Linear metallopolymers from ruthenium(II)-2,6-di(quinolin-8-yl)pyridine complexes by electropolymerization - Formation of redox-stable and emissive films

Article abstract of DOI:10.1002/ejic.201300359

Two rodlike ruthenium(II) complexes of 2,6-di(quinolin-8-yl)pyridines (dqp) were synthesized and possess a pair of 2-thienyl moieties attached to the 4-positions of the quinoline units of one ligand or at the 4-positions of the pyridine rings of both ligands. The heteroleptic and homoleptic complexes were characterized by UV/Vis absorption and emission spectroscopy as well as electrochemical and X-ray crystallographic means. The subsequent electropolymerization of the thiophene units led to the controlled formation of thin solid films onto electrode surfaces as proven by X-ray photoelectron spectroscopy (XPS). The electrochemical analysis of the films was complemented by UV/Vis spectroscopy and UV/Vis/NIR spectroelectrochemistry, which revealed their stability towards oxidation, red emission, and reversible redox switching of their optical properties. Density functional theory (DFT) calculations were executed on the monomer complexes and respective dimeric systems to gain insight into their spectroscopic and electrochemical properties. Rodlike polymer assemblies containing thiophene-equipped ruthenium(II)-2,6-di(quinolin-8-yl) pyridine complexes are deposited onto electrode surfaces by applying an electropolymerization approach. The films display the defined and reversible electrochemical switchability of the ruthenium center as well as red luminescence under ambient conditions.

Full text of DOI:10.1002/ejic.201300359

FULL PAPER
DOI:10.1002/ejic.201300359
Linear Metallopolymers from Ruthenium(II)-2,6-
di(quinolin-8-yl)pyridine Complexes by
Electropolymerization – Formation of Redox-Stable and
Emissive Films
Christian Friebe,[a,b] Helmar Görls,[c] Michael Jäger,*[a,b] and
Ulrich S. Schubert*[a,b]
Keywords: Electropolymerization / Ruthenium / N ligands / Thin films / Density functional calculations
Two rodlike ruthenium(II) complexes of 2,6-di(quinolin-8-yl)-
pyridines (dqp) were synthesized and possess a pair of 2-
thienyl moieties attached to the 4-positions of the quinoline
units of one ligand or at the 4-positions of the pyridine rings
of both ligands. The heteroleptic and homoleptic complexes
were characterized by UV/Vis absorption and emission spec-
films onto electrode surfaces as proven by X-ray photoelec-
tron spectroscopy (XPS). The electrochemical analysis of the
films was complemented by UV/Vis spectroscopy and UV/
Vis/NIR spectroelectrochemistry, which revealed their sta-
bility towards oxidation, red emission, and reversible redox
switching of their optical properties. Density functional
troscopy as well as electrochemical and X-ray crystallo- theory (DFT) calculations were executed on the monomer
graphic means. The subsequent electropolymerization of the
thiophene units led to the controlled formation of thin solid
complexes and respective dimeric systems to gain insight
into their spectroscopic and electrochemical properties.
precludes efficient utilization of the excited states.[5] The ob-
served short lifetimes are caused by the efficient interaction
of the triplet metal-to-ligand charge-transfer (3MLCT) ex-
cited state with the triplet metal-centered (3MC) states and
their fast and radiationless deactivation towards the singlet
Introduction
Ruthenium(II) polypyridyl complexes are favorable
building units for photophysical applications, in particular
as sensitizers within light-harvesting devices, because of
their ability to undergo efficient and long-lived light-in-
2+
ground state within Ru(tpy)2 and analogous systems.[6]
2+
duced charge separation.[1] The prototypical Ru(bpy)3
The different strategies to overcome this limitation
(bpy = 2,2Ј-bipyridine) shows a long excited-state lifetime
of 860 ns, which allows subsequent redox reactions to oc-
cur.[2] However, the tris(bidentate) ligand assembly can lead
to the formation of enantiomeric and diastereomeric mix-
tures and impedes the ideal trans arrangement of donor and
acceptor moieties with maximal separation to prevent back-
reactions.[3] On the other hand, the bis(tridentate) congener
3
mainly focus on the enlargement of the MLCT–3MC en-
3
ergy gap by stabilization of the MLCT state, destabiliza-
3
tion of the MC state, or both. The most promising ap-
proaches are based on the incorporation of strong electron
donors by cyclometalation[6a,7] or N-heterocyclic carbenes[8]
and structural modifications,[9] whereas the attachment of
additional energy-storing chromophores only enhances the
apparent lifetime.[10] Hammerström et al. demonstrated that
the utilization of tridentate ligands with a bite angle of 180°
leads to a less distorted octahedral complex geometry,
which causes a decreased admixing of metal d orbitals into
the lowest unoccupied molecular orbital (LUMO). Thus,
the orbital overlap between the ligand-centered LUMOs
and the eg orbitals is diminished and a deactivation via the
3MC state becomes less probable.[6b] Applying this concept,
the family of substituted Ru(dqp)22+-based complexes [dqp
= 2,6-di(quinolin-8-yl)pyridine] combines excellent proper-
ties for light-harvesting applications with an excited-state
lifetime in the μs time scale.[11]
2+
Ru(tpy)2 (tpy = 2,2Ј:6Ј,2ЈЈ-terpyridine) allows for the
stringent trans alignment of the substituents,[4] but pos-
sesses a very short excited-state lifetime of 0.25 ns, which
[a] Laboratory of Organic and Macromolecular Chemistry
(IOMC), Friedrich Schiller University Jena,
Humboldtstr. 10, 07743 Jena, Germany
E-mail: michael.jager.iomc@uni-jena.de
ulrich.schubert@uni-jena.de
Homepage: http://www.schubert-group.com/
[b] Jena Center for Soft Matter (JCSM), Friedrich Schiller
University Jena,
Philosophenweg 7, 07743 Jena, Germany
[c] Laboratory of Inorganic and Analytic Chemistry (IAAC),
Friedrich Schiller University Jena,
The subsequent incorporation into photovoltaic devices
requires thin-film processing of the respective sensitizers,
that is, the deposition of a layer that is thin and homogen-
Lessingstr. 8, 07743 Jena, Germany
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201300359.
Eur. J. Inorg. Chem. 2013, 4191–4202
4191
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
2+
Scheme 1. Type II and type III arrangement of metal-containing polymers as applied for Ru(dqp)2 complexes in this contribution.
eous enough to allow efficient charge transport onto an
The structural features of the complexes closely resemble
2+
electrode surface. For this purpose, a convenient and wide- those previously obtained for the parent Ru(dqp)2 with a
spread technique is the electrochemical polymerization of different basis set (see Supporting Information).[11b] No-
a metal-containing monomer solution to enable the direct tably, both substitution patterns differ in the degree of co-
preparation of an insoluble polymeric coating on the sur- planarity (dihedral angle) between the dqp and the outer
face[12] and, thus, numerous ruthenium(II) polypyridyl-type phenyl ring, which suggests an increased interaction with
complexes have been processed by electropolymerization.[13] the dqp-based molecular orbitals (MOs) from 1 (49°) to 2
2+
In this contribution, we present two Ru(dqp)2 complexes (30°). However, no significant differences between the
featuring 2-thienyl units linked either to the quinoline (1) phenyl and thiophene twists were observed (a comparison
or the pyridine moieties (2) and their subsequent incorpora- to the experimental data is provided below). The related
tion into a photoactive film by electropolymerization dimeric complexes display very similar structural param-
(Scheme 1). The two investigated substitution patterns en- eters and have identical Ru–N bond lengths and torsional
able a lateral attachment to the polymer main chain or a angles. The twist between the phenyl and thiophene unit
direct incorporation into the backbone (metal-containing is slightly smaller, whereas the torsion angle between the
polymers of type II and III, respectively, Scheme 1)[14] with bridging thiophene units is 10°. On the basis of the dihedral
equal spatial separation of the ruthenium(II) centers. First angles, phenyl–bithiophene–phenyl conjugation within the
density functional theory (DFT) and time-dependent DFT bridge is reasonable (Ͻ22°) and also allows coupling to the
2+
(TD-DFT) calculations were performed to explore the Ru(dqp)2 fragments (Ͻ49°).
structural, spectroscopic, and electrochemical properties of
The frontier MO energies and respective contributions of
the designed systems. The second part describes the synthe-
the involved fragments of the complexes in the ground state
sis and characterization of the monomer complexes as well
are depicted in the Supporting Information. Importantly,
as the resulting polymer films by UV/Vis spectroscopy and
the highest occupied molecular orbitals (HOMO and
cyclic voltammetry. Additionally, X-ray crystallographic
HOMO–1) are mainly metal-based, whereas the nearby or-
analysis for the monomers and UV/Vis/NIR spectroelectro-
bitals (HOMO–2 and HOMO–3) show significant admixing
chemical experiments and X-ray photoelectron spec-
of the thiophene units. The lowest unoccupied molecular
troscopy (XPS) studies of the obtained polymer films were
orbitals (LUMO to LUMO+5) are primarily localized on
the dqp ligands. For the asymmetric complex 1, the contri-
executed.
bution of the thiophene-substituted dqp fragment to the
LUMO and LUMO+1 is larger than that of nonfunction-
alized dqp, whereas the opposite holds for the LUMO+2
and LUMO+3. For the symmetric complex 2, the contri-
Results and Discussion
bution of each dqp ligand to the LUMOs is equal because
DFT Calculations
of the symmetry of the molecule. Upon dimerization, the
DFT and TD-DFT calculations were performed to inves- orbital energy of the bithiophene fragments is significantly
tigate the electronic structures and transitions of the two lowered and constitutes the new HOMO with only minor
different conjugation paths. Firstly, the geometries of the admixing of d orbitals from both ruthenium centers. How-
complexes were optimized with the given charge and multi- ever, the lower occupied orbitals remain metal-based. The
plicity and subsequently confirmed by vibrational analysis. LUMOs of the dimers are ligand-based and display a
The optical properties were further investigated by TD- higher contribution of the dqp fragment connected to the
DFT to obtain absorption spectra. In addition to the bridge. The electronic structure of the complexes can be
monomers, the respective dimetallic complexes were investi- qualitatively summarized as follows: (1) The HOMO mani-
gated to explore the effect of dimerization on the electronic fold is essentially ruthenium-based, (2) the dimers show a
structure (for optimized geometries, see Supporting Infor- delocalized HOMO on the bridge with minor contributions
mation).
of both metal centers, (3) the LUMO manifold is ligand-
Eur. J. Inorg. Chem. 2013, 4191–4202
4192
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Figure 1. EDDM plots of the lowest-energy ground state absorptions (isovalue = 0.0016, decrease in blue, increase in cyan) of 1 (top
left) and 2 (top right). Electron density from spin-density difference calculations of dimer(1) (middle) and dimer(2) (bottom, isovalue =
0.001) in the singly oxidized state.
based throughout the series and shows a larger contribution ficient conjugation through the metal-based d orbitals in
from the substituted dqp ligand. However, the computa- dimer(2), as described above.
tional results of the HOMOs of the dimers should be inter-
preted with care owing to the well-known artificial stabili- upon oxidation were examined by spin-density difference
zation of delocalized states.[15]
plots (see Supporting Information). Upon oxidation of the
The changes in the electronic structures of the complexes
The vertical excitations were investigated by TD-DFT monomeric complexes, the first electron is removed from a
calculations (see Supporting Information). Electron-density metal-centered orbital, whereas the second one originates
difference maps (EDDM) are often used to visualize elec- from one 4-(2-thienyl)phenyl unit. Hence, the intended re-
tronic redistribution,[16] in particular if the discussion is dox-mediated coupling becomes reasonable upon a second
complicated by many contributing MOs. The monomers 1 oxidation, according to the accepted mechanism involving
and 2 display low-energy transitions with MLCT character an oxidized thiophene radical.[17] In contrast, both dimers
(Figure 1), whereas the higher-energy transitions revealed show first oxidations that reside primarily on the bridge
an extended delocalization across the aromatic units and an with minor contributions from both ruthenium centers. The
increased admixing of thiophene-based orbitals. Further- second and third oxidation processes are localized on the
more, the EDDM analysis shows the principle effect of the individual ruthenium centers (see Supporting Information).
substitution pattern. The alignment of the thiophene- and
Notably, the order of the successive oxidation steps may
metal-based d orbitals leads to higher oscillator strengths be reversed, owing to the discussed artificial stabilization of
of the respective transitions. For complex 1, this combina- delocalized states.[15] However, the observed low energetic
tion is less favorable and the strong MLCT transition is differences between the metal- and bridge-based orbitals as
found at 505 nm, whereas this transition in complex 2 well as their spatial overlap suggest significant interactions
shows a pronounced redshift to 536 nm, in line with the between them.
lower dihedral angle between the dqp fragment and the
phenyl ring (see above).
The combined computational results of the monomeric
and dimeric structures show their promising potential in
The dimers display a similar behavior, except for an ad- electropolymerized films. As the polymerization process is
ditional very intense intraligand charge-transfer (ILCT) expected to start upon the oxidation of the thiophene unit,
transition of the bis(thiophene). For dimer(1), the asymmet- the (easier) ruthenium-centered oxidation may serve as a
ric substitution pattern induces a localization on the quinol- valuable tool to monitor the course of the electropolymeri-
ine unit connected to the bridge, whereas the accepting zation. The MO analysis of the dimeric structures reveals
LUMO of dimer(2) is almost evenly distributed over both an extended delocalization across the bridge, that is, a weak
dqp ligands. This behavior is consistent with the more ef- communication between the two metal centers in the oxid-
Eur. J. Inorg. Chem. 2013, 4191–4202
4193
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
ized state, which is beneficial for efficient charge migration fits from the direct access to the bis-functionalized ligand.
within the films. However, a more detailed analysis, for ex- The related pyridine-substituted ligand 8,8Ј-[4-(4-bromo-
ample, modeling of polymeric structures, is beyond the phenyl)pyridine-2,6-diyl]diquinoline was synthesized by a
scope of this study.
Kröhnke reaction of acetylquinoline and p-bromobenzalde-
hyde as described in the literature.[21] The next step involved
the coordination to a suitable ruthenium precursor and sep-
aration from facial isomers[11a] by fractionalized crystalli-
Syntheses of Monomers
The synthetic strategy towards the electropolymerizable zation. Heating ligand 5 with [Ru(dqp)(CH3CN)3](PF6)2 af-
thiophene-equipped complexes is shown in Scheme 2 and is forded complex 6 in excellent yield (95%). The subsequent
based on the synthesis of the bromo-functionalized ligands, conversion to the thiophene-functionalized complex (1) was
followed by the stepwise coordination to ruthenium, and quantitative without any detectable debromination accord-
the introduction of the thiophene moieties in the final step. ing to the 1H NMR signal intensity ratios of the thiophene-
This sequence was chosen to prevent any side reactions of and the quinoline-related protons. The related homoleptic
the thiophene units owing to the harsh conditions during complex 2 was synthesized in a similar stepwise fashion; the
the coordination steps and to enable easy removal of the initial reaction of RuCl3 with 8,8Ј-[4-(4-bromophenyl)-
small amounts of the inevitably formed facial isomers by pyridine-2,6-diyl]diquinoline was followed by a reduction
crystallization. In addition, this route also explores the ver- and halide abstraction with AgI to yield the intermediate
satility of the intermediate bromo complex, for example, for complex 7. The next step is the coordination of a second
subsequent cross-coupling reactions. To construct the equivalent of 8,8Ј-[4-(4-bromophenyl)pyridine-2,6-diyl]di-
framework of the quinoline-functionalized ligand, the origi- quinoline; however, some debromination occurred and iso-
nal route by C–C coupling was adjusted to tolerate the reac- lation of the bis(bromo) complex by chromatography and
tive peripheral bromo substituents. In this regard, 2-nitro- crystallization failed. We tentatively assign this failure to
benzoic acid represents a valuable quinoline precursor, as the small structural difference between the H and Br sub-
shown by the efficient decarboxylative cross coupling with stituents among the complexes. Hence, the crude material
a variety of aryl halides.[18] To suppress the protonation of was converted into the thiophene-equipped complexes,
the formal C nucleophile after decarboxylation, the potas- which were successfully separated by crystallization. The
sium salt was used instead.[19] The twofold coupling with origin of the unexpected debromination is unknown and
2,6-dibromopyridine gave 3 (48% yield), which was sub- the isolated yields of 2 are lower (38% over both steps).
1
sequently reduced to the corresponding bis(aniline) 4 by
The H NMR spectra of 1 and 2 show the characteristic
using hydrazine hydrate and palladium on charcoal. Finally, pattern of the coordinated ligands in the aromatic region
the quinoline ring formation was achieved by a twofold (Figure 2) and are supported by 2D NMR spectroscopic
Skraup reaction with a commercial bromophenyl-substi- data (see Supporting Information). The spectrum of the
tuted C3 synthon.[20] Although this route gave only a low asymmetric complex 1 has two overlapping sets of signals
yield (15%) of bis(bromophenyl)-substituted dqp (5), it is for each ligand. However, the subunits can be identified ac-
comparable with related single Skraup reactions and bene- cording to their characteristic chemical shifts and coupling
Scheme 2. Schematic representation of the synthetic route towards the heteroleptic complex 1 (top) and the homoleptic complex 2
(bottom): (i) CuI, Pd/C, 1,10-phenanthroline, N-methyl-2-pyrrolidinone (NMP), 180 °C, 15 h, 48%; (ii) Pd/C, hydrazine hydrate, EtOH,
80 °C, 1 h, 86%; (iii) 1-(4-bromophenyl)-3-chloropropan-1-one, arsenic pentoxide, phosphoric acid, 140 °C, 2 h, 15%;
(iv) [Ru(dqp)(CH3CN)3](PF6)2, ethylene glycol, 140 °C, 14 h, 95%; (v) thiophen-2-ylboronic acid, Pd(dba)2, S-PHOS, CH3CN/H2O, potas-
sium carbonate, 100 °C, 16 h, 92 (1) and 55% (2); (vi) 1. RuCl3·3H2O, EtOH, 115 °C, 13 h; 2. AgNO3, CH3CN/EtOH/H2O, 90 °C, 17 h,
then NH4PF6, 60%; (vii) 8,8Ј-[4-(4-bromophenyl)pyridine-2,6-diyl]diquinoline, ethylene glycol, 140 °C, 14 h, 68%.
Eur. J. Inorg. Chem. 2013, 4191–4202
4194
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
patterns, in agreement with the literature.[11a,21] The proton plexes within the unit cell,[11b] which is seen in the large
in the 3-position of the quinoline unit is noticeable as a absolute difference of the respective bond lengths. However,
doublet in ligand 5 or as a double doublet in unsubstituted the average value agrees well with the numbers derived from
dqp. The protons in the 4-position of the pyridine rings complexes 6 and 2. In general, the Ru–N bond length of
appear as overlapping triplets owing to a small chemical the central pyridine ring is shortened by 0.026 Å compared
shift difference. Two of the thiophene protons are well re- with that of the outer quinoline unit. A similar behavior
solved and can be used to validate the thiophene content. was found for the calculated structures, despite the known
A direct comparison by means of the adjacent phenyl ring typical overestimation of the calculated bond lengths
is complicated owing to superposition with the remaining (+0.04 Å). The internal N–Ru–N angles are all close to the
signals. Complex 2 shows a less complicated spectrum ow- ideal octahedral coordination, as are the dihedral angles be-
ing to the axial symmetry of the molecule. Similar features tween the quinoline and pyridine unit as calculated from
of the quinoline and thiophene units are present, whereas their respective mean planes.
the 3- and 5-protons of the pyridine rings appear as singlets
in the spectrum.
1
Figure 2. H NMR spectra (CD3CN, 300 MHz, expanded region)
of 1 (top) and 2 (bottom) with assignment of the characteristic
protons.
X-ray Crystallography
Figure 3. Top: Solid-state structure of 6 (ellipsoids drawn at 50%
probability, hydrogen atoms are omitted for clarity). Bottom: Solid-
The structures of the complexes with both substitution
patterns were also investigated by X-ray crystal analysis
atoms are omitted and only one conformation of the thiophene
state structure of 2 (ellipsoids drawn at 50% probability, hydrogen
(Figure 3 and Table 1). All attempts to crystallize complex
1 gave only plates, and the quality of the collected X-ray
data was not sufficient to allow discussion of the structure
units is displayed for clarity, spherical carbon atoms are isotropic).
The numbers agree well with those of the computed
beyond the configuration and conformation (see Support- structures (37°); somewhat larger variations were found for
ing Information). Hence, the related bromo precursor 6 was 2. This observation is tentatively attributed to crystal pack-
also investigated to supplement the structural discussion ing effects and indicates a certain flexibility of the coordi-
with respect to 2. The geometrical features of the coordina- nated ligand. This distortion is also visible from the “bend-
tion site, that is, the mutual arrangement of the N and Ru ing” away from the C2 axis of the molecule in 2 (Figure 3).
atoms, are preserved in comparison to the parent Ru- In addition, two conformations of the thiophene unit with
(dqp)22+. Reinvestigation of the available X-ray data of respect to the neighboring phenyl group were found in the
2+
Ru(dqp)2 showed a large deviation between the two com- solid state and arise from the very similar steric demand
Table 1. Solid-state structural data for Ru(dqp)22+, 6, and 2.
Compound
Ru–N[a] [Å]
N–Ru–N [°]
Adjacent
Torsion angle[c] [°]
py–quin py–ph ph–tph
dqp–thiophene[b]
dqp
Npy
Axial
Npy
Nquin
Nquin
2+[d]
Ru(dqp)2
6
2
–
–
2.034 (0.030)
2.039 (–)
–
2.063 (0.050)
2.070 (0.005)
–
88–92
89–90
89–91
176–180
179–180
179–180
35–40
37–41
29–44
–
–
2.036 (–)
2.028 (–)
2.073 (0.002)
2.064 (0.007)
46–47 4–33[e]
26 13
[a] Numbers in parentheses are the absolute differences between the experimental minimum and maximum values. [b] Thiophene-contain-
ing ligand {i.e., 3 or 8,8Ј-[4-(4-bromophenyl)pyridine-2,6-diyl]diquinoline}. [c] Calculated angles spanned by the mean planes defined by
all heteroatoms of the aromatic unit (py = pyridine, quin = quinoline, ph = phenyl, tph = thiophene). [d] Calculated from the crystallo-
graphic data (average of both complexes of the unit cell) from ref.[11b] [e] Data taken from the structural motif of 1.
Eur. J. Inorg. Chem. 2013, 4191–4202
4195
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
2+
after a 180° rotation along the connecting C–C bond. The and agree with the reported data of the parent Ru(dqp)2
structural motif of 1 displays a similar torsion angle be- complex.[11b]
tween the pyridine and phenyl rings (45°) as that in 6, but
Cyclic voltammetric (CV) measurements (Table 2 and
a large variation for the phenyl and thiophene unit (4 and Supporting Information) showed reversible first oxidations
33°). The average experimental values agree very well with at 0.70 V vs. Fc+/Fc for 1 and 0.67 V for 2, which are attrib-
the DFT calculations; however, the larger numerical varia- uted to single-electron RuIII/RuII redox processes. A second,
tions, for example, between the pyridine and quinoline units irreversible oxidation wave was observed at ca. 1.3 V for
and the peripheral aromatic units, may be induced by the both complexes and is assigned to the formation of thio-
crystal packing (see Supporting Information).
phene radical moieties. Furthermore, the first reduction sig-
nals at ca. –1.70 V featured reversible behavior for both
complexes.
Photophysical and Electrochemical Characterization of
Monomers
The optical changes of 1 and 2 upon oxidation were in-
vestigated by UV/Vis/NIR spectroelectrochemistry. Firstly,
a decrease of the MLCT absorption band at 500 nm was
observed as well as the formation of a broad absorption
band between 600 and 1100 nm featuring several unas-
signed low-energy transitions from lower-lying occupied or-
bitals to the metal-based singly occupied molecular orbital
(SOMO, see the Supporting Information). These results are
in line with those for the parent Ru(dqp)22+ and are further
discussed with regard to the electropolymerized films (see
below).
The monomeric complexes were further characterized by
UV/Vis spectroscopy (Figure 4 and Table 2). The absorp-
tion spectrum of the heteroleptic complex 1 shows charac-
teristic bands between 400 and 600 nm, which are assigned
to MLCT transitions. Complex 2 exhibits similar absorp-
tion bands, but displays a pronounced shoulder at longer
wavelength and an additional peak at 425 nm. The experi-
mental absorption data are in very good agreement with
the DFT results (see also the electronic transitions in the
DFT Calculations section). The UV/Vis emission measure-
ments revealed structureless room-temperature emission
bands at 698 (1) and 678 nm (2) with Stokes shifts of 4900
and 3300 cm–1, respectively, which indicate significant dif-
ferences between the two conjugation pathways in the emis-
sive excited states. Furthermore, photoluminescence quan-
tum yields (ΦPL) of 0.8 and 3.1% for 1 and 2, respectively,
could be determined in deaerated solutions at room tem-
perature. The excitation spectra support the assignments
Electropolymerization
Complex 1 was electropolymerized in acetonitrile con-
taining 5 vol.-% BF3·OEt2 and 0.1 m Bu4NPF6 as conduc-
tive electrolyte, according to the proposed reaction in
Scheme 1. The electropolymerization was not possible in
pure acetonitrile/Bu4NPF6. Hence, the Lewis acid boron
trifluoride diethyl etherate was used as it interacts with the
aromatic system of the thiophene and reduces its aromatic-
ity and, thus, leads to a lowered oxidation potential for the
thiophene moieties and enables electropolymerization.[22]
The polymerization was conducted potentiodynamically by
cycling between –0.5 and 1.7 V vs. Fc+/Fc; the thiophene
moieties are oxidized at ca. 1.2 V to form reactive thienyl
cation radicals.[17] Figure 5 shows the development of the
cyclic voltammogram over the first 50 cycles: It exhibits a
well-defined growth of the characteristic electrochemical re-
sponse at 0.7 V, and the signal corresponding to oxidation
of the mono-thiophene decreases over the first cycles owing
to the consumption of monomeric complexes near the elec-
trode surface. Notably, a small cathodic peak at ca. 0.9 V
occurs after two cycles and is tentatively assigned to the re-
reduction of oxidized thiophene moieties, which may origi-
nate from trapped units within the (formed) film or from
Figure 4. UV/Vis absorption (hollow symbols) and emission (filled
symbols) spectra of the monomeric complexes 1 (blue) and 2 (red)
(10–6 m in CH3CN).
Table 2. UV/Vis spectroscopic properties and electrochemical data of the monomer complexes (in CH3CN, 10–6 m for UV/Vis spectroscopy,
10–4 m with 0.1 m Bu4NPF6 for electrochemistry).
λAbs [nm] (ε [103 m–1 cm–1])[a]
λEm [nm] ΦPL [%]
E1/2 [V] (ipa/ipc, ΔEp [mV])
[b]
[c]
2+ Ǟ3+
2+ Ǟ1+
1
2
521s (16.6), 500 (18.5), 350 (43.9), 285 (46.4)
553s (15.0), 507 (17.1), 425 (14.7), 345 (64.7), 290 (41.7) 678
698
0.8
3.1
0.70 (1.05, 70)
0.67 (1.06, 68)
–1.68 (0.94, 79)
–1.70 (0.90, 73)
[a] s = shoulder. [b] Measured by using [Ru(dqp)2](PF6)2 (ΦPL = 2% in MeOH/EtOH, 1:4) as reference.[11b] [c] Measured vs. Fc+/Fc.
Eur. J. Inorg. Chem. 2013, 4191–4202
4196
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
unreacted units owing to a decrease in the rate of oxidative thenium, carbon, nitrogen, and sulfur) as well as Ru/S ra-
dimerization as the monomer is depleted. The increase of tios of 1:2.1 and 1:1.9 for poly(1) and poly(2), respectively,
the peak current corresponding to the ruthenium(II)- and/ consistent with the theoretical value.
or bis(thiophene) oxidation shows a slope change around
The obtained CV data prove a successful electrochemical
the fifteenth cycle, most likely because of the complete polymerization process and show a continuous growth of
coverage of the electrode surface with polymer, and, thus, the peak current of the redox wave of the ruthenium(II)
a decreased charge transport, which causes a diminished complex. This is attributed to deposition of the complex
polymerization rate.[23]
moieties and is accompanied by the disappearance of the
thienyl-related signal, which is assigned to irreversible con-
secutive reactions of the formed thienyl radical cations,
namely coupling reactions to generate oligomeric and poly-
meric chains. The utilization of boron trifluoride diethyl
etherate, which is known to enhance the oxidative electro-
polymerization ability of aromatics by lowering of the redox
potential, was necessary to enable the polymerization pro-
cess.
Electrochemical and Photophysical Characterization of the
Polymers
The films were rinsed with pure solvent to remove solu-
ble monomer species after electropolymerization, and the
coated working electrodes were immersed in fresh solvent
with 0.1 m Bu4NPF6 and showed no dissolution at all. The
electrochemical and photophysical data are summarized in
Table 3. Figure 6 shows cyclic voltammograms of the oxi-
dation of the polymers at different scan rates. The half-wave
potential of poly(1) is only marginally shifted (towards
0.76 V vs. Fc+/Fc) compared to that of the dissolved mono-
mer complex. However, the involvement of the bis(thio-
phene) unit in the oxidation process cannot be excluded as
the respective potential of the bis(phenylthienyl) (1.14 V vs.
SCE, ca. 0.73 V vs. Fc+/Fc)[24] is close to the observed redox
potential. Furthermore, the redox process is reversible: The
charge density is the same for oxidation and reduction (ca.
5ϫ10–4 Ccm–2). The linear relationship between peak cur-
rent and applied scan rate up to 500 mVs–1 indicates the
formation of a conductive film with redox processes that
are only weakly limited by charge diffusion.[17,25] For
poly(2), the oxidation signal appears well-defined and re-
versible (charge transfer of around 3ϫ10–4 Ccm–2 for both
oxidation and subsequent reduction) at a potential of
0.72 V vs. Fc+/Fc. In contrast to poly(1), the linearity of
the peak-current–scan-rate function is retained up to the
maximum applied scan rate of 2000 mVs–1 (Figure 6). This
behavior can be explained by a higher charge mobility than
in poly(1), which is further supported by the smaller Epa–
Epc separation as well as the already mentioned smaller de-
Figure 5. First 50 CV cycles of the electropolymerization of 1 (top)
and 2 (bottom) on a glassy carbon disk electrode. Insets: Peak cur-
rent increase at 0.70 V with cycle number (10–4 m in CH3CN with
5 vol.-% BF3·OEt2 and 0.1 m Bu4NPF6).
Likewise, electropolymerization of homoleptic 2 was per-
formed and monitored by recording the respective cyclic
voltammograms (Figure 5). As for the heteroleptic counter-
part, a decrease of the current slope is observable at around
the fifteenth cycle, but is less pronounced than for 1. How-
ever, further studies are required to reveal the effect of other
factors, such as counterion diffusion and charge mobility,
on the film growth.
Table 3. UV/Vis spectroscopic and electrochemical data of electro-
polymerized films.
[a]
[a]
[b]
λAbs, poly λAbs, mono
λEm, poly
λEm, mono
E1/2,ox
[nm]
[nm]
[nm]
[nm]
[V]
The elemental composition of the deposited films on in-
dium tin oxide (ITO) coated glass substrates was investi-
gated by X-ray photoelectron spectroscopy (XPS, see Sup-
porting Information). Analysis of the spectra revealed the
Poly(1) 537
Poly(2) 567
527
551
767
745
746
717
0.76
0.72
[a] UV/Vis properties of spin-coated films of the monomer com-
signals for the expected characteristic elements (namely, ru- plexes. [b] Measured vs. Fc+/Fc.
Eur. J. Inorg. Chem. 2013, 4191–4202
4197
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
crease of the current slope during electropolymerization for spectral shifts towards higher wavelengths were observed
poly(2).
likewise for ruthenium(II) polypyridyl systems in previous
studies and are assigned to the presence of low-energy trap
sites, which are available through electronic interaction be-
tween the ligand π systems of the closely packed complexes
in the solid state.[26] For the polymerized systems, an even
more efficient interaction is plausible and leads to the more
pronounced redshift. However, no significant effect of the
conjugation path on the excited-state properties was found.
Figure 6. Cyclic voltammograms of the electropolymerized film of
poly(1) (top) and poly(2) (bottom) at different scan rates. Insets:
Peak-current dependence on scan rate [coated glassy carbon elec-
trode in CH3CN with 0.1 m Bu4NPF6; linear fit for poly(1) is valid
up to 500 mVs–1].
The UV/Vis absorption and emission characteristics of
films of poly(1) and poly(2) on ITO-coated glass substrates
were determined. The films of poly(1) and its respective
monomer 1 (Figure 7) show a significant redshift of absorp-
tion of about 1400 and 1000 cm–1, respectively, in compari-
son to that of the monomer dissolved in acetonitrile, but
there are only marginal differences between the monomer
and the polymer in the solid state. Similarly, the thin-film
absorption of poly(2) exhibits a large redshift between the
Figure 7. UV/Vis absorption (hollow symbols) and emission (filled
symbols) spectra of films of poly(1) (green), 1 (blue) (top), poly(2)
(orange), and 2 (red) (bottom) on ITO-coated glass substrates.
The electrosynthesized polymer films were studied by
dissolved monomer complex 2 and the spin-coated film UV/Vis/NIR spectroelectrochemistry [Figure 8 shows an
thereof, and only a small shift of 500 cm–1 occurs for the exemplary spectra for poly(1), and those of the other com-
polymer relative to the monomer film (Figure 7). Further- pounds are in the Supporting Information]. Poly(1) showed
more, the additional peak at 425 nm, which was observed a bleaching of the low-energy MLCT absorption band,
for 2 in solution, is also present for the films as an absorp- caused by the depletion of the respective metal-located or-
tion shoulder at ca. 430 nm. Notably, the absorbance of the bitals, as well as the appearance of a broad band at 600–
formed bis(phenylthienyl) moieties is expected at ca. 1100 nm, similar to that of the related monomer species.
374 nm[24] and is, thus, overlaid with features of the rutheni- This leads to a color change of the polymer film from deep
um(II) complexes. Both polymeric films showed weak pho- red to light yellow. However, no clear evidence of the oxid-
toluminescence. In comparison to the emission of the com- ized bis(thiophene) moieties can be deduced from the data
plexes in solution, the solid-state emission of the spin- owing to the spectral overlap [the bis(phenylthienyl) radical
coated monomers is bathochromically shifted by approxi- cation absorbs at ca. 932 nm],[24] despite the occurrence of
2+
mately 800 to 900 cm–1, and the emission of the electropoly- some distinct bands in comparison to Ru(dqp)2 and the
merized films is shifted by ca. 1300 cm–1 (see Figure 7). The respective monomers. Subsequent application of a potential
Eur. J. Inorg. Chem. 2013, 4191–4202
4198
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
properties. A small difference between both conjugation
paths was observed, namely a faster electrochemical re-
sponse, which is attributed to a higher charge-carrier mo-
bility for the type III polymer (obtained from the homolep-
tic monomer complex); this finding is in accordance with
the higher film growth rate and DFT calculations, which
revealed a more efficient π conjugation within the type III
system.
However, additional parameters (e.g., counterion mo-
bility, film morphology, and thickness) may also contribute
and are under investigation. The presented approach to
photoredox-active films benefits from facile instrumen-
tation, the modular design of the monomers, and the pres-
ervation of photophysical and electrochemical properties.
Figure 8. Exemplary change of the UV/Vis absorption spectrum of
a film of poly(1) during the oxidation process. Inset: Change of Experimental Section
transmission at 515 nm over 20 cycles of switching between initial
and oxidized state.
General Methods: All starting materials were purchased from com-
mercial sources [dba is dibenzylideneacetone, S-PHOS is dicyclo-
hexyl(2Ј,6Ј-dimethoxybiphenyl-2-yl)phosphane, Pd/C is activated
palladium on charcoal (10 wt.-% from Aldrich)] and were used as
obtained unless otherwise noted; potassium 2-nitrobenzoate,[19]
able to reduce the oxidized species recovered fully the start-
ing spectrum; the repeatable change of film transmission
at 515 and 810 nm with switching potential over 20 cycles
indicates a reversible and stable redox process with switch-
ing times (defined by achieving 95% of the full transmission
change[27]) of approximately 2 s.
Likewise, the application of an oxidative potential to
poly(2) films caused the disappearance of the bands be-
tween 400 and 600 nm and the increase of absorption inten-
8,8Ј-[4-(4-bromophenyl)pyridine-2,6-diyl]diquinoline,[21]
and
[11a,21]
[Ru(dqp)(MeCN)3](PF6)2
were prepared according to litera-
ture procedures. Flash column chromatography was conducted
with a Biotage Isolera One System with Biotage SNAP Cartridges
KP-Sil and a UV/Vis detector. Microwave reactions were per-
formed with a Biotage Initiator Sixty Microwave synthesizer.
UV/Vis absorption spectra of solutions and films were recorded
sity in the NIR region with a peak at 800 nm. The re-re- with a Perkin–Elmer Lambda 750 UV/Vis spectrophotometer,
emission spectra of solutions were recorded with a Jasco FP6500,
and emission spectra of films were recorded with a Tecan infinite
M200 Pro microplate reader. Solution measurements were per-
formed by using concentrations of 10–6 m in respective solvents
(spectroscopy grade; deaerated for emission measurements) in 1 cm
quartz cuvettes at 25 °C; emission spectra were taken by excitation
at the longest-wavelength absorption maximum.
duction of poly(2) produced the initial spectrum nearly
completely and, at least over 20 cycles, a reversible and
stable redox switching with response times of ca. 2.5 s could
be observed.
Conclusions
Electrochemical measurements were performed with a Metrohm
Autolab PGSTAT30 potentiostat with a standard three-electrode
configuration (a glassy carbon disk working electrode, a platinum-
Two new ruthenium(II) complexes based on 2,6-di(quin-
olin-8-yl)pyridines were synthesized by a straightforward
and efficient modular route to introduce electropolymeriz- rod auxiliary electrode, and a Ag/AgCl reference electrode); scan
rates from 20 to 2000 mVs–1 were applied. The experiments were
able 2-thienyl units. The linear arrangement within the
heteroleptic and homoleptic complexes assures identical
spatial separation of the metal centers within the rodlike
performed at concentrations of 10–4 m in degassed solvents (spec-
troscopy grade) containing 0.1 m Bu4NPF6 (dried previously by
heating at 110 °C and stored under vacuum). At the end of each
type II (lateral incorporation) and type III (incorporation
measurement, ferrocene was added as an internal standard.
into the backbone) metallopolymers.
Electropolymerization experiments were executed with the same
The monomer complexes were structurally investigated
set-up by using either a glassy carbon disk electrode or an ITO-
by X-ray crystallography and show a sizeable deviation
coated glass slide (Sigma Aldrich, 0.5ϫ1") as working electrode.
from the ideal linear arrangement. The principle physico-
The polymerization was performed potentiodynamically by apply-
2+
chemical characteristics of the incorporated Ru(dqp)2
ing velocities of 200 mVs–1.
moiety were preserved, that is, broad and strong long-wave-
length UV/Vis absorption features, photoluminescence
quantum yields up to 3% as well as reversible redox pro-
cesses. Both monomers allowed well-defined potentiodyn-
Spectroelectrochemical experiments were performed in a thin-layer
quartz cuvette containing a 0.1 m Bu4NPF6 dichloromethane solu-
tion with an ITO-coated glass slide with the deposited polymer as
the working electrode, a platinum wire auxiliary electrode, and a
amic electropolymerization in the presence of a Lewis acid
Ag/AgCl reference electrode. The potential was controlled by using
to directly yield insoluble polymer films on the electrode
surfaces. The (spectro)electrochemical measurements con-
firmed the stability of the films towards oxidation and re-
a Metrohm Autolab PGSTAT30 potentiostat. The oxidation pro-
cess was monitored by UV/Vis spectroscopy by using a Perkin–
Elmer Lambda 750 UV/Vis spectrophotometer and was considered
vealed reversible redox-triggered switching of their optical complete when there were no further spectral changes.
Eur. J. Inorg. Chem. 2013, 4191–4202
4199
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
For comparison, films of the monomer complexes were prepared
by spin-coating solutions of the monomers (5 mgmL–1 in acetoni-
trile) with a spin coater from Laurell Technologies Corporation
(30 s at 1500 rpm) onto ITO-coated glass substrates (Sigma Ald-
rich, 1ϫ1").
was removed under reduced pressure. The crude product was puri-
fied by flash chromatography (silica gel, SNAP 50 g, adsorbed on
silica, eluted with a gradient of hexane/EtOAc, 10:90 to 0:100, fol-
lowed by evaporation of solvent under reduced pressure). The solid
was triturated and heated to reflux in ethanol; the solids were col-
lected from the hot solution by filtration. The obtained solids were
dried under reduced pressure to yield 5 (0.150 g, 0.233 mmol,
15%).[29] 1H NMR (CD2Cl2, 300 MHz): δ = 8.97 (d, J = 4.3 Hz, 2
H), 8.19 (dd, J = 7.1, 1.3 Hz, 2 H), 8.06 (d, J = 7.8 Hz, 2 H), 7.92
(dd, J = 8.5, 1.5 Hz, 2 H), 7.92 (dd, J = 8.4, 7.2 Hz, 2 H), 7.69
(dm, J = 8.4 Hz, 4 H), 7.62 (dd, J = 8.3, 7.3 Hz, 2 H), 7.44 (dm, J
= 8.4 Hz, 4 H), 7.36 (d, J = 4.3 Hz, 2 H) ppm. 13C NMR (CD2Cl2,
75 MHz): δ = 157.1, 149.8, 147.5, 146.6, 140.0, 137.5, 134.5, 131.9,
131.5, 131.3, 127.0, 126.7, 126.4, 125.7, 122.8, 121.4 ppm. HRMS
(ESI): [M + H]+ calcd. for C35H2279Br81BrN3 644.0158; found
644.0080.
NMR spectra were recorded with a 250 or 300 MHz NMR spec-
trometer (Bruker AVANCE) with samples in deuterated solvents at
25 °C, if not noted otherwise. Chemical shifts are reported in parts
per million (ppm, δ scale) relative to the residual solvent signal.[28]
ESI HRMS spectrometry was performed with an ESI-(Q)-TOF-
MS MICROTOF II (Bruker Daltonics GmbH) mass spectrometer.
X-ray photoelectron spectra were recorded with an EA200-ESCA
system (SPECS) by using a non-monochromatic Al-Kα radiation
source (hυ = 1486.6 eV).
2,6-Bis(2-nitrophenyl)pyridine (3): A flask was charged with potas-
sium 2-nitrobenzoate (13.171 g, 64.2 mmol), 2,6-dibromopyridine
(6.901 g, 29.1 mmol), Pd/C (0.340 g, 0.287 mmol), copper(I) iodide
(0.380 g, 1.995 mmol), 1,10-phenanthroline (0.367 g, 2.037 mmol),
and N-methyl-2-pyrrolidinone (20 mL). The reaction mixture was
heated to 180 °C for 15 h and allowed to cool to room temperature.
The reaction mixture was quenched by addition of water and ex-
tracted three times with dichloromethane. The combined organic
layers were dried with sodium sulfate, filtered, and the excess sol-
vent was removed under reduced pressure (20 mbar) at 70 °C. The
crude product was adsorbed on silica and purified by flash
chromatography (silica gel, SNAP 50 g, eluted with hexane/
CH2Cl2, 30:70 to 0:100, followed by evaporation of solvent under
reduced pressure) to yield 3 (4.501 g, 14.01 mmol, 48%). 1H NMR
(CD2Cl2, 300 MHz): δ = 7.89 (t, J = 7.8 Hz, 1 H), 7.86 (d, J =
7.8 Hz, 2 H), 7.71–7.61 (m, 4 H), 7.55 (td, J = 7.5, 2.0 Hz, 2 H),
7.49 (d, J = 7.9 Hz, 2 H) ppm. 13C NMR (CD2Cl2, 75 MHz): δ =
155.3, 149.3, 137.8, 134.7, 132.7, 131.5, 129.5, 124.4, 122.0 ppm.
HRMS (ESI): calcd. for C17H11N3O4Na [M + Na]+ 344.0642;
found 344.0658. C17H11N3O4 (321.29): calcd. C 63.55, H 3.45, N
13.08; found C 63.31, H 3.39, N 12.91.
[Ru(5)(dqp)](PF6)2 (6):
A flask was charged with [Ru(dqp)-
(CH3CN)3](PF6)2 (0.165 g, 0.156 mmol), 5 (0.100 g, 0.155 mmol),
and ethylene glycol (4 mL), purged with N2 for 30 min, and heated
to 140 °C for 14 h under a N2 atmosphere. The crude reaction mix-
ture was allowed to cool to room temperature and added dropwise
into aqueous NH4PF6. The red solid was filtered, washed with
water, and purified by column chromatography (silica, SNAP,
CH3CN/H2O/neat KNO3, 40:4), the red band was collected and
the counterion exchanged with NH4PF6. Recystallization by vapor
diffusion of diethyl ether into an acetonitrile solution yielded 6
(0.202 g, 0.148 mmol, 95%). 1H NMR (CD3CN, 300 MHz): δ =
8.20 (apparent td, J = 8.1, 1.2 Hz, 2 H), 8.14 (d, J = 5.5 Hz, 2 H),
8.11–8.07 (m, 4 H), 7.94 (d, J = 8.1 Hz, 2 H), 7.91 (d, J = 8.0 Hz,
4 H), 7.80 (d, J = 8.3 Hz, 4 H), 7.73 (d, J = 7.2 Hz, 2 H), 7.69–
7.63 (m, 4 H), 7.50–7.39 (m, 8 H), 7.09 (dd, J = 8.2, 5.5 Hz, 2 H),
7.01 (d, J = 5.3 Hz, 2 H) ppm. 13C NMR (CD3CN, 75 MHz): δ =
159.6, 158.8, 157.9, 157.7, 149.4, 148.0, 147.8, 139.4, 139.2, 138.8,
136.2, 134.2, 133.9, 133.2, 133.1, 133.1, 132.6, 131.9, 129.3, 129.2,
129.2, 128.1, 127.8, 127.8, 125.8, 124.4, 123.3, 123.2 ppm. HRMS
(ESI): calcd. for C58H3679Br81BrN6Ru [M – 2PF6]2+ 539.0200;
found 539.0232.
2,2Ј-(Pyridine-2,6-diyl)dianiline (4): A flask was charged with 3
(4.501 g, 14.01 mmol), Pd/C (0.100 g, 0.084 mmol), ethanol
(60 mL), and hydrazine hydrate (3 mL, 61.5 mmol) in portions. The
reaction mixture was heated to 80 °C for 1 h. The reaction mixture
was allowed to cool to room temperature, filtered, and the remain-
ing solid was washed with dichloromethane. The filtrates were com-
bined, and the excess solvent was removed under reduced pressure
[Ru{2,6-bis(4-{4-(thiophen-2-yl)phenyl}quinolin-8-yl)pyridine}(dqp)]-
(PF6)2 (1): A microwave vial was charged with 6 (0.117 g,
0.086 mmol), thiophen-2-ylboronic acid (0.043 g, 0.336 mmol),
Pd(dba)2 (0.0025 mg, 4.35 μmol), dicyclohexyl(2Ј,6Ј-dimethoxybi-
phenyl-2-yl)phosphane (0.0035 g, 8.53 μmol), and potassium carb-
onate (0.076 g, 0.550 mmol). After the addition of acetonitrile
(3 mL) and water (1.5 mL), the vial was sealed, purged for 10 min
with N2, and conventionally heated to 100 °C for 16 h. The crude
reaction mixture was allowed to cool to room temperature and was
purified as described for 6 to yield 1 (0.108 g, 0.079 mmol, 92%).
1H NMR (CD3CN, 250 MHz): δ = 8.22 (t, J = 8.1 Hz, 1 H), 8.21
(t, J = 8.1 Hz, 1 H), 8.14–8.05 (m, 6 H), 7.99–7.85 (m, 10 H), 7.74
(d, J = 8.1 Hz, 4 H), 7.69 (d, J = 8.2 Hz, 2 H), 7.58 (d, J = 3.6 Hz,
2 H), 7.56–7.39 (m, 10 H), 7.19 (dd, J = 4.9, 3.6 Hz, 2 H), 7.10
(dd, J = 8.1, 5.1 Hz, 2 H), 7.04 (d, J = 5.5 Hz, 2 H) ppm. 13C
NMR (CD3CN, 62 MHz): δ = 159.5, 158.8, 157.9, 157.8, 150.0,
148.0, 147.8, 143.8, 139.4, 139.2, 138.8, 136.3, 136.1, 134.2, 133.9,
133.3, 133.2, 131.9, 131.6, 129.8, 129.4, 129.3, 129.2, 127.9, 127.8,
127.8, 127.4, 127.0, 125.9, 125.5, 123.2, 123.2 ppm. HRMS (ESI):
calcd. for C66H42N6RuS [M – 2PF6]2+ 542.0981; found 542.1057.
to yield
4
(3.150 g, 12.05 mmol, 86%). 1H NMR (CD2Cl2,
300 MHz): δ = 7.87 (t, J = 8.0 Hz, 1 H), 7.53 (d, J = 8.0 Hz, 2 H),
7.52 (dd, J = 7.8, 1.3 Hz, 2 H), 7.17 (td, J = 7.7, 1.4 Hz, 2 H),
6.80–6.72 (m, 4 H), 5.32 (br. s, 4 H) ppm. 13C NMR (CD2Cl2,
75 MHz): δ = 157.8, 146.3, 138.2, 130.0, 129.8, 122.8, 120.1, 117.6,
116.9 ppm. HRMS (ESI): calcd. for C17H16N3 [M + H]+ 284.1158;
found 284.1092. C17H15N3 (261.33): calcd. C 78.13, H 5.59, N
16.08; found C 77.77, H 5.68, N 16.07.
2,6-Bis[4-(4-bromophenyl)quinolin-8-yl]pyridine (5): A flask was
charged with 4 (0.400 g, 1.531 mmol), arsenic pentoxide (1.055 g,
4.59 mmol), and phosphoric acid (85%, 15 mL). The reaction mix-
ture was heated to 100 °C to form a yellow solution, and 1-(4-
bromophenyl)-3-chloropropan-1-one (0.871 g, 3.52 mmol) was
added in portions. After 30 min, the reaction mixture was heated
to 140 °C for 2 h. The mixture was cooled to room temperature,
neutralized by dropwise addition of aqueous sodium hydroxide (to
pH 8), and extracted three times with dichloromethane. The com-
bined organic layers were washed with water and brine, dried with
sodium sulfate, filtered, and adsorbed on silica. The excess solvent
[Ru(8,8Ј-{4-(4-bromophenyl)pyridine-2,6-diyl}diquinoline)2-
(CH3CN)3](PF6)2 (7): A microwave vial was charged with 8,8Ј-
[4-(4-bromophenyl)pyridine-2,6-diyl]diquinoline[21] (0.300 g,
0.614 mmol), ruthenium trichloride hydrate (0.164 g, 0.585 mmol),
and ethanol (10 mL). The vial was sealed and conventionally
Eur. J. Inorg. Chem. 2013, 4191–4202
4200
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
heated to 115 °C for 13 h. The reaction mixture was allowed to
cool to room temperature, filtered, and washed thoroughly with
ethanol and dichloromethane. The dark brown solid was dried un-
der reduced pressure to yield crude Ru{8,8Ј-[4-(4-bromophenyl)-
pyridine-2,6-diyl]diquinoline}Cl3 (0.352 g, 0.506 mmol, 86%).
data. We will not deposit the data in the Cambridge Crystallo-
graphic Data Centre. Crystallographic data as well as structure
solution and refinement details are summarized in the Supporting
Information. XP (Siemens Analytical X-ray Instruments, Inc.) was
used for structure representations.
A flask was charged with crude Ru{8,8Ј-[4-(4-bromophenyl)pyr-
idine-2,6-diyl]diquinoline}Cl3 (0.352 g, 0.506 mmol), silver nitrate
(0.301 g, 1.771 mmol), acetonitrile (7 mL), ethanol (1.5 mL), and
water (1.5 mL). The suspension was heated to 90 °C for 17 h. The
white solids were removed by filtration, and the orange-brown solu-
tion was reduced in volume under reduced pressure and purified
as described for 6 to yield 7 (0.305 g, 0.304 mmol, 60%). 1H NMR
(CD3CN, 300 MHz): δ = 9.08 (dd, J = 5.2, 1.2 Hz, 2 H), 8.67 (br.
d, J = 7.5 Hz, 2 H), 8.62 (dd, J = 8.4, 1.2 Hz, 2 H), 8.28 (d, J =
8.2 Hz, 2 H), 8.18 (s, 2 H), 7.93 (t, J = 7.8 Hz, 2 H), 7.87 (dm, J
= 8.5 Hz, 2 H), 7.77 (dm, J = 8.5 Hz, 2 H), 7.67 (dd, J = 8.2,
5.2 Hz, 2 H), 2.45 (s, 3 H), 1.96 (s, 6 H) ppm. 13C NMR (CD3CN,
63 MHz): δ = 159.8, 158.6, 150.0, 147.3, 139.7, 136.2, 135.3, 135.2,
133.5, 133.0, 130.4, 129.7, 129.2, 127.8, 127.3, 126.3, 125.5, 123.2,
4.5, 3.8 ppm. HRMS (ESI): calcd. for C35H2779BrN6Ru [M –
2PF6]2+ 356.0259; found 355.8804.
CCDC-888462 (for 6) and -888463 (for 2) contain the supplemen-
tary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
Computational Methods: All calculations were performed with the
Gaussian 09 (G09) program package[32] by employing the DFT
method and using Becke’s three-parameter hybrid functional[33]
and the Lee–Yang–Parr gradient-corrected correlation func-
tional[34] (B3LYP). The ruthenium atoms were treated by the 28-
electron relativistic effective core potential MWB[35] for the inner
shells, whereas the outer shells (4s, 4p, 4d, and 5s electrons) were
treated separately. The remaining atoms (C, H, N, and S) were
treated with the 6-31G(d) double-ζ basis set.[36] Bulk solvent effects
(acetonitrile) were included by using the integral equation formal-
ism of the polarizable continuum model of Tomasi and co-
workers.[37] The geometry optimizations of the singlet and triplet
states were performed without any constraints and the true nature
was confirmed by normal-mode analysis. The molecular orbitals
and electron/spin densities were visualized by using the
GaussView 5.0 package.[38] The vertical excitations were computed
by TD-DFT at the same level of theory. The electronic transitions
were determined from the changes in electronic distribution by
using electron-density difference maps (EDDMs),[39] which were
computed with the GaussSum 2.2 package.[40] The triplet excited
states were visualized by spin-density plots, expressed as a differ-
ence between α and β spin densities, by using the GaussView 5.0
package.[38] The density of states (DOS) and crystal orbital overlap
population (COOP) analysis was performed by using the
GaussSum 2.2 package.[40]
[Ru{8,8Ј-(4-{4-(thiophen-2-yl)phenyl}pyridine-2,6-diyl)diquinoline}2]-
(PF6)2 (2): A flask was charged with 7 (0.195 g, 0.195 mmol), 8,8Ј-
[4-(4-bromophenyl)pyridine-2,6-diyl]diquinoline
(0.095 g,
0.195 mmol), and ethylene glycol (10 mL) and heated to 140 °C
under N2 for 14 h. The crude reaction mixture was allowed to cool
to room temperature and was worked up as described for 6 to yield
crude
[Ru(8,8Ј-{4-(4-bromophenyl)pyridine-2,6-diyl}diquin-
oline)2](PF6)2 containing some monodebrominated product
(0.180 g, 0.132 mmol, 68%).
A microwave vial was charged with crude [Ru(8,8Ј-{4-(4-bromo-
phenyl)pyridine-2,6-diyl}diquinoline)2](PF6)2
(0.180 g,
0.132 mmol), thiophen-2-ylboronic acid (0.067 g, 0.526 mmol),
Pd(dba)2 (0.0038 g, 6.58 μmol), dicyclohexyl(2Ј,6Ј-dimethoxybi-
phenyl-2-yl)phosphane (0.0054 g, 0.013 mmol), and potassium
carbonate (0.111 g, 0.790 mmol). After the addition of acetonitrile
(3 mL) and water (1.5 mL), the vial was sealed, purged for 10 min
with N2, and conventionally heated to 100 °C for 16 h. The reaction
mixture was worked up as described for 6 to yield the title com-
Supporting Information (see footnote on the first page of this arti-
cle): Spectral data (UV/Vis, UV/Vis/NIR spectroelectrochemistry,
XPS), DFT calculation results, and X-ray crystallographic data.
2
(0.100 g, 0.073 mmol, 55%). 1H NMR (CD3CN,
Acknowledgments
pound
300 MHz): δ = 8.18–8.13 (m, 8 H), 8.09 (dd, J = 8.1, 1.2 Hz, 4 H),
8.01 (d, J = 8.5 Hz, 4 H), 7.93 (dd, J = 7.4, 1.2 Hz, 4 H), 7.88 (d,
J = 8.5 Hz), 7.72 (dd, J = 8.2, 1.2 Hz, 4 H), 7.58 (dd, J = 3.6,
1.0 Hz, 2 H), 7.50 (t, J = 7.5 Hz, 4 H), 7.50 (dd, J = 5.1, 1.0 Hz, 2
H), 7.18 (dd, J = 5.1, 3.7 Hz, 2 H), 7.09 (dd, J = 8.2, 5.2 Hz, 4HH)
ppm. 13C NMR (CD3CN, 75 MHz): δ = 159.6, 158.1, 149.8, 147.7,
143.7, 138.6, 137.2, 135.8, 134.5, 133.0, 131.6, 129.7, 129.2, 127.9,
127.6, 127.5, 127.4, 126.1, 125.7, 123.1 ppm. HRMS (ESI): calcd.
for C66H42N6RuS2 [M – 2PF6]2+ 542.0981; found 542.0990.
The authors acknowledge the Bundesministerium für Bildung und
Forschung, the European Social Fund (ESF), the Thüringer Auf-
baubank (TAB), and the Thuringian Ministry of Economy, Em-
ployment and Technology (TMWAT) for financial support. M. J.
is grateful for financial support from the Carl-Zeiss-Stiftung. The
authors also thank Esra Altuntas for ESI measurements and
Dr. Bernd Schröter for XPS measurements.
[1] a) A. Juris, V. Balzani, F. Barigelletti, S. Campagna, P. Belser,
A. von Zelewsky, Coord. Chem. Rev. 1988, 84, 85–277; b) T. J.
Meyer, Pure Appl. Chem. 1986, 58, 1193–1206.
[2] B. Durham, J. V. Caspar, J. K. Nagle, T. J. Meyer, J. Am. Chem.
Soc. 1982, 104, 4803–4810.
[3] J.-P. Sauvage, J.-P. Collin, J.-C. Chambron, S. Guillerez, C. Co-
udret, V. Balzani, F. Barigelletti, L. De Cola, L. Flamigni,
Chem. Rev. 1994, 94, 993–1019.
[4] a) E. C. Constable, Chem. Soc. Rev. 2007, 36, 246–253; b) H.
Hofmeier, U. S. Schubert, Chem. Soc. Rev. 2004, 33, 373–399.
[5] J. R. Winkler, T. L. Netzel, C. Creutz, N. Sutin, J. Am. Chem.
Soc. 1987, 109, 2381–2392.
X-ray Crystallography: The intensity data for the compounds were
collected with a Nonius KappaCCD diffractometer with graphite-
monochromated Mo-Kα radiation. The data were corrected for Lo-
rentz and polarization effects but not for absorption effects.[30]
The structures were solved by direct methods (SHELXS[31]) and
2
refined by full-matrix least-squares techniques against Fo
(SHELXL-97[31]). All hydrogen atoms were included at calculated
positions with fixed thermal parameters. All non-hydrogen, non-
disordered atoms were refined anisotropically.[31] The crystals of 1
were extremely thin and of low quality and resulted in a substan-
dard data set; however, the structure is sufficient to show the con-
nectivity and geometry despite the high final R value. We will only
publish the conformation of the molecule and the crystallographic
[6] a) B. Schulze, D. Escudero, C. Friebe, R. Siebert, H. Görls, S.
Sinn, M. Thomas, S. Mai, J. Popp, B. Dietzek, L. González,
U. S. Schubert, Chem. Eur. J. 2012, 18, 4010–4025; b) O. A.
Eur. J. Inorg. Chem. 2013, 4191–4202
4201
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Borg, S. S. M. C. Godinho, M. J. Lundqvist, S. Lunell, P.
Persson, J. Phys. Chem. A 2008, 112, 4470–4476; c) A. Amini,
A. Harriman, A. Mayeux, Phys. Chem. Chem. Phys. 2004, 6,
1157–1164; d) D. W. Fink, W. E. Ohnesorge, J. Am. Chem. Soc.
1969, 91, 4995–4998.
a) M. Jäger, A. Smeigh, F. Lombeck, H. Görls, J.-P. Collin, J.-
P. Sauvage, L. Hammarström, O. Johansson, Inorg. Chem.
2010, 49, 374–376; b) S. H. Wadman, M. Lutz, D. M. Tooke,
A. L. Spek, F. Hartl, R. W. A. Havenith, G. P. M. van Klink,
G. van Koten, Inorg. Chem. 2009, 48, 1887–1900; c) P. G.
Bomben, K. C. D. Robson, P. A. Sedach, C. P. Berlinguette, In-
org. Chem. 2009, 48, 9631–9643; d) J.-P. Collin, M. Beley, J.-P.
Sauvage, F. Barigelletti, Inorg. Chim. Acta 1991, 186, 91–93.
a) H.-J. Park, K. H. Kim, S. Y. Choi, H.-M. Kim, W. I. Lee,
Y. K. Kang, Y. K. Chung, Inorg. Chem. 2010, 49, 7340–7352;
b) B. Schulze, D. Escudero, C. Friebe, R. Siebert, H. Görls, U.
Köhn, E. Altuntas, A. Baumgaertel, M. D. Hager, A. Winter,
B. Dietzek, J. Popp, L. González, U. S. Schubert, Chem. Eur.
J. 2011, 17, 5494–5498.
a) L. Hammarström, O. Johansson, Coord. Chem. Rev. 2010,
254, 2546–2559; b) A. Breivogel, C. Förster, K. Heinze, Inorg.
Chem. 2010, 49, 7052–7056; c) F. Schramm, V. Meded, H. Fli-
egl, K. Fink, O. Fuhr, Z. Qu, W. Klopper, S. Finn, T. E. Keyes,
M. Ruben, Inorg. Chem. 2009, 48, 5677–5684.
[21] M. Jäger, L. Eriksson, J. Bergquist, O. Johansson, J. Org.
Chem. 2007, 72, 10227–10230.
[22] S. Jin, G. Xue, Macromolecules 1997, 30, 5753–5757.
[23] P. J. Hochgesang, R. D. Bereman, Inorg. Chim. Acta 1988, 149,
69–76.
[24] J. J. Apperloo, L. B. Groenendaal, H. Verheyen, M. Jayakan-
nan, R. A. J. Janssen, A. Dkhissi, D. Beljonne, R. Lazzaroni,
J.-L. Brédas, Chem. Eur. J. 2002, 8, 2384–2396.
[25] a) A. J. Bard, L. R. Faulkner, Electrochemical Methods, 2nd
ed., John Wiley & Sons, New York, 2001; b) the relationship
between the peak current and scan rate may also be strongly
influenced by film thickness.
[26] a) M. Devenney, L. A. Worl, S. Gould, A. Guadalupe, B. P.
Sullivan, J. V. Caspar, R. L. Leasure, J. R. Gardner, T. J. Meyer,
J. Phys. Chem. A 1997, 101, 4535–4540; b) J. L. Colón, C. Y.
Yang, A. Clearfield, C. R. Martin, J. Phys. Chem. 1988, 92,
5777–5781.
[7]
[8]
[9]
[27] C. L. Gaupp, D. M. Welsh, R. D. Rauh, J. R. Reynolds, Chem.
Mater. 2002, 14, 3964–3970.
[28] H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997,
62, 7512–7515.
[29] A small amount of inseparable impurities remained and were
readily removed after the coordination step.
[30] a) COLLECT, Data Collection Software, Nonius B. V., The Ne-
therlands, 1998; b) Z. Otwinowski, W. Minor, in: Processing of
X-ray diffraction data collected in oscillation mode, in: Methods
in Enzymology, vol. 276: Macromolecular Crystallography, Part
A (Eds.: C. W. Carter, R. M. Sweet), Academic Press, New
York, 1997, p. 307–326.
[10]
[11]
a) J. Wang, Y.-Q. Fang, L. Bourget-Merle, M. I. J. Polson, G. S.
Hanan, A. Juris, F. Loiseau, S. Campagna, Chem. Eur. J. 2006,
12, 8539–8548; b) J. Wang, G. S. Hanan, F. Loiseau, S. Cam-
pagna, Chem. Commun. 2004, 2068–2069.
[31] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
[32] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B.
Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li,
H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Son-
nenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hase-
gawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai,
T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Starov-
erov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell,
J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M.
Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Ad-
amo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Mar-
tin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B.
Foresman, J. V. Ortiz, J. Cioslowski, D. J. Fox, Gaussian 09,
revision A.1, Gaussian, Inc., Wallingford, 2009.
a) M. Jäger, R. J. Kumar, H. Görls, J. Bergquist, O. Johansson,
Inorg. Chem. 2009, 48, 3228–3238; b) M. Abrahamsson, M.
Jäger, T. Österman, L. Eriksson, P. Persson, H.-C. Becker, O.
Johansson, L. Hammarström, J. Am. Chem. Soc. 2006, 128,
12616–12617.
a) C. Friebe, M. D. Hager, A. Winter, U. S. Schubert, Adv. Ma-
ter. 2012, 24, 332–345; b) M. O. Wolf, J. Inorg. Organomet. Po-
lym. Mater. 2006, 16, 189–199.
a) Y.-W. Zhong, C.-J. Yao, H.-J. Nie, Coord. Chem. Rev. 2013,
257, 1357–1372; b) C.-J. Yao, Y.-W. Zhong, H.-J. Nie, H. D.
Abruña, J. Yao, J. Am. Chem. Soc. 2011, 133, 20720–20723; c)
X. J. Zhu, B. J. Holliday, Macromol. Rapid Commun. 2010, 31,
904–909; d) J. Hjelm, R. W. Handel, A. Hagfeldt, E. C. Con-
stable, C. E. Housecroft, R. J. Forster, Inorg. Chem. 2005, 44,
1073–1081; e) R. M. Leasure, T. Kajita, T. J. Meyer, Inorg.
Chem. 1996, 35, 5962–5963; f) C. P. Horwitz, Q. Zuo, Inorg.
Chem. 1992, 31, 1607–1613; g) T. F. Guarr, F. C. Anson, J.
Phys. Chem. 1987, 91, 4037–4043; h) C. D. Ellis, L. D.
Margerum, R. W. Murray, T. J. Meyer, Inorg. Chem. 1983, 22,
1283–1291; i) P. Denisevich, H. D. Abruña, C. R. Leidner, T. J.
Meyer, R. W. Murray, Inorg. Chem. 1982, 21, 2153–2161.
[12]
[13]
[33] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[34] C. T. Lee, W. T. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–
789.
[35] D. Andrae, U. Häußermann, M. Dolg, H. Stoll, H. Preuß,
Theor. Chim. Acta 1990, 77, 123–141.
[14] M. O. Wolf, Adv. Mater. 2001, 13, 545–553.
[15] a) M. L. Naklicki, S. I. Gorelsky, W. Kaim, B. Sarkar, R. J.
Crutchley, Inorg. Chem. 2012, 51, 1400–1407; b) M. Lundberg,
P. E. M. Siegbahn, J. Chem. Phys. 2005, 122, 224103.
[16] Y. Sun, M. El Ojaimi, R. Hammitt, R. P. Thummel, C. Turro,
J. Phys. Chem. B 2010, 114, 14664–14670.
[36] P. C. Hariharan, J. A. Pople, Theor. Chim. Acta 1973, 28, 213–
222.
[37] J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 2005, 105,
2999–3094.
[38] GaussView, v. 5.0.8, R. D. Dennington, T. A. Keith, J. M. Mil-
lan, Semicchem, Inc., Shawnee KS, 2008.
[39] W. R. Browne, N. M. O’Boyle, J. J. McGarvey, J. G. Vos, Chem.
Soc. Rev. 2005, 34, 641–663.
[17] J. Heinze, B. A. Frontana-Uribe, S. Ludwigs, Chem. Rev. 2010,
110, 4724–4771.
[18] L. J. Goossen, N. Rodríguez, B. Melzer, C. Linder, G. Deng,
L. M. Levy, J. Am. Chem. Soc. 2007, 129, 4824–4833.
[19] L. J. Gooßen, B. Zimmermann, C. Linder, N. Rodríguez, P. P.
Lange, J. Hartung, Adv. Synth. Catal. 2009, 351, 2667–2674.
[20] D. Pomeranc, V. Heitz, J.-C. Chambron, J.-P. Sauvage, J. Am.
Chem. Soc. 2001, 123, 12215–12221.
[40] a) GaussSum, v. 2.2.5, N. M. O’Boyle, J. G. Vos, Dublin City
University, Dublin, Ireland, 2009; available at http://gausssum.-
sourceforge.net ; b) N. M. O’Boyle, A. L. Tenderholt, K. M.
Langner, J. Comput. Chem. 2008, 29, 839–845.
Received: March 15, 2013
Published Online: June 26, 2013
Eur. J. Inorg. Chem. 2013, 4191–4202
4202
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim