458
2.036
2.389
1.351
1.474
1.412
1.443
1.170
Ru-C1-C2
122.6
126.9
13.0
135.1
127.3
14.4
121.8
127.5
0.1
C1-C2-C3
C1-C2-C3-C4
a Average values from ref 79 for [{(Ph3P)2(py)(CO)ClRu}3(µ-HCdCH)3-
C6H3-1,3,5].
Jia and co-workers.79 The latter, however, contains six-
coordinate metal centers. We also note the close agreement with
respect to the dihedral angles between the central benzene ring
and the vinyl groups of m-2 (calculated value, 13°; experimental
values, 13.3-15.0°).79 The calculated dihedral angle for p-2Me
approaches 0°. Upon sequential oxidation we note a stepwise
shortening of the Ru-C (average ∆r(Ru-C) ≈ 0.07 Å) and
CPh-Cvinyl bonds (∆r ≈ 0.03 Å) and a concomitant lengthening
of the former vinylic CdC bonds (∆r ≈ 0.03 Å) for each
oxidation step (see Tables S3 and S4 in the Supporting
Information). This agrees well with the observed red shift of
Spectroscopic parameters were calculated for the p-2Me and
m-2Me models of the PiPr3 complexes in their various oxidation
states. Table 4 lists the calculated stretching frequencies for both
compounds in their most stable transoid forms, i.e. the form
with the chloro and carbonyl ligands at opposite sides, for all
oxidation states. Preliminary calculations indicate that differ-
ences in the cisoid forms are negligibly small. As for the νCO
ν
CdC in IR spectroelectrochemistry. There is also some quinoidal
distortion of the central phenylene ring that increases with the
overall oxidation state. Similar observations have been made
for the mono- and dioxidized forms of tetraphenylethenes80,81
and for the various reduced forms of m- and p-phenylene-
vinylenes.64 Note that in these latter systems oxidation and
(82) Obierski, J. M.; Greiner, A.; Ba¨ssler, H. Chem. Phys. Lett. 1991,
184, 391.
(83) Al Salih, T.; Duarte, M. T.; Frausto da Silva, J. J. R.; Galva˜o, A.
M.; Guedes da Silva, M. F. C.; Hitchcock, P. B.; Hughes, D. L.; Pickett,
C. J.; Pombeiro, A. J. L.; Richards, R. L. J. Chem. Soc., Dalton Trans.
1993, 3015.
(84) Guedes da Silva, M. F. C.; Ferreira, X. M. P.; Frau´sto da Silva, J.
J. R.; Pombeiro, A. J. L. J. Chem. Soc., Dalton Trans. 1998, 439.
(85) Carvalho, M. F. N. N.; Duarte, M. T.; Galva˜o, A. M.; Pombeiro,
A. J. L.; Henderson, R.; Fuess, H.; Svoboda, I. J. Organomet. Chem. 1999,
583, 56.
(79) Xia, H.; Wen, T. B.; Hu, Q. Y.; Wang, X.; Chen, X.; Shek, L. Y.;
Williams, I. D.; Wong, K. S.; Wong, G. K. L.; Jia, G. Organometallics
2005, 24, 562.
(80) Rathore, R.; Lindeman, S. V.; Kumar, A. S.; Kochi, J. K. J. Am.
Chem. Soc. 1998, 120, 6931.
(81) Sun, D.-L.; Rosokha, S. V.; Lindeman, S. V.; Kochi, J. K. J. Am.
Chem. Soc. 2003, 125, 15950.

DiVinylphenylene-Bridged Diruthenium Complexes
Organometallics, Vol. 25, No. 15, 2006 3711
Table 4. Selected G03/BPW91 Calculated Stretching
Frequencies (cm-1) for m-2Me n+ and p-2Me n+
frequency
n ) 0
n ) 1
n ) 2
n+
m-2Me
ν1 (CO)
ν2 (CO)
ν1 (CC)
1924.7
1928.2
1546.5
1948.5
1958.9
1491.5
1978.2
1987.2
1477.9
p-2Me n+
ν1 (CO)
ν2 (CO)
ν1 (CC)
1925.9
1927.3
1555.2
1955.8
1960.5
1513.5
1986.8
1990.8
1519.9
values, the calculations reproduce the overall behavior of the
CO bands rather well in that they predict a stepwise shift to
higher energies as the oxidation state increases. One should note
here that two CO bands are calculated for each isomer in every
oxidation state. They correspond to the symmetrical (ν1) and
the antisymmetrical (ν2) combinations of the CO stretching
vibrations, which are obviously nondegenerate. The ν1 and ν2
frequencies are rather close and are probably not resolved in
solution. The presence of various conformers in solution owing
to rotation of the {Ru(PiPr3)2(CO)Cl} entities around the Ru-
vinyl bonds may also contribute to some broadening of the CO
bands. Our observation of just one band in the neutral forms
and the dications is therefore still compatible with the compu-
tational results. A major discrepancy between calculated and
experimental data, however, arises in the case of m-2+. Quantum
Figure 9. IR spectra recorded during the stepwise oxidation of
m-2 to m-2+ on an extended energy scale.
calculated transitions for neutral p-2Me and m-2Me systems. TD
DFT calculations on the model complexes predict weak transi-
tions in the visible region at slightly lower energies than
observed (Tables S5 and S6 in the Supporting Information).
According to the calculations, this transition involves the set
of the closely spaced HOMO/HOMO-1 and LUMO/LUMO+1
levels (c1A in Figures 7, 8). Considering the composition of
these orbitals, the transition may be described as a mixed
bridging ligand-to-metal (LMCT) and bridging ligand-to-
phosphine (LLCT) electron transfer band. A transition at higher
energies which is only visible as a shoulder in the experimental
spectra has d-d character (d1A in Figures 7, 8), while the most
intense absorption in the near-UV region can be interpreted as
a π f π* transition within the extended delocalized chro-
mophore (i1A in Figures 7, 8). TD DFT calculations on closed
shell dication p-22+ indicate a red shift of the low-energy
transition, in agreement with experiment.
Observation of a Low-Energy Electronic Band for m-2+.
When monitoring the spectroscopic changes of vibrational
spectra over an extended energy range, we observed the
reversible growth (m-2 f m-2+; see Figure 9) and disappear-
ance (m-2+ f m-22+; see Figure S6 in the Supporting
Information) of a broad absorption covering the entire range
from 7500 to 2000 cm-1. The overall band shape was best
reproduced by invoking three overlapping bands at ca. 6550,
4300, and 2950 cm-1 with half-widths of 2600, 2100, and 1100
cm-1, respectively. Considerable noise in the experimental
spectrum, however, limits the accuracy of the absolute values.
chemistry gives ν1 and ν2, which differ by just about 10 cm-1
,
practically the same value as for the dication. This implies a
symmetrical structure of the radical with electronically equiva-
lent {Ru(CO)Cl(PR3)2(vinyl)} moieties, in stark contrast to the
experimental results. The much larger separation observed by
IR spectroelectrochemistry indicates two rather distinct metal
moieties, one of them resembling the reduced state and the other
the fully oxidized state. This inherent unsymmetry of the radical
cation m-2+ is thus not adequately reproduced by the present
calculations. Considering the well-known preference of the DFT
method for symmetrical structures and electron distributions,
this result is not wholly unexpected. It also suggests that the
calculated structural parameters of m-2+ are more or less the
average of distinct and dissimilar halves of the molecule. We
are going to address this issue by applying the broken symmetry
approach. In agreement with experiment, the calculations also
predict a stronger overall effect on the CO band shift for the
second oxidation and nicely reproduce the shift of the HCdCH
band upon sequential oxidation. Atom displacements that
accompany this vibration have been calculated and are shown
in Figure S5 in the Supporting Information. They clearly prove
a combined HCdCH stretch and dCH bend origin of this
fundamental.
ESR parameters calculated for the monooxidized forms of
m-2+ and p-2+ also agree reasonably with the experimental
values (Table 2), especially with regard to a small g anisotropy.
ADF/BP calculations, including spin-orbit coupling, give the
isotropic g value giso ) 2.039 and a splitting of the g tensor
with g1 ) 2.069, g2 ) 2.034, and g3 ) 2.016 for m-2Me+ and
giso ) 2.029 and g1 ) 2.050, g2 ) 2.026, and g3 ) 2.014 for
p-2Me. We note again the significant deviation of the g values
from that of the free electron. This finding provides direct
evidence for some metal contribution to the SOMO, as is also
borne out by the observed carbonyl shifts and the calculated
compositions of the SOMO orbitals.
Electronic absorption bands in this energy range are a typical
feature of weakly coupled mixed-valent systems where two or
more identical or similar redox sites are present in formally
different oxidation states. Similar bands have on occasion also
been observed for unsymmetrical complexes that combine
redox-active metals and “noninnocent ligands” with similar
redox potentials as the metal ions. Such bands have either ligand-
to-metal (LMCT) or metal-to-ligand (MLCT) character, depend-
ing on which of the two possible states, oxidized metal/reduced
ligand or reduced metal/oxidized ligand, is energetically more
favorable.86-88 These bands are conceptually important, in that
they provide the electronic coupling parameter HMM or HLM
,
respectively, on the basis of the Hush model.89-92
(86) Desjardins, P.; Yap, G. P. A.; Crutchley, R. J. Inorg. Chem. 1999,
38, 5901.
(87) Mosher, P. J.; Yap, G. A. P.; Crutchley, R. J. Inorg. Chem. 2001,
40, 1189.
While time-dependent DFT (TD DFT) calculations on the
open-shell systems m-2+ and p-2+ are currently being pursued,
we restrict ourselves to the assignment of the optical transitions
observed for the neutral forms. Arrows in Figures 7 and 8
indicate the main contributions to the lowest allowed TD DFT
(88) Lloveras, V.; Caballero, A.; Ta´rraga, A.; Velasco, M. D.; Espinosa,
A.; Wurst, K.; Evans, D. J.; Vidal-Gancedo, J.; Rovira, C.; Molina, P.;
Veciana, J. Eur. J.Inorg. Chem. 2005, 2436.
(89) Hush, N. S. Prog. Inorg. Chem. 1967, 8, 391.
(90) Hush, N. S. Coord. Chem. ReV. 1985, 64, 135.
(91) Creutz, C. Prog. Inorg. Chem. 1983, 30, 1.

3712 Organometallics, Vol. 25, No. 15, 2006
Maurer et al.
Interpretation of low-energy electronic absorption bands
within the framework of this model assumes that electronic
excitation involves the transfer of charge, either from the
formally reduced to the formally oxidized metal ion or between
the metal and the redox-active ligand. The radical cation m-2+
probably conforms to none of these cases. Here, the electronic
transition in the near-infrared region likely originates from the
excitation of an electron from the SOMO-1 level to the SOMO.
Preliminary calculations of m-2+ suggest that such a transition
may be observed. Since the SOMO and SOMO-1 levels of m-2+
are both metal-ligand antibonding combinations of the π3 level
of the divinylphenylene ligand and appropriate d orbitals, the
low-energy band of m-2+ is attributed to a π f π* type
transition. Its appearance in the near-IR/IR regime is then the
result of the near-degeneracy of these levels. This parallels the
appearance of similar HOMO-n f SOMO transitions in the
same spectral region for the radical cations of diethynylarylene
(arylene ) 2,5-thiophenylene, 9,10-anthracenylene)- and octa-
tetraynediyl-bridged diruthenium complexes.17 A similar band
should also be observed for the monoxidized form of the para
isomer. In p-2+ the energy gap between the SOMO-1 and the
SOMO levels is, however, much larger than in m-2+ and the
corresponding SOMO-1 f SOMO transition may thus be
obscured by the strong electronic band at 1255 nm. In this
context we note that this band is about four times more intense
for p-2+, as is the case for its meta isomer. The appearance of
three overlapping features instead of just one and their large
half-widths may point to different rotamers and to large
structural reorganization following excitation.
ClRu}2(µ-HCdCHC6H4CHdCH-1,4)] (p-2) and their PPh3-
substituted counterparts shows how the electronic balance of
bridge and metal contribution to the occupied frontier levels
can be controlled by the appropriate choice of coligands.
Decreasing the gap between the metal d orbitals and the higher
lying π levels of the bridge leads to a greater metal character
of the oxidized forms. This follows from the larger CO band
shifts upon stepwise oxidation of the PiPr3-substituted complexes
and from the observation of resolved hyperfine splittings
between the unpaired spin and the ruthenium nuclei. Replace-
ment of PPh3 by the more basic PiPr3 also stabilizes the partially
and fully oxidized forms to such an extent that it was possible
to investigate and characterize them by vibrational and electronic
spectroscopy. Solutions of the monooxidized forms are suf-
ficiently long-lived that these systems are viable synthetic
targets. Further studies that aim at their isolation and a more
comprehensive assessment of their electronic and magnetic
properties are currently underway in our laboratories.
Acknowledgment. Support of this work by the Deutsche
Forschungsgemeinschaft (Grants Wi 1262/7-1 and 436 TSE 113/
45/0-1), the EU (COST actions D35), and the Grant Agency of
the Academy of Sciences of the Czech Republic (S.Z., Grant
1ET400400413) is gratefully acknowledged. We also acknowl-
edge contributions from Barbara Vogel during an advanced
course in our laboratories.
Supporting Information Available: One-electron energies and
percentage compositions of selected orbitals of p-2Me and m-2Me
(Tables S1 and S2), symmetry-averaged bond lengths and angles
for p-2Me and m-2Me in their various oxidation states (Tables S3
and S4), lowest calculated excitation energies for p-2Me and m-2Me
as calculated by TDDFT (Tables S5 and S6), the ESR spectrum of
m-2Me+ (Figure S1) and the UV/vis/near-IR and IR spectra of
p-2Me+ and m-2Me+ generated by bulk electrolysis (Figures S2 and
S3), charge density difference plots for the first oxidations of p-2Me
and m-2Me (Figure S4), calculated atomic displacements for the
CdC stretch/CH bend of p-2Me (Figure S5), and IR spectro-
electrochemistry for further oxidation of m-2+ to m-22+ on a larger
energy scale (Figure S6). This material is available free of charge
via the Internet at http://pubs.acs.org.
Conclusions
Our investigations on divinylphenylene-bridged diruthenium
complexes establish that σ-bonded divinylphenylenes constitute
a unique class of noninnocent ligands for organometallic
compounds. As is shown by IR, UV/vis/near-IR, and ESR
spectroscopy, oxidation of these systems is largely centered on
the organic bridge. Comparison of (E,E)-[{(PiPr3)2(CO)ClRu}2-
(µ-HCdCHC6H4CHdCH-1,3)] (m-2) and (E,E)-[{(PiPr3)2(CO)-
(92) Brunschwig, B.; Creutz, C.; Sutin, N. Chem. Soc. ReV. 2002, 31,
168.
OM0602660