A donor - Acceptor - Donor bridging ligand in a class III mixed-valence complex

Article abstract of DOI:10.1021/ic001268u

The novel mononuclear and dinuclear complexes [Ru(trpy)(bpy)(apc)][PF₆] and [{Ru(trpy)(bpy)}₂(μ -adpc)][PF₆]₂ (bpy = 2,2′-bipyridine, trpy = 2,2′:6′,2″-terpyridine, apc^- = 4-azo(phenylcyanamido)benzene, and adpc^2- = 4,4′ -azodi(phenylcyanamido)) were synthesized and characterized by ¹H NMR, UV-vis, and cyclic voltammetry. Crystallography showed that the dinuclear Ru(II) complex crystallizes from diethyl ether/acetonitrile solution as [{Ru(trpy)(bpy)}₂gt-adpc)][PF₆] ₂·2(acetonitrile)·2(diethyl ether). Crystal structure data are as follows: crystal system triclinic, space group P1, with a, b, and c = 12.480(2), 13.090(3) and 14.147(3) A, respectively, α, β, and γ = 79.792(3), 68.027(3), and 64.447(3)°, respectively, V = 1933.3(6) A, and Z = 1. The structure was refined to a final R factor of 0.0421. The mixed-valence complex with metal ions, separated by a through-space distance of 19.5 A, is a class III system, having the comproportionation constant K_c. = 1.3 × 10¹³ and an intervalence band at 1920 nm (ε_max = 10 000 M^-1 cm^-1), in dimethylformamide solution. The results of this study strongly suggest that the bridging ligand adpc^2- can mediate metal-metal coupling through both hole-transfer and electron-transfer superexchange mechanisms.

Full text of DOI:10.1021/ic001268u

Inorg. Chem. 2001, 40, 1189-1195
1189
A Donor-Acceptor-Donor Bridging Ligand in a Class III Mixed-Valence Complex
Peter J. Mosher,† Glenn P. A. Yap,‡ and Robert J. Crutchley*,†
Ottawa-Carleton Chemistry Institute, Carleton University, 1125 Colonel By Drive, Ottawa,
Ontario, Canada K1S 5B6, and University of Ottawa, Ottawa, Ontario, Canada K1N 6N5
ReceiVed September 26, 2000
The novel mononuclear and dinuclear complexes [Ru(trpy)(bpy)(apc)][PF6] and [{Ru(trpy)(bpy)}2(µ-adpc)][PF6]2
(bpy ) 2,2′-bipyridine, trpy ) 2,2′:6′,2′′-terpyridine, apc- ) 4-azo(phenylcyanamido)benzene, and adpc2-
)
4,4′-azodi(phenylcyanamido)) were synthesized and characterized by 1H NMR, UV-vis, and cyclic voltammetry.
Crystallography showed that the dinuclear Ru(II) complex crystallizes from diethyl ether/acetonitrile solution as
[{Ru(trpy)(bpy)}2(µ-adpc)][PF6]2‚2(acetonitrile)‚2(diethyl ether). Crystal structure data are as follows: crystal
system triclinic, space group P1h, with a, b, and c ) 12.480(2), 13.090(3) and 14.147(3) Å, respectively, R, â, and
γ ) 79.792(3), 68.027(3), and 64.447(3)°, respectively, V ) 1933.3(6) Å3, and Z ) 1. The structure was refined
to a final R factor of 0.0421. The mixed-valence complex with metal ions, separated by a through-space distance
of 19.5 Å, is a class III system, having the comproportionation constant Kc ) 1.3 × 1013 and an intervalence
band at 1920 nm (ꢀmax ) 10 000 M-1 cm-1), in dimethylformamide solution. The results of this study strongly
suggest that the bridging ligand adpc2- can mediate metal-metal coupling through both hole-transfer and electron-
transfer superexchange mechanisms.
Introduction
mixed-valence complex [III,II] relative to its oxidized and
reduced forms can be quantified by the comproportionation
The creation of hybrid polymeric materials that possess useful
optical, magnetic, and/or electronic properties requires multi-
disciplinary research and is a very active field.1 In this regard,
the optoelectronic properties of a polymer depend on the degree
of conjugation between units making up the polymer chain. One
way of achieving a high degree of extended conjugation, and
hence a narrow band gap, has been to synthesize polythiophenes
made up of alternating donor/acceptor repeat units.2 Alterna-
tively, increasing the quinoid character of the polythiophene
backbone has been found to have a similar effect.3 For example,
isothianaphthene exhibits a band gap of ca. 1.0 eV, while that
of the parent polythiophene is ca. 2.0 eV.4 At the molecular
level, the energy between the highest occupied molecular orbital
(HOMO) and the lowest unoccupied molecular orbital (LUMO)
is analogous to the macroscopic band gap energy. In addition,
the mediation of electronic interactions by a bridging molecule
can be quantitatively modeled by the Marcus-Hush theory for
mixed-valence systems.5 Thus, the study of mixed-valence
complexes provides the knowledge to purposefully synthesize
novel polymeric materials.
equilibrium constant Kc:
Kc
[III,III] + [II,II] y
\z 2[III,II]
(1)
This permits the calculation of the free energy of compropor-
tionation (∆Gc ) - RT ln(Kc)), which is made up of the free
energy terms7
∆Gc ) ∆Gs + ∆Ge + ∆Gi + ∆Gr + ∆GAF
(2)
where ∆Gs reflects the statistical distribution of the compro-
portionation equilibrium, ∆Ge accounts for the electrostatic
repulsion of the two like-charged metal centers, ∆Gi is an
inductive factor dealing with competitive coordination of the
bridging ligand by the metal ions, ∆Gr is the free energy of
resonance exchange, and ∆GAF is the free energy of antiferro-
magnetic exchange. When metal-metal coupling is very strong,
the resonance exchange term ∆Gr largely determines the
magnitude of ∆Gc.7
Pentaammineruthenium dinuclear complexes incorporating
various bridging ligands have played an important role in
understanding how the medium influences coupling in mixed-
valence systems.8,9 For example, [{Ru(NH3)5}2(µ-4,4′-bipyri-
dine)]5+ is a class II system having Kc ) 20 and an MMCT
band centered at 1030 nm (ꢀ ) 920 M-1 cm-1).10 Its solvent-
dependent properties are largely consistent with the predictions
of Hush theory.8 Inserting an azo group between the pyridine
groups of the bridging ligand does not appear to dramatically
effect coupling. Thus, [{Ru(NH3)5}2(µ-4,4′-azopyridine)]5+ is
The strength of donor-acceptor coupling can range from
nonexistent (class I) to weak (class II) to very strong (class III).6
For a class III complex, there is no thermal barrier for electron
transfer between metal ions and, thus, the odd electron exists
in a delocalized state. Metal-metal coupling gives rise to a
metal-to-metal charge transfer (MMCT) transition in the com-
plex’s electronic absorption spectrum, and the stability of a
† Carleton University.
‡ University of Ottawa.
(1) Allcock, H. Science 1992, 255, 1106.
(2) Zhang, Q. T.; Tour, J. M. J. Am. Chem. Soc. 1998, 120, 5355.
(3) Kitamura, C.; Tanaka, S.; Yamashita, Y. Chem. Mater. 1996, 8, 570.
(4) Kobayashi, M.; Colaneri, M.; Boysel, M.; Wudl, F.; Heeger, A. J. J.
Phys. Chem. 1985, 82, 5717.
(7) Evans, C. E. B.; Naklicki, M. L.; Rezvani, A. R.; White, C. A.;
Kondratiev, V. V.; Crutchley, R. J. J. Am. Chem. Soc. 1998, 120,
13096.
(8) Creutz, C. Prog. Inorg. Chem. 1983, 30, 1.
(5) Brunschwig, B. S.; Sutin, N. Coord. Chem. ReV. 1999, 187, 233.
(6) Robin, M. B.; Day, P. AdV. Inorg. Chem. Radiochem. 1967, 10, 247.
(9) Crutchley, R. J. AdV. Inorg. Chem. 1994, 41, 273.
(10) Sutton, J. E.; Sutton, P. M.; Taube, H. Inorg. Chem. 1979, 18, 1017.
10.1021/ic001268u CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/07/2001

1190 Inorganic Chemistry, Vol. 40, No. 6, 2001
Mosher et al.
with gold-foil working and counter electrodes and a silver/silver chloride
reference electrode. OTTLE work done in DMF used ITO (indium-
tin oxide) coated glass for the working and counter electrodes. Elemental
analyses were performed by Canadian Microanalytical Services.
Materials. All of the reagents and solvents used were reagent grade
or better. 4,4′-Azodianiline was purchased from ACROS Organics and
used as received. Synthesis of [Ru(trpy)(bpy)Cl][PF6] has been previ-
ously described.16 ITO glass plates were purchased from Delta-
Technologies.
Preparation of 4,4′-Azodi(phenylcyanamide)‚H2O (adpcH2). Am-
monium thiocyanate (1.5 g, 20 mmol) was dissolved in 20 mL of
acetone and brought to a boil. To this was added dropwise a solution
of benzoyl chloride (2.8 g, 20 mmol) in 20 mL of acetone, and the
mixture was refluxed for an additional 15 min. 4,4′-Azodianiline (2.1
g, 9.9 mmol) was dissolved in 150 mL of boiling acetone and then
slowly added to the refluxing mixture above. The mixture was refluxed
for 1 h and then poured into a beaker containing 800 mL of water.
The brown benzoyl thiourea precipitate was removed by suction
filtration and dissolved in 300 mL of boiling 2 M NaOH. The resulting
deep red solution was boiled for 5 min and then cooled to 60 °C. Lead
acetate (7.5 g, 20 mmol) was dissolved in 20 mL of water and added
to the thiourea solution. The temperature was maintained at 60 °C for
15 min, forming a black PbS precipitate, and the mixture was then
filtered by suction into an ice-cooled flask. The filtrate was acidified
to pH 5 with the addition of 25 g of glacial acetic acid, and a green
precipitate was removed by suction filtration. Recrystallization from
700 mL of boiling acetone afforded a greenish gold solid. Yield: 1.3
g, 51%. Anal. Calcd for C14H12N6O: C, 59.99; H, 4.32; N, 29.98.
Found: C, 59.70; H, 4.39; N, 29.76. IR: ν(NCN) 2225 cm-1. 1H NMR
(400 MHz): 10.63 (2H, broad singlet), 7.89 (4H, doublet), 7.13 (4H,
doublet) ppm. Mp: 200-205 °C dec.
a class II system having Kc ) 40 and an MMCT band centered
at 1400 nm (ꢀ ) 500 M-1 cm-1).11 However, if a disulfide group
is inserted, coupling dramatically increases. [{Ru(NH3)5}2(µ-
4,4′-disulfidebipyridine)]5+ is a strongly coupled class II system
having Kc ) 8 × 104 and an MMCT band centered at 1500 nm
(ꢀ ) 4300 M-1 cm-1).12 As pyridine possesses low-energy π*
orbitals that permit it to act as a π-acceptor ligand and disulfide
is expected to be a π donor, this suggested to us that alternating
donor and acceptor groups in a bridging ligand might lead to
strong metal-metal coupling.
In this study, we have applied the above rationale to previous
work13 on the weakly coupled (Kc ) 16) mixed-valence
complex, [{Ru(NH3)5}2(µ-bp)]3+, where bp2- is
and have synthesized the novel bridging ligand 4,4′-azodi-
(phenylcyanamide) dianion, adpc2-
The Ru(II) dinuclear complex [{Ru(trpy)(bpy)}2(µ-adpc)][PF6]2
(bpy ) 2,2′-bipyridine, trpy ) 2,2′:6′,2′′-terpyridine, and adpc2-
) 4,4′-azodi(phenylcyanamido)) was also synthesized and
1
characterized by H NMR, UV-vis, cyclic voltammetry, and
Preparation of Tl2[adpc]. A 0.25 g (0.95 mmol) portion of adpcH2
was dissolved in 300 mL of boiling 2/1 acetone/water solution. To
this was added a 50 mL 2/1 acetone/water solution of thallium acetate
(0.53 g, 2.0 mmol). An orange precipitate quickly formed with the
addition of 1 mL of triethylamine. The mixture was boiled for 2 min
longer and then slowly cooled to -20 °C. Filtration afforded a bright
orange microcrystalline solid, which was washed with cold water and
acetone. Yield: 0.53 g, 83%. Anal. Calcd for C14H8N6Tl2: C, 25.13;
H, 1.21; N, 12.56. Found: C, 25.51; H, 1.31; N, 12.57. IR: ν(NCN)
crystallography. One-electron oxidation of this complex yields
a mixed-valence system whose properties are consistent with
class III behavior, despite metal ions being separated by 19.5
Å. The nature of the bridging ligand adpc2- was investigated
by ab initio methods to understand its effectiveness at mediating
metal-metal coupling.
Experimental Section
1
2078 and 2042 cm-1. H NMR (200 MHz): 7.41 (4H, doublet), 6.62
Equipment. UV-vis spectroscopy was performed on a CARY 5
UV-vis-near-IR spectrophotometer. IR spectra were taken with a
BOMEM Michelson-100 FT-IR spectrophotometer (KBr disks). 1H
NMR data from dimethyl sulfoxide-d6 solutions were obtained by using
a Bruker AMX-400 spectrometer or a Varian Gemini 200 spectrometer.
Cyclic voltammetry was performed using a BAS CV-27 voltammograph
and plotted on a BAS XY recorder. The sample cell consisted of a
double-walled glass crucible with an inner volume of ∼15 mL, which
was fitted with a Teflon lid incorporating a three-electrode system and
argon bubbler. The cell temperature was maintained at 25.0 ( 0.1 °C
by means of a HAAKE D8 recirculating bath. BAS 2013 Pt electrodes
(1.6 mm diameter) were used as the working and counter electrodes.
A silver wire functioned as a pseudo-reference electrode. Acetonitrile
(MeCN) was dried over P2O5 and vacuum-distilled. Nitromethane
(MeNO2) and dimethylformamide (DMF) were dried with anhydrous
alumina and vacuum-distilled. Tetrabutylammonium hexafluorophos-
phate (TBAH), purchased from Aldrich, was twice recrystallized from
1:1 ethanol/water and vacuum-dried at 110 °C. Ferrocene (E° ) 0.665
V vs NHE) and cobaltocenium hexafluorophosphate (E° ) -0.664 V
vs NHE) were used as internal references.14 An OTTLE cell was used
to perform the spectroelectrochemistry.15 The cell had interior dimen-
sions of roughly 1 × 2 cm with a path length of 0.2 mm and was fitted
(4H, doublet) ppm. Mp: 200-205 °C dec.
The poor solubility of Tl2[adpc] in DMF limited the accuracy of
the cyclic voltammetry study. This was corrected by preparing the
tetraphenylarsonium salt.
Preparation of [AsPh]2[adpc]‚H2O. adpcH2 (0.098 g, 0.37 mmol)
was dissolved in 50 mL of 2 M NaOHaq. To this was added a solution
of [AsPh4]Cl‚H2O (0.35 g, 0.80 mmol) in 20 mL of water, and the
mixture was stirred for 15 min. The red precipitate was collected and
washed with 30 mL of water and then recrystallized from boiling
MeCN. Yield: 0.25 g, 66%. Anal. Calcd for C62H50N6OAs2: C, 71.26;
1
H, 4.82; N, 8.04. Found: C, 70.91; H, 4.95; N, 8.06. The H NMR
spectrum of adpc2- is nearly identical with that obtained for Tl2[adpc].
1H NMR (200 MHz): 7.78 (40H, multiplet), 7.39 (4H, doublet), 6.59
(4H, doublet) ppm.
Preparation of [{Ru(trpy)(bpy)}2(µ-adpc)][PF6]2‚(ether). [Ru-
(bpy)(trpy)Cl][PF6] (0.60 g, 0.89 mmol) was dissolved in 100 mL of
DMF in a 250 mL round-bottom flask. Tl2[adpc] (0.30 g, 0.45 mmol)
was added, and the deep red solution was refluxed for 48 h. The reaction
mixture was then chilled to -20 °C and filtered to remove a fine white
TlCl precipitate. The filtrate was then concentrated to 30 mL by rotary
evaporation, and 600 mL of ether was added to precipitate the crude
product, which was then collected by suction filtration (0.56 g). The
crude product (0.3 g) was dissolved in 60 mL of 1:1 MeCN/toluene,
filtered, and purified by chromatography on a 50 cm × 3 cm diameter
(11) Launay, J.-P.; Tourrel-Paagis, M.; Libskier, J.-F.; Marvaud, V.;
Joachim, C. Inorg. Chem. 1991, 30, 1033.
(12) de Sousa Moreira, I.; Franco, D. W. J. Chem. Soc., Chem. Commun.
1992, 450.
(13) Aquino, M. A. S.; White, C. A.; Bensimon, C.; Greedan, J. E.;
Crutchley, R. J. Can. J. Chem. 1996, 74, 2201.
(14) Gennett, T.; Milner, D. F.; Weaver, M. J. J. Phys. Chem. 1985, 89,
2787.
(15) (a) Krejcik, M.; Danek, M.; Hartl, F. J. Electroanal. Chem. Interfacial
Electrochem. 1991, 317, 179. (b) Evans, C. E. B. Ph.D. Thesis,
Carleton University, 1997.
(16) Takeuchi, K. J.; Thompson, M. S.; Pipes, D. W.; Meyer, T. J. Inorg.
Chem. 1984, 23, 1845.

A Class III Mixed-Valence Complex
Inorganic Chemistry, Vol. 40, No. 6, 2001 1191
Table 1. Crystal Data for
column containing ∼250 g grade III alumina (Brockmann I, weakly
acidic, 150 mesh). Three bands containing starting material and a
mononuclear complex were eluted with 1:1 MeCN/toluene. A fourth
band was eluted with 500 mL of 3:1 MeCN/toluene and contained the
target complex, which crystallized from the solution upon evaporation
of the MeCN. Recrystallization was achieved by the slow diffusion of
ether into a saturated solution of the complex in MeCN. Yield: 0.16
g (43%). Anal. Calcd for C68H56N16OP2F12Ru2: C, 50.88; H, 3.52; N,
[{Ru(trpy)(bpy)}2(µ-adpc)][PF6]2‚2(acetonitrile)‚2(diethyl ether)
formula
fw
cryst syst
space group
unit cell dimens
a, b, c/Å
C76H72F12N18O2P2Ru2
1761.60
triclinic
P1h
12.480(2), 13.090(3), 14.147(3)
13.96. Found: C, 50.81; H, 3.33; N, 14.04. IR: ν(NCN) 2161 cm-1
.
R, â, γ/deg
79.792(3), 68.027(3), 64.477(3)
1933.3(6)
V/Å3
1H NMR (400 MHz): 9.65 (2H, doublet), 8.96 (2H, doublet), 8.90
(4H, doublet), 8.78 (4H, doublet), 8.69 (2H, doublet), 8.40 (2H, triplet),
8.32 (2H, triplet), 8.09 (6H, multiplet), 7.85 (2H, triplet), 7.73 (4H,
doublet), 7.46 (6H, multiplet), 7.27 (4H, doublet), 7.15 (2H, triplet),
6.02 (4H, doublet) ppm.
Z
1
calcd density/(g/cm3)
cryst dimens/mm
θ range/deg
limiting indices
no. of rflns collected
no. of unique rflns
abs cor
1.513
0.30 × 0.15 × 0.04
1.72-28.83
-15 e h e 16, -17 e k e 17, 0 e l e 18
15 188
Preparation of 4-Azo(phenylcyanamide)benzene Hydrate (apcH).
The same procedure as that for adpcH2 was followed. Recrystallization
from acetone/water (3/1) gave orange needles and crystalline flakes of
the product. Yield: 60%. Anal. Calcd for C13H11N4O0.5: C, 67.52; H,
4.79; N, 24.23. Found: C, 67.55; H, 4.79; N, 24.66. IR: ν(NCN) 2222
8763
semiempirical from equivalents
0.928 078-0.690 280
R1 ) 0.0421, wR2 ) 0.0682
1.044
transmissn range
R1, wR2 (I > 2σ(I))a
goodness-of-fit on F2
cm-1 1H NMR (400 MHz): 10.71 (1H, broad singlet), 7.95 (2H,
.
doublet), 7.87 (2H, doublet), 7.58 (3H, multiplet), 7.17 (2H, doublet)
ppm. Mp: 154-157 °C.
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|; wR2 ) (∑w(|Fo| - |Fc|)2/∑w|Fo| )1/2
.
Preparation of Tl[apc]. A procedure identical with that employed
to synthesize Tl2[adpc] was used. Orange crystalline flakes were isolated
directly from the reaction solution. Yield: 73%. Anal. Calcd for
C13H9N4Tl: C, 36.69; H, 2.13; N, 13.16. Found: C, 36.53; H, 2.14; N,
Table 2. Selected Crystal Structure Data for
[{Ru(trpy)(bpy)}2(µ-adpc)][PF6]2‚2(acetonitrile)‚2(diethyl ether)
Bond Lengthsa/Å
13.29. IR: ν(NCN) 2075 and 2042 cm-1
.
Ru-N(1)
Ru-N(2)
Ru-N(3)
Ru-N(4)
Ru-N(5)
Ru-N(6)
2.041(2)
2.069(2)
2.065(2)
1.964(2)
2.062(2)
2.046(2)
N(6)-C(26)
N(7)-C(26)
N(7)-C(27)
N(8)-C(30)
N(8)-N(8A)
1.164(3)
1.281(4)
1.394(3)
1.420(3)
1.274(4)
Preparation of [Ru(trpy)(bpy)(apc)][PF6]. The mononuclear com-
plex was synthesized in the same manner as the dinuclear species above,
except that only a 1/1 ratio of [Ru(trpy)(bpy)Cl]+ and ligand was used.
Yield: 53%. Anal. Calcd for C38H28N9F6PRu: C, 53.27; H, 3.29; N,
14.71. Found: C, 53.48; H, 3.42; N, 14.41. IR: ν(NCN) 2161 cm-1
.
1H NMR (400 MHz): 9.65 (1H, doublet), 8.95 (1H, doublet), 8.91
(2H, doublet), 8.79 (2H, doublet), 8.69 (1H, doublet), 8.40 (1H, triplet),
8.34 (1H, triplet), 8.10 (3H, multiplet), 7.86 (1H, triplet), 7.72 (4H,
multiplet), 7.46 (8H, multiplet), 7.15 (1H, triplet), 6.08 (2H, doublet)
ppm.
Bond Anglesa/deg
N(1)-Ru-N(2)
N(1)-Ru-N(5)
N(3)-Ru-N(2)
N(4)-Ru-N(3)
78.98(9) N(4)-Ru-N(5)
91.37(9) N(6)-C(26)-N(7)
99.81(9) C(26)-N(6)-Ru
79.75(9) C(26)-N(7)-C(27)
79.50(10)
173.1(3)
164.4(2)
118.7(2)
Crystallography. Plates of [{Ru(trpy)(bpy)}2(µ-adpc)][PF6]2‚
2(acetonitrile)‚2(diethyl ether) were grown by slow diffusion of diethyl
ether into an acetonitrile solution of the complex. The data were
collected on a 1K Siemens Smart CCD using Mo KR radiation (λ )
0.710 73 Å) at 203(2) K using an ω-scan technique and corrected for
absorptions using equivalent reflections.17 No symmetry higher than
triclinic was observed, and solution in the centric space group option
yielded chemically reasonable and computationally stable results of
refinement. The structure was solved by direct methods and refined
with full-matrix least-squares procedures. The molecular cation is
located at an inversion center. A molecule of acetonitrile and a molecule
of diethyl ether are each located cocrystallized in the asymmetric unit.
Anisotropic refinement was performed on all non-hydrogen atoms. All
hydrogen atoms were calculated. Scattering factors are contained in
the SHELXTL 5.1 program library.
N(5)-Ru-N(2) 100.94(9) N(8A)-N(8)-C(30) 114.0(3)
a Estimated standard deviations are in parentheses.
thenium(III) dinuclear complex but were unsuccessful in obtain-
ing a product that gave a reasonable elemental analysis.
Deep red-brown plates of the dinuclear Ru(II) complex [{Ru-
(trpy)(bpy)}2(µ-adpc)]2+ that were suitable for crystallography
were grown by diffusion of ether into a solution of the complex
in acetonitrile. Crystallography data and bond lengths and angles
are reported in Tables 1 and 2, respectively. An ORTEP drawing
of the complex cation is shown in Figure 1. The Ru(II) ion
occupies a pseudo-octahedral coordination sphere of nitrogen
donor atoms where the cyanamide group of the phenylcyana-
mide ligand is trans to a pyridine moiety of the bipyridine ligand.
The bridging adpc2- ligand is approximately planar, with the
cyanamide groups in an anti conformation relative to each other
and the azo group adopting the more thermodynamically stable
trans conformation. The planar geometry of the adpc2- bridging
ligand is suggested to allow for a more effective π delocalization
of the cyanamide groups with phenyl and azo groups. This
rationale has also been used to explain the planar conformation
of the 1,4-dicyanamidobenzene dianion and its derivatives,
observed in crystal structures of tetraphenylarsonium salts20 and
Ru(III) complexes.21 The cyanamide group is approximately
Results
The complexes of this study were synthesized by the
metathesis reaction of [Ru(trpy)(bpy)Cl]+ with the thallium salts
of apc- or adpc2- in refluxing DMF. Some purification by
column chromatography was required, but the complexes were
nevertheless isolated in good yields. We observed no evidence
for a photoinduced cis-trans isomerization of the azo group,18
but these studies were preliminary, and it may be that isomer-
ization back to the more stable trans isomer is rapid.19 Both
Ru(II) complexes are air-stable and can be readily recrystallized.
We also attempted to prepare the analogous pentaammineru-
(20) Aquino, M. A. S.; Crutchley, R. J.; Lee, F. L.; Gabe, E. J.; Bensimon,
C. Acta Crystallogr. 1993, C49, 1543.
(17) Blessing, R. Acta Crystallogr. 1995, A51, 33.
(18) Yutaka, T.; Kurihara, M.; Kubo, K.; Nishihara, H. Inorg. Chem. 2000,
(21) (a) Aquino, M. A. S.; Lee, F. L.; Gabe, E. J.; Bensimon, C.; Greedan,
J. E.; Crutchley, R. J. J. Am. Chem. Soc. 1992, 114, 5130. (b) Rezvani,
A. R.; Bensimon, C.; Cromp, B.; Reber, C.; Greedan, J. E.; Kondratiev,
V.; Crutchley, R. J. Inorg. Chem. 1997, 36, 3322. (c) Evans, C. E. B.;
Yap, G. P. A.; Crutchley, R. J. Inorg. Chem. 1998, 37, 6161.
39, 3438.
(19) Chambers, E. J.; Haworth, I. S. J. Chem. Soc., Chem. Commun. 1994,
1631.

1192 Inorganic Chemistry, Vol. 40, No. 6, 2001
Mosher et al.
Figure 1. ORTEP drawing of [{Ru(trpy)(bpy)}2(µ-adpc)]2+. The counterions, hydrogen atoms, and solvents of crystallization (two acetonitrile,
two diethyl ether) have been omitted for clarity.
Table 3. Cyclic Voltammetry Dataa of Azo Ligands and
Ru(trpy)(bpy) Complexes
the bridging ligand at these potentials appears to result in loss
of reversibility, as there is evidence of electrode deposition (a
cathodic spike appeared at -1.44 V vs NHE). This chemistry
ligand/complex
Ru(III/II) L(1-/2-) L(0/1-)
was not pursued.
Tl[apc]
0.86b,d
In Table 3, we have placed the Ru(III/II) couples25,26 of [Ru-
(trpy)(bpy)(2,4-Cl2pcyd)]+ and [{Ru(trpy)(bpy)}2(µ-dicyd)]2+
for comparison with the analogous complexes of apc- and
adpc2- ligands. The Ru(III/II) couple of a mononuclear ruthe-
nium complex approximates the energy of a ruthenium ion that
is not stabilized by the resonance exchange experienced by
ruthenium ions in a dinuclear mixed-valence complex, provided
their coordination spheres are similar. Qualitatively, it is
expected that, for a dinuclear complex, the Ru(III/II) couple
forming the [III,II] complex should be more positive of this
value, while the Ru(III/II) couple forming the [II,II] complex
should be more negative of this value. Thus, in Table 3, the
Ru(III/II) couples of the apc- and adpc2- complexes ap-
proximately follow the expected behavior, but the phenylcy-
anamide and dicyd2- complexes do not. The reason for this is
that antiferromagnetic exchange in [{Ru(trpy)(bpy)}2(µ-di-
cyd)]4+ stabilizes the [III,III] complex relative to the mixed-
valence complex, thereby shifting the corresponding Ru(III,II)
couple to more negative potential.27 This implies that antifer-
romagnetic exchange is not important in [{Ru(trpy)(bpy)}2(µ-
adpc)]4+. Unfortunately, the instability of this [III,III] complex
prevented solid-state magnetic measurements.
The electronic absorption data of the thallium(I) salts of apc-
and adpc2- and their Ru(II) and Ru(III) complexes in DMF have
been placed in Table 4. The anionic ligands in DMF show a
strong visible absorption, which we have assigned to a
cyanamide-to-azo group intraligand charge transfer (ILCT)
transition. When these ligands are incorporated into the Ru(II)
complex, this transition is still observed, but it is now ap-
proximately coincident in energy with metal-to-ligand charge
transfer (MLCT) transitions of the Ru(II)-bpy and Ru(II)-
trpy chromophores.
[AsPh4]2[adpc]
0.48b
0.63b,e
[{Ru(trpy)(bpy)(apc)]+
0.96c
1.06b
[{Ru(trpy)(bpy)}2(µ-adpc)]2+
0.79,c 1.57c
0.91,b ∼1.6b
[{Ru(trpy)(bpy)(2,4-Cl2pcyd)]+ f 1.0b
[{Ru(trpy)(bpy)}2(µ-dicyd)]2+ g 0.69,b 0.25b
a Conditions: in V vs NHE at 25 °C; scan rate of 0.1 V/s; 0.1 M
TBAH electrolyte; ferrocene (Fc+/Fc ) 0.665 V vs NHE) or cobalto-
cenium hexafluorophosphate (E° ) -0.664 V vs NHE) used as internal
references. b In DMF. c In MeCN. d Anodic wave only. e Partially
reversible. f Data from ref 25. g Data from ref 26.
linear (173.1(3)°), as expected, while the angle describing the
coordination of its terminal nitrogen to Ru(II) (164.4(2)°) is sig-
nificantly bent. In contrast, four crystal structures of Ru(III)-
cyanamide complexes showed an average Ru(III)-cyanamide
bond angle of 174.6°.21,22 We have suggested23 that the π
bonding between Ru(III) and cyanamide is optimized when this
angle is linear, a condition which evidently relaxed when π
bonding is not as important.
Cyclic voltammetry data for the thallium salt of apc-, the
tetraphenylarsonium salt of adpc2-, and the two complexes have
been placed in Table 3, and a figure showing the cyclic
voltammograms of the complexes has been placed in the
Supporting Information. The Ru(III/II) couple of [Ru(trpy)(bpy)-
(apc)]+ in acetonitrile is partially reversible and occurs at a
potential of 0.96 V vs NHE. Scanning to more positive potentials
resulted in decomposition of the complex and loss of the
cathodic wave for the Ru(III/II) couple. The voltammogram of
the dinuclear complex [{Ru(trpy)(bpy)}2(µ-adpc)]2+ in aceto-
nitrile shows two Ru(III/II) oxidation waves at 1.57 and 0.79
V vs NHE. The Ru(III/II) couples were quasi-reversible, with
the separation between cathodic and anodic waves dependent
upon the scan rate. On the basis of the difference between
Ru(III/II) couples, the comproportionation equilibrium constant
for [{Ru(trpy)(bpy)}2(µ-adpc)]3+ is very large at Kc ) 1.3 ×
1013 and implies strong metal-metal coupling in the mixed-
valence complex.
Spectroelectrochemical studies of the thallium salt of adpc2-
in DMF solution are shown in Figure 2. Upon one-electron
oxidation of adpc2-, the intense absorption of the intraligand
(24) (a) Calvert, J. M.; Schmehl, R. H.; Meyer, T. J. Inorg. Chem. 1983,
22, 2151. (b) Berger, R. M.; McMillin, D. R. Inorg. Chem. 1988, 27,
4245.
(25) Mosher, P.; Yap, G. P. A.; Crutchley, R. J. Inorg. Chem., in press.
(26) Rezvani, A. R.; Evans, C. E. B.; Crutchley, R. J. Inorg. Chem. 1995,
34, 4600.
(27) Antiferromagnetic exchange constants have been measured for [{NH3)5-
Ru}2(µ-L)]4+, where L is a 1,4-dicyanamidobenzene derivative, and
so antiferromagnetic exchange seems certain to make a considerable
contribution in [{Ru(trpy)(bpy)}2(µ-dicyd)]4+. See refs 21a and:
Naklicki, M. L.; White, C. A.; Plante, L. L.; Evans, C. E. B.; Crutchley,
R. J. Inorg. Chem. 1998, 37, 1880.
The bpy and trpy ligand reductions in similar complexes occur
between -1 and -2 V vs NHE, with the first reduction wave,
assigned24 to trpy (0/-1), showing some reversibility. In the
case of the dinuclear complex, reduction of the azo group of
(22) Crutchley, R. J.; McCaw, K.; Lee, F. L.; Gabe, E. J. Inorg. Chem.
1990, 29, 2576.
(23) Zhang, W.; Bensimon, C.; Crutchley, R. J. Inorg. Chem. 1993, 32,
5808.

A Class III Mixed-Valence Complex
Inorganic Chemistry, Vol. 40, No. 6, 2001 1193
Table 4. Electonic Spectral Dataa for apc- and adpc2- and Their
Ru(II) and Ru(III) Complexes and the Radical Anion adpc•-
compd
absorption
[apc]c
325b (4500), 490 (40 300)
340 (6700), 500 (54 400),
523 (51 500)
421 (18 400), 498 (31 100),
658 (5030), 714 (17 400),
1310 (3900)
293 (43 200), 318 (38 000),
469 (35 000)
289 (45 300), 311 (41 600),
448 (10 600), 1179 (3960)
293 (81 900), 317 (72 100),
490 (69 100)
292 (86 900), 314 (76 700),
471 (41 300), 652 (13 900),
812 (12 700), 1920 (10 000)
312 (73 400), 456 (41 900),
856 (12 000)
[adpc]2-
[adpc]-
[Ru(trpy)(bpy)(apc)]+
[Ru(trpy)(bpy)(apc)]2+
[{Ru(trpy)(bpy)}2(µ-adpc)]2+
[{Ru(trpy)(bpy)}2(µ-adpc)]3+
[{Ru(trpy)(bpy)}2(µ-adpc)]4+
a Conditions: in nm; molar extinction coefficient (L mol-1 cm-1
)
appears in parentheses; in 0.1 M TBAH in DMF. b Shoulder. c DMF
solution.
Figure 3. OTTLE cell vis-near-IR spectra of [{Ru(trpy)(bpy)}2(µ-
adpc)]2+ under increasing anodic potential in 0.1 M TBAH in DMF at
25 °C: (A) 0-0.75 V; (B) 0.80-1.05 V (versus Ag|AgCl).
LMCT chromophore, as the band at 448 nm is derived from
ILCT and possibly LMCT transitions.
For [{Ru(trpy)(bpy)}2(µ-adpc)]2+, oxidation to the mixed-
valence complex results in the appearance of three new spectral
features at λmax 652, 812, and 1920 nm (Figure 3A). The band
at 812 nm is assigned to the Ru(III)-cyanamide chromo-
phore, as it appears to gain absorptivity with the formation of
the [III,III] complex (Figure 3B). However, the band at 652
nm decreases in intensity with the formation of [III,III] (Figure
3B). We tentatively assign this band to a Ru(II)-adpc MLCT
transition, as ab initio studies of adpc2- (see Discussion) do
show a low-energy LUMO of π symmetry. The intense
absorption centered at 1920 nm (Figure 3A) is absent in the
spectrum of [II,II] and [III,III] complexes (parts A and B of
Figure 3) and is not present in the spectrum of adpc- (Figure
2). As the appearance of this band is most likely associated
with the oxidation of Ru(II), we have assigned this band to a
metal-to-metal charge-transfer MMCT transition.29 When the
Figure 2. OTTLE cell vis-near-IR spectra showing the oxidation of
adpc2- to the radical anion adpc-, in 0.1 M TBAH in DMF, at 25 °C.
charge-transfer band centered at 498 nm loses intensity and two
new bands of weaker intensity appear at 714 and 1310 nm that
we assign to π f π transitions of the radial anion adpc-. The
spectral changes that are seen in Figure 2 are reversible if
reduction is performed rapidly. However, further oxidation to
adpc0 is irreversible, as even partial regeneration of the adpc2-
spectrum was not achieved.
(29) This assignment depends on whether Ru(II) or adpc2- is oxidized in
the study shown in Figure 3A. We note that the energies, intensities,
and widths of the bands seen for the radical anion adpc- (Figure 2)
are not observed in the spectrum of the mixed-valence complex (Figure
3A). Second, in DMF solution, the potential of the first oxidation
couple of the dinuclear complex at 0.91 V vs NHE is close to that of
the mononuclear complex at 1.06 V vs NHE (Table 3) and of the
Ru(III/II) couples of [Ru(trpy)(bpy)L]+, where L is a phenylcyanamide
derivative.25 It is therefore reasonable to assign the couple at 0.91 V
to a Ru(III/II) couple. Finally, we have shown in studies of another
redox-active ligand that the redox potentials of the 1,4-dicyanamide-
benzene dianion shift to positive potentials by more than 1 V when
coordinated to pentaammineruthenium(III). This positive shift is even
greater when the ammines are replaced by pyridine moieties.26 We
expect a similar shift in ligand redox potentials when adpc2- is
coordinated to terpyridine(bipyridine)ruthenium(III), and this would
take the ligand couples outside the potential range that was studied.
The spectroelectrochemistry of the [II,II] complex in DMF
to generate [III,II] and [III,III] spectra is shown in Figure 3 while
the spectra resulting from the stepwise oxidation of [Ru(trpy)-
(bpy)(apc)]+ have been placed in the Supporting Information.
For [Ru(trpy)(bpy)(apc)]+, oxidation of Ru(II) is expected to
result in the loss of the MLCT transition in the visible and the
appearance of two new absorption bands due to the Ru(III)-
cyanamide LMCT chromophore.28 Only the band at 1179 nm
in the spectrum of [Ru(trpy)(bpy)(apc)]2+ (Table 4 and Sup-
porting Information) can be unambiguously assigned to this
(28) (a) Crutchley, R. J.; Naklicki, M. L. Inorg. Chem. 1989, 28, 1955. (b)
Evans, C. E. B.; Ducharme, D.; Naklicki, M. L.; Crutchley, R. J. Inorg.
Chem. 1995, 34, 1350.

1194 Inorganic Chemistry, Vol. 40, No. 6, 2001
Mosher et al.
Figure 4. Metal-metal charge-transfer band of [{Ru(trpy)(bpy)}2(µ-
adpc)]3+ in 0.1 M TBAH in DMF.
Table 5. Electrochemical and MMCT Band Spectral Data for
[{Ru(trpy)(bpy)}2(µ-adpc)]3+ in Various Solvents
solvent
acetonitrile
DMF
nitromethane
λmaxa/nm
1800 (6900)
[0.0935]
2710
1920 (10 000)
[0.167]
2870
1830 (8400)
[0.119]
2640
Figure 5. Calculated geometries for the highest occupied molecular
orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) of
adpc2- (A) and dicyd2- (B).35
∆ν/cm-1
ΗΜΜ′b/cm-1
∆Ec/V
2780
2600
2730
0.775
∼0.7
∼0.8
Kc
1.3 × 1013
1012-1013
1013-1014
observed. The reversibility from the [III,III] complex in ni-
tromethane or acetonitrile was noticeably poorer.
a Wavelength of maximum absorbance (in nm) with the extinction
coefficient (M-1 cm-1) in parentheses. The oscillator strength is shown
in brackets. HMM′ ) EMMCT/2. c The oxidation limits of DMF and
b
Discussion
nitromethane made precise determinations of the second Ru(III/II)
oxidation couples difficult.
The observation of class III properties in [{Ru(trpy)(bpy)}2-
(µ-adpc)]3+, despite a metal-metal ion through-space distance
of 19.5 Å, is extraordinary. The only comparable class III
complex is [{Ru(NH3)5}2(µ-1,4-benzoquinonediimine)]5+, in
which the ruthenium ions are separated by 10 Å.33 Indeed, the
magnitude of the comproportionation equilibrium constant of
[{Ru(trpy)(bpy)}2(µ-adpc)]3+ (Kc ) 1.3 × 1013) is second only
to that of [{Ru(NH3)4}2(µ-bptz)]5+, where bptz is 3,6-bis(2-
pyridyl)tetrazine, with Kc ) 1015 but at a metal-metal separation
of 6.8 Å.34 Why then is adpc2- so effective at mediating metal-
metal coupling?
MMCT band is plotted against an energy axis (Figure 4), the
band shape is non-Gaussian with a pronounced high-energy tail,
a characteristic of MMCT band properties of a class III com-
plex.30 This result together with the very large comproportion-
ation constant observed for [{Ru(trpy)(bpy)}2(µ-adpc)]3+ des-
ignates this complex as a class III system despite metal ions
that are separated by 19.5 Å.31 We note as well that for a class
III complex, the metal-metal coupling element HMM is given
by the MMCT band energy band energy:8,9
When metal ion separation precludes direct metal-metal
orbital overlap, coupling between metal ions necessarily involves
bridging ligand orbitals and is therefore defined by superex-
change processes. A complete description of superexchange
should include the sum of all possible contributions involving
the bridging ligand’s orbitals. However, it is usually possible
to simplify the description to a dominant pathway for super-
exchange, which in the case of a π-acceptor bridging ligand
(e.g., pyridine, pyrazine, etc.) is electron transfer through the
lowest unoccupied molecular orbital (LUMO),8 while for a
π-donating ligand (e.g., 1,4-dicyanamidobenzene dianion), it is
hole transfer through the highest occupied molecular orbital
(HOMO).7 Because of the cyanamide anion groups, the adpc2-
bridging ligand is a π-donor system and might be expected to
mediate metal-metal coupling by using the hole transfer
mechanism, as does the dicyd2- bridging ligand, but there are
important differences.
EMMCT
HMM′
)
(3)
2
Thus, for [{Ru(trpy)(bpy)}2(µ-adpc)]3+ in DMF, HMM′ ) 2600
cm-1. This is considerably smaller than what might have been
predicted on the basis of the magnitude of the comproportion-
ation constant but is not unprecedented in the literature.32
Class III complex properties should also be solvent-
independent, and so we examined the MMCT properties of
[{Ru(trpy)(bpy)}2(µ-adpc)]3+ in nitromethane and acetonitrile
as well as DMF (Table 5). Both the energy and bandwidth of
the MMCT band change very little with solvent. However, the
oscillator strength is significantly greater in DMF compared to
that in either acetonitrile or nitromethane.
The [III,III] spectrum in Figure 3B shows the growth of the
LMCT bands, as there are now two Ru(III)-cyanamide chro-
mophores per complex. Unfortunately, the [III,III] complex is
unstable, as only 90% of the [II,II] spectrum was recovered by
reducing [III,III]. The temperature of this experiment was
reduced to 0 °C, but no improvement in reversibility was
The results of an ab initio calculation35 of the HOMO and
LUMO of adpc2- and dicyd2- are shown in Figure 5. There
(33) Joss, S.; Reust, H.; Ludi, A. J. Am. Chem. Soc. 1981, 103, 981.
(34) (a) Johnson, J. E. B.; de Groff, C.; Ruminski, R. R. Inorg. Chim. Acta
1991, 187, 73. (b) Poppe, J.; Moscherosch, M.; Kaim, W. Inorg. Chem.
1993, 32, 2640.
(30) Nelsen, S. F. Chem. Eur. J. 2000, 6, 581.
(31) Obtained from the Ru(II) dinuclear crystal structure.
(32) See Table 15 of ref 8.
(35) Spartan version 5.0.1 calculation using the Hartree-Fock model with
the 6-31G** basis set and geometry optimization.

A Class III Mixed-Valence Complex
Inorganic Chemistry, Vol. 40, No. 6, 2001 1195
are two important comparisons to make. For dicyd2-, the
HOMO-LUMO energy gap is large at 2.59 eV and it is clear
that only the HOMO can interact simultaneously with ruthenium
donor and acceptor dπ orbitals. For adpc2-, both the HOMO
and LUMO have the correct symmetry to interact simultaneously
with ruthenium dπ orbitals and the HOMO-LUMO energy gap
is only 1.07 eV. Thus, superexchange via both the HOMO and
LUMO may be energetically favorable.
that of the electron-transfer pathway (∆ELM ) 12 200 cm-1),
and so it seems likely that hole-transfer superexchange makes
the greatest contribution to metal-metal coupling in [{Ru(trpy)-
(bpy)}2(µ-adpc)]3+
.
Conclusion
We have shown that by alternating donor and acceptor groups
within a bridging ligand it is possible to dramatically enhanced
metal-metal coupling within a mixed-valence complex. The
alternation of donor and acceptor groups reduces the energy
gap between the π HOMO and LUMO of adpc2- and, as these
orbitals can interact simultaneously with ruthenium ion dπ donor
and acceptor orbitals, superexchange via both hole and electron-
transfer mechanisms is possible. The magnitude of coupling seen
for [{Ru(trpy)(bpy)}2(µ-adpc)]3+ (HMM′ ) 2600 cm-1 and Kc
) 1.3 × 1013), despite a separation of 19.5 Å between ruthenium
ions, suggests that the properties of the adpc2- bridging ligand
approach that of a truly conducting molecular wire.
The general equation36 for superexchange coupling of the
metal centers
H
MLHM′L
HLMHLM′
HMM′
)
+
(4)
2∆EML
∆ELM
is illustrative of the various factors controlling the metal-metal
coupling element HMM′. The coupling elements in the two terms
on the right of eq 4 are associated with metal-ligand interactions
of the electron-transfer and hole-transfer pathways, respectively.
The denominators are reduced energy gaps between metal and
ligand orbitals. The subscript nomenclature may be understood
given that the electron-transfer pathway is associated with metal-
to-ligand charge transfer (MLCT) bands in the electronic
spectrum of the complex, while the hole-transfer pathway is
associated with ligand-to-metal (LMCT) bands.
In the present case, the reduced energy gaps can be readily
calculated from charge-transfer data. However, metal-ligand
coupling elements will greatly depend on the value chosen for
the transition dipole moment length. The largest value it may
have is 9.75 Å, which is the halfway point between the two
metal ions bridged by adpc2-. Depending upon the degree of
covalency in the metal-ligand interaction, the transition dipole
moment will be smaller and so a calculation of HMM may not
be useful at this time. Nevertheless, the reduced energy gap of
the hole-transfer pathway (∆ELM ) 9000 cm-1) is smaller than
Acknowledgment. We are grateful to the Natural Sciences
and Engineering Research Council of Canada (NSERC) for
financial support.
Supporting Information Available: Figures showing the cyclic
voltammograms of [{Ru(trpy)(bpy)}2(µ-adpc)]2+.and [Ru(trpy)(bpy)-
(apc)]+ in acetonitrile and absorption spectra of the spectroelectro-
chemical oxidation of [Ru(trpy)(bpy)(apc)]+ in DMF and full listings
of crystal structure data, atomic parameters, anisotropic thermal
parameters, bond lengths, and bond angles, as well as electronic files
in CIF format, for [{Ru(trpy)(bpy)}2(µ-adpc)]2+. This material is
available free of charge via the Internet at http://pubs.acs.org.
IC001268U
(36) Creutz, C.; Newton, M. D.; Sutin, N. J. Photochem. Photobiol. A:
Chem. 1994, 82, 47.