Novel Ru(ii) sensitizers bearing an unsymmetrical pyridine-quinoline hybrid ligand with extended π-conjugation: Synthesis and application in dye-sensitized solar cells

Article abstract of DOI:10.1039/c3dt33063j

Heteroleptic ruthenium(ii) sensitizers DV42 and DV51, encompassing a novel unsymmetrical pyridine-quinoline hybrid ligand with extended π-conjugation, were synthesized, characterized, and utilized in nanocrystalline dye-sensitized solar cells. Due to the extended conjugation of DV42 and DV51, the absorption of the corresponding sensitized TiO₂ films extends into the red spectral range, shifted by 30-40 nm relative to the absorption of TiO ₂ films sensitized with the standard Z907 ruthenium(ii) dye. Contact angle measurements of DV42- and DV51-sensitized TiO₂ films suggest that these films are hydrophilic with contact angle values commonly observed upon sensitization with the standard N3 ruthenium(ii) dye. Electrochemical studies of the novel ruthenium(ii) dyes show that their first oxidation potentials lie well below the I^-/I₃^- redox potential allowing easy regeneration. The excited-state oxidation potentials of both dyes lie above the TiO₂ conduction band, permitting efficient electron injection from the excited dye molecules into the semiconductor conduction band. Liquid electrolyte dye-sensitized solar cells incorporating DV42- or DV51-sensitized TiO₂ photoelectrodes afford overall power conversion efficiencies of 3.24 or 4.36% respectively. These efficiencies are up to 56% of the power conversion efficiencies attained by TiO₂ photoelectrodes sensitized by the benchmark Z907 ruthenium(ii) dye under similar experimental conditions.

Full text of DOI:10.1039/c3dt33063j

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Novel Ru(II) sensitizers bearing an unsymmetrical
pyridine-quinoline hybrid ligand with extended
π-conjugation: synthesis and application in
dye-sensitized solar cells
Cite this: Dalton Trans., 2013, 42, 6582
Georgios C. Vougioukalakis,* Thomas Stergiopoulos, Athanassios G. Kontos,
Eleftherios K. Pefkianakis, Kyriakos Papadopoulos and Polycarpos Falaras*
Heteroleptic ruthenium(II) sensitizers DV42 and DV51, encompassing a novel unsymmetrical pyridine-
quinoline hybrid ligand with extended π-conjugation, were synthesized, characterized, and utilized in
nanocrystalline dye-sensitized solar cells. Due to the extended conjugation of DV42 and DV51, the
absorption of the corresponding sensitized TiO2 films extends into the red spectral range, shifted by
30–40 nm relative to the absorption of TiO2 films sensitized with the standard Z907 ruthenium(II) dye.
Contact angle measurements of DV42- and DV51-sensitized TiO2 films suggest that these films are
hydrophilic with contact angle values commonly observed upon sensitization with the standard N3
ruthenium(II) dye. Electrochemical studies of the novel ruthenium(II) dyes show that their first oxidation
potentials lie well below the I−/I3 redox potential allowing easy regeneration. The excited-state oxi-
−
dation potentials of both dyes lie above the TiO2 conduction band, permitting efficient electron injection
from the excited dye molecules into the semiconductor conduction band. Liquid electrolyte dye-
sensitized solar cells incorporating DV42- or DV51-sensitized TiO2 photoelectrodes afford overall power
conversion efficiencies of 3.24 or 4.36% respectively. These efficiencies are up to 56% of the power con-
version efficiencies attained by TiO2 photoelectrodes sensitized by the benchmark Z907 ruthenium(II) dye
under similar experimental conditions.
Received 1st August 2012,
Accepted 18th February 2013
DOI: 10.1039/c3dt33063j
www.rsc.org/dalton
The dye, which is the heart of the DSC device, undergoes
transition from its ground to its excited state upon light
Introduction
Dye-sensitized solar cells (DSCs) provide an appealing alterna- absorption. One electron is then injected from the dye into the
tive to the conventional solid-state cells.1 This is mainly due to conduction band of the semiconductor, resulting in an oxi-
their ability to function indoors and under low light con- dized dye molecule. Ideally, the photoinjected electrons perco-
ditions, their potential transparency and flexibility, their invar- late through the nanoparticulate network and get collected at
iant efficiency with the operating temperature, and their the back contact. They subsequently flow through the external
relatively low production cost.2 A DSC device is comprised of circuit performing useful work and eventually arrive at the
two facing electrodes: the photoanode, which is transparent, counter electrode. The oxidized dye is regenerated to its
consists of a thin nanoparticulate film of a mesoporous large ground state by taking an electron from the redox couple,
band gap semiconductor deposited on conductive glass. which is in turn reduced at the counter electrode completing
A monolayer of sensitizer (dye) molecules is chemically grafted the circuit. The device maximum voltage (Vmax) that can be
onto the semiconductor through anchoring functionalities generated equals the difference between the semiconductor
such as carboxylic or phosphonic acids. An electrolyte contain- Fermi level and the electron mediator species’ redox potential.
ing a redox couple is placed between the photoelectrode and The result of this regenerative cycle is the conversion of
the cathode, typically comprised of platinum deposited onto photons to electrons, while no chemical species undergo per-
conductive glass, to transfer the charges.
manent transformations.
Charge generation, i.e., the injection of an electron from
the dye into the conduction band of the semiconductor, is one
of the most important steps in the DSC photoelectrochemical
cycle. The “ideal” dye has to fulfill a variety of photophysical,
Division of Physical Chemistry, IAMPPNM, NCSR Demokritos, Agia Paraskevi Attikis,
15310 Athens, Greece. E-mail: vougiouk@chem.demokritos.gr,
papi@chem.demokritos.gr; Fax: +30-210-6511766; Tel: +30-210-6503644
6582 | Dalton Trans., 2013, 42, 6582–6591
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
structural, and electrochemical requirements: (i) a broad,
strong absorption extending from the visible to the near-infra-
red; (ii) chemical stability in its ground, excited, and oxidized
states; (iii) its excited state should not be deactivated through
the emission of heat or light; (iv) an irreversible adsorption
onto the surface of the semiconductor; (v) a strong electronic
coupling between its excited state and the semiconductor’s
conduction band to ensure efficient electron injection; (vi) an
excited-state oxidation potential sufficiently higher (by about
150–200 mV)3 than the semiconductor conduction band edge
in order to result in an effective electron injection; and (vii) a
ground-state reduction potential sufficiently lower (by about
200–300 mV)3 than the redox potential of the electron
mediator species, in order to be regenerated rapidly.
Fig. 2 Ru(II) dyes DV42 and DV51 bearing a novel bidentate pyridine-quino-
line hybrid ligand with extended aromatic conjugation.
Although many different compounds have been utilized as
sensitizers in DSC applications,2 Ru(II) polypyridyl complexes
comprise the most successful family of DSC dyes in terms of
both photovoltaic performance and long-term stability. N3 and
N719 (Fig. 1) were the first high performance Ru(II) sensitizers
to be reported.4 N3 and N719, which differ only in their proto-
nation state, afford a very high conversion of incident photons
into electric current over a large spectral range, giving DSCs
with overall power conversion efficiencies of 10.0% and 11.2%
respectively. The higher power conversion efficiency of N719,
when compared to the tetra-protonated parent dye N3, origi-
nates from the increased cell voltage. The structural fine-
tuning of these early Ru(II) polypyridyl dyes has led to even
more efficient sensitizers with amphiphilic properties and
extended conjugation that achieve power conversion efficien-
cies up to 12.3%. Thus, CYC-B1 (Fig. 1), reported in 2006,
bears a highly conjugated ligand substituted with alkyl bithio-
phene groups.5 The overall power conversion efficiency of
CYC-B1 was reported to be 8.5%, 10% higher than that of N3
under the same cell fabrication and measuring conditions.
Z991, the mono(tetrabutylammonium) salt of CYC-B1 (Fig. 1),
holds the 12.3% DSC power conversion efficiency record. This
DSC applications,7 we herein report the synthesis, characteri-
zation, and sensitizing ability evaluation of two new Ru(II) dyes
(DV42 and DV51, Fig. 2). DV42 and DV51 bear a novel pyri-
dine-quinoline hybrid ligand with extended aromatic conju-
gation. The unsymmetrical, bidentate pyridine-quinoline
ligand in DV42 and DV51 was designed with a long aromatic
moiety. This novel architecture should facilitate an increased
extinction coefficient of the main metal-to-ligand charge trans-
fer (MLCT) band of its Ru(II) complexes and an increased
response to red light in the corresponding sensitized TiO2 elec-
trodes.5,8 Moreover, this ligand scaffold was constructed in
order to enable the detailed study of the influence of exceed-
ingly long aromatic moieties on the hydrophobicity and
wettability of the corresponding Ru(II) polypyridyl dyes.
Experimental
Materials
was achieved with phosphinate amphiphile dineohexyl bis- Unless otherwise stated, all reagents were purchased from
(3,3-dimethyl-butyl)-phosphinic acid (DINHOP) as the co- Aldrich. Solvents were purchased from Fluka or Panreac and
adsorbent.6
used without further purification. A dichloro(p-cymene)ruthe-
The structural and photoelectrochemical properties of a nium(II) dimer was purchased from Alfa Aesar. 2,2′-Bipyridine-
Ru(II) polypyridyl dye and, therefore, its efficiency in a DSC are 4,4′-dicarboxylic acid (H2dcbpy) was obtained from Dyesol.
dictated by the coordinated ligands. As part of our ongoing Acetonitrile, used in the electrochemical studies, was degassed
work in the development of novel Ru(II) photosensitizers for by a freeze–pump–thaw cycle and stirred overnight over P2O5,
at room temperature, under an argon atmosphere. It was then
distilled under argon and degassed again via another freeze–
pump–thaw cycle, before being stored over activated 4 Å molecular
sieves. DMF, used in the electrochemical studies, was degassed
by a freeze–pump–thaw cycle and stirred overnight over acti-
vated MgSO4, at room temperature, under an argon atmos-
phere. A further quantity of activated MgSO4 was added and
DMF was then distilled under reduced pressure, before being
stored over activated 4 Å molecular sieves. All synthetic reac-
tions were performed under an argon atmosphere in degassed
(15 min of argon bubbling) solvents. cis-Diisothiocyanato-(2,2′-
bipyridyl-4,4′-dicarboxylic acid)-(4,4′-dinonyl-2,2′-bipyridine)-
ruthenium(II) dye (Z907, Fig. 3) was purchased from Dyesol.
Fig. 1 Molecular structures of Ru(II) polypyridyl dyes N3, N719, CYC-B1, and
Z991.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6582–6591 | 6583

View Article Online
Paper
Dalton Transactions
deoxygenated solutions purified by N2 gas (99.9%) for 30 min
prior to use. During the experiments, N2 was passed over the
solution surface.
Electrodes preparation and devices assembly-characterization
Titania films were prepared using the doctor-blade technique
by depositing one transparent layer (7–8 µm, measured with
an AMBIOS XP-2 profilometer) of
a commercial paste
(DSL18NR-T, Dyesol) onto a pre-cleaned fluorine-doped tin
oxide (FTO) conductive glass (Pilkington Solar, 3.2 mm thick-
ness). The DSL18NR-T paste solely consists of anatase particles
and elongated nanorods with an average size of 17 nm, while
its pore size distribution is wide, presenting a maximum at
Fig. 3 Ru(II) complex Z907 was utilized as a benchmark dye in the present
study.
Analytical measurements
1H, 13C, HSQC, and COSY Nuclear Magnetic Resonance (NMR) 30 nm. The films were then dried at 125 °C for 6 minutes.
spectra were obtained with a Bruker Avance 500 MHz spec- A second layer (4.5–5.0 µm) was deposited on top by doctor-
trometer in (CD3)2SO or CDCl3. Chemical shifts are reported in blading a paste of large scattering particles (WER2-O, average
ppm downfield from Me4Si, by using the residual solvent peak size of cubic-like particles: 150–250 nm, Dyesol). Next, the
as an internal standard. Data for NMR spectra are reported as films were gradually annealed at 125 °C (5 min), 325 °C
follows: chemical shift (δ ppm), multiplicity, and integration. (15 min) and then at 525 °C for 30 min. The double layered
Electronic spectra (UV-vis) in solution [5 × 10−5 M in ethanol films were post-treated at 70 °C for 60 min with an aqueous
(DV51 and Z907) or DMF (DV42)] were recorded with a Hitachi solution of 0.04 M TiCl4, then thoroughly cleaned with water
3010 spectrophotometer in standard 1 cm path length cuv- and ethanol and re-annealed at 450 °C for 1 h. The thermally-
ettes. Diffuse reflectance (R%) and transmittance (T%) UV-vis treated titania films were sensitized by overnight immersion in
spectra of the dye-sensitized TiO2 films (for this purpose one 0.3 mM dye solutions. The solvents utilized were DMF for
layer of a transparent Dyesol paste was doctor-bladed onto DV42, ethanol for DV51, and acetonitrile/tert-butanol for Z907.
microscopy glass) were obtained on the same instrument In the case of Z907 the corresponding solution also contained
using an integrating sphere with 60 mm diameter and were an equimolar quantity of chenodeoxycholic acid as a co-
transformed in absorbance (A%) units (A% = 100 − R% − T%). adsorbent.
Infrared spectra (FTIR) in the powder form were obtained
In order to evaluate their photovoltaic performance, the
using an FTIR Nicolet 6700 spectrometer. Data for FTIR dye-sensitized films were incorporated as the active photoelec-
spectra are reported as follows: frequency (cm−1) and strength trode (PE) in a lamellar solar cell configuration of the following
(s = strong; m = medium; w = weak). Despite our repeated structure: conductive glass/TiO2 + dye/electrolyte/Pt/conductive
attempts, we were unable to get reliable Elemental Analysis glass. 1.1 mm thick TCO10-10 conductive glasses (Solaronix),
data for the newly synthesized complexes, possibly due to platinized by sputtering (resulting in a thickness for the
insufficient combustion. High Resolution Mass Spectra Pt layer of about 100 nm), were used as counter electrodes
(HRMS) were recorded using a Bruker Daltonics APEXIV 4.7 (CE). Non-sealed DSCs were fabricated by placing a drop of a
Tesla Fourier Transform Ion Cyclotron Resonance Mass spec- liquid electrolyte (1.0 M 1,3-dimethylimidazolium iodide,
trometer. Micro-Raman spectra were measured in the backscat- 50 mM LiI, 15 mM I2, 0.5 M tert-butylpyridine, and 0.1 M
tering configuration using a Renishaw inVia spectrometer with guanidinium thiocyanate dissolved in acetonitrile–valeronitrile
an excitation at 514.5 nm by an Ar+ ion laser. In all cases, very (85 : 15)) onto the photoelectrode and sandwiching the CE on
low laser power density (0.01 mW μm−2) was applied in order top of the PE.
to avoid dye degradation and thermal shifts of the modes. The
Current–voltage (I–V) measurements were performed by
advancing contact angle of water droplets on the films was illuminating the DSCs using solar-simulated light (1 sun AM
measured with a Contact Angle Meter (CAM) 100, KSV Instru- 1.5 G, 1000 W m−2), from a 300 W Xe lamp. Illumination
ments, Ltd, equipped with a horizontal microscope and a pro- under 400 nm was excluded using a cut-off UV filter (Oriel).
tractor eyepiece. Cyclic voltammetry measurements were The active area of the DSCs was set at 0.15 cm2, using a black
performed in 10−3 M solutions of the corresponding com- metallic mask of very large area to exclude parasitic light in
plexes in acetonitrile (DV42) or DMF (DV51) containing 0.1 M front of the cell (while the aperture area was larger, 0.36 cm2).9
tetrabutylammonium tetrafluoroborate (TBATFB) as a support- I–V characteristics were obtained using linear sweep voltam-
ing electrolyte, at a sweep rate of 100 mV s−1, using the metry (scan speed: 50 mV s−1) on an Autolab PGSTAT-30
Autolab PGSTAT-30 potentiostat (Ecochemie). A three-elec- potentiostat.
trode, one-compartment electrochemical cell was used, with
Metrohm dot (6.0302.000) and planar (6.0305.000) platinum
Synthesis
working and counter electrodes, respectively, as well as a
Synthesis
of
6-bromo-4-phenyl-2-pyridin-2-ylquinoline
Metrohm Ag/AgCl 6.0726.100 reference electrode (LiCl satu- (1).10 In a round-bottomed flask equipped with a reflux con-
rated in EtOH). The experiments were performed in denser and a magnetic stirrer under an argon atmosphere
6584 | Dalton Trans., 2013, 42, 6582–6591
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
were added 2-amino-5-bromobenzophenone
Paper
(5.5
g, (m), 783.0 (m), 730.9 (s), 698.1 (s), 617.1 (m), 590.1 (m), 520.7
20.0 mmol), acetic acid (20 mL), and H2SO4 (98%, 0.1 mL), (m), 424.3 (w). Anal. Calcd for C42H28N2: C, 89.97; H, 5.03;
along with an excess of 2-acetylpyridine (4.5 mL, 40.0 mmol). N, 5.00. Found: C, 89.98; H, 5.06; N, 4.96.
This reaction mixture was heated at 110 °C under stirring for
48 h. Upon cooling to room temperature, a needle-like solid necked 50 mL round-bottomed flask, equipped with a stirring
precipitated from the resulting reaction mixture. This was fil- bar and reflux condenser, was charged with [RuCl-
Synthesis of cis-[Ru(LP1)(H2dcbpy)Cl2] (DV42). A two-
a
tered and washed repeatedly with hot water and methanol. (p-cymene)]2 (76.5 mg, 0.125 mmol), LP1 (140.2 mg,
Subsequent drying under vacuum afforded 1 (6.5 g, 90%). 0.250 mmol), and degassed DMF (35 mL) at room temperature
1H NMR (CDCl3, 500 MHz): δ 8.72 (d, 1H), 8.67 (d, 1H), 8.54 under argon. The reaction mixture was heated in the dark at
(s, 1H), 8.11 (d, 1H), 8.08 (d, 1H), 7.87 (t, 1H), 7.80 (dd, 1H), 60 °C for 4 hours. H2dcbpy (61.1 mg, 0.250 mmol) was then
7.49–7.56 (m, 5H), 7.36 (t, 1H).
added and the reaction mixture was heated in the dark under
Synthesis of 4-phenyl-2-pyridin-2-yl-6-(4-vinyl-phenyl)-quino- argon at 140 °C for 4 hours. This solution was cooled to room
line (3). In a degassed three-necked flask equipped with a temperature with a water bath, and DMF was removed on a
reflux condenser and a magnetic stirrer were added 1 (4.0 g, rotary evaporator. The resulting black solid was taken up and
11.0 mmol), (4-vinylphenyl)boronic acid (4.5 g, 22.0 mmol), washed on a G4 sintered glass crucible with Et2O (3 × 10 mL),
and Pd(PPh3)4 (0.63 g, 0.55 mmol). To this flask were then EtOH (6 × 5 mL), and again Et2O (3 × 5 mL). The remaining
added degassed toluene (200 mL) and a degassed 2 M aqueous black powder was dissolved in the minimum amount of DMF,
solution of Na2CO3 (50 mL). The mixture was heated under precipitated with Et2O and washed on a G4 sintered glass cru-
reflux and vigorously stirred for 48 h. After cooling to room cible with Et2O (5 × 10 mL), to afford complex DV42 as a black
temperature, the resulting mixture was filtered and the organic powder that was dried overnight under high vacuum (178 mg,
layer was separated, washed with H2O (3 × 50 mL), and evapo- 73%). 1H NMR (DMSO-d6, 500 MHz, due to the existence of
rated under reduced pressure. The obtained solid was sus- two diastereoisomeric pairs of enantiomers, each having
pended in methanol and filtered to afford a pale white solid. twenty six diastereotopic protons in the aromatic region,
Drying under vacuum resulted in the final product 3 (3.4 g, proton signals cannot be fully assigned and integrations are
1
80%). H NMR (CDCl3, 500 MHz): δ 8.71 (m, 2H), 8.59 (s, 1H), not given): δ 10.42 (d), 10.21 (d), 10.11 (s), 10.10 (s), 10.05 (d),
8.29 (d, 1H), 8.14 (s, 1H), 8.01 (d, 1H), 7.88 (t, 1H), 7.46–7.64 9.89 (d), 9.47 (d), 9.25–9.11 (m), 9.08 (s), 8.98–8.93 (m), 8.90
(m, 9H), 7.35 (t, 1H), 6.75 (dd, 1H), 5.81 (d, 1H), 5.27 (d, 1H).
(s), 8.72 (s), 8.67 (s), 8.61–8.59 (m), 8.38–8.13 (m), 8.00–7.51
Synthesis of 6-{4-[(E)-2-(9-anthryl)vinyl]phenyl}-4-phenyl-2- (m), 7.24 (d), 7.19–7.17 (m), 7.08–6.94 (m), 6.78 (d). 13C{1H}
pyridin-2-ylquinoline (LP1). In a two-necked round-bottomed NMR (DMSO-d6, 125 MHz): δ 165.83, 165.71, 165.60, 165.34,
flask equipped with a reflux condenser and a magnetic stirrer, 165.11, 162.33, 162.16, 161.50, 161.30, 160.61, −159.93, 159.27,
under an argon atmosphere, were added 3 (1.50 g, 3.90 mmol), 158.76, 158.06, 155.35, 154.56, 154.39, 153.84, 153.58, 153.09,
an excess of 9-bromoanthracene (1.51 g, 5.87 mmol), Pd(OAc)2 153.04, 152.10, 147.72, 146.89, 146.79, 137.02, 136.76, 136.22,
(18 mg, 71.3 μmol), and tris(3-methylphenyl)phosphine 135.88, 135.52, 132.11, 132.06, 130.93, 130.34, 129.46, 129.16,
(72 mg, 235 μmol). DMF (45 mL) along with Et3N (6 mL) were 128.92, 128.21, 128.15, 127.96, 127.92, 126.91, 126.69, 126.62,
then added and the system was carefully degassed again. The 126.42, 126.13, 125.93, 125.65, 125.11, 124.84, 124.62, 123.74,
reaction mixture was heated under reflux for 72 h and after 123.48, 122.61, 121.25, 120.38. FTIR vmax (cm−1) 1710.6 (m),
cooling to room temperature was poured into a mixture of 1598.7 (m), 1477.2 (m), 1403.9 (w), 1249.7 (s), 1211.1 (s),
crushed ice and water (500 mL). A yellowish solid started to 1184.1 (s), 1137.8 (m), 1062.6 (w), 931.5 (m), 885.2 (s), 827.3
form immediately. This mixture was filtered and repeatedly (s), 781.0 (s), 759.8 (s), 734.8 (s), 702.0 (m), 663.4 (m), 524.5
washed with water to remove any excess of solvents. The solid (w). HRMS (ESI−): calculated for C54H36Cl2N4O4Ru [M − H]−
crude product was vigorously stirred for 24 h in ethanol and 975.1088, observed 975.1068.
filtered again. Crude LP1, obtained after drying under reduced
pressure at 50 °C for 24 h, was redissolved in the minimum necked 50 mL round-bottomed flask, equipped with a stirring
amount of DMF and precipitated with H2O to afford ligand bar and reflux condenser, was charged with [RuCl-
Synthesis of Na-cis-[Ru(LP1)(Hdcbpy)(NCS)2] (DV51). A two-
a
LP1 as a yellow powder (1.31 g, 60%). 1H NMR (CDCl3, (p-cymene)]2 (76.5 mg, 0.125 mmol), LP1 (140.2 mg,
500 MHz): δ 8.88 (s, 1H), 8.80 (d, 1H), 8.65 (s, 1H), 8.50 (s, 1H), 0.250 mmol), and degassed DMF (35 mL) at room temperature
8.41–8.39 (m, 2H), 8.27 (s, 1H), 8.15 (d, 1H), 8.08–8.04 (m, 2H), under argon. The reaction mixture was heated in the dark at
8.00–7.98 (m, 2H), 7.81–7.76 (m, 5H), 7.73–7.71 (m, 2H), 60 °C for 4 hours. H2dcbpy (61.1 mg, 0.250 mmol) was then
7.64–7.58 (m, 3H), 7.54–7.50 (m, 4H), 7.46–7.44 (m, 1H), 7.03 added and the reaction mixture was heated in the dark under
(d, 1H). 13C{1H} NMR (CDCl3, 125 MHz): δ 156.36, 155.65, argon at 140 °C for 4 hours. NH4NCS (475.8 mg, 6.3 mmol)
149.41, 149.22, 148.05, 140.14, 138.92, 138.38, 136.99, 136.71, was added to the reaction mixture and heating at 140 °C con-
131.54, 130.81, 129.72, 128.88, 128.74, 128.64, 128.46, 127.84, tinued for 4 more hours. This solution was cooled to room
127.15, 126.59, 125.96, 125.56, 125.33, 125.22, 124.09, 123.48, temperature with a water bath, and DMF was removed on a
121.92, 119.78. FTIR vmax (cm−1) 1673.9 (w), 1587.1 (m), 1546.6 rotary evaporator. The resulting black solid was taken up and
(w), 1488.8 (m), 1469.5 (m), 1440.6 (m), 1361.5 (m), 1257.4 (w), washed on a G4 sintered glass crucible with H2O (5 × 10 mL),
1072.2 (w), 977.8 (w), 966.2 (w), 887.1 (m), 833.1 (m), 798.4 MeOH (10 × 1 mL), and Et2O (5 × 5 mL). The remaining solid
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6582–6591 | 6585

View Article Online
Paper
Dalton Transactions
was dissolved in the minimum amount of DMF, precipitated
with Et2O and washed on a G4 sintered glass crucible with
Et2O (5 × 5 mL), to afford complex DV44 as a black powder
that was dried overnight under high vacuum (186 mg, 73%).
Chromatographed DV51 was obtained from DV44 by the fol-
lowing procedure: 90 mg of DV44 were dissolved in a 0.02 M
NaOH solution in MeOH (5 mL) and MeOH (2.5 mL). This solu-
tion was filtered, loaded on a Sephadex LH-20 column (equili-
brated with MeOH), and purified with MeOH as an eluent. The
collected main band (deep red) was concentrated on a rotary
evaporator to about 7 mL and slowly titrated with a 0.02 M
HNO3 solution in MeOH to pH 5.2 (monitored by a pH meter).
The solvent was then removed on a rotary evaporator and the
remaining black solid was dissolved in DMF, precipitated with
Et2O, and washed on a G4 sintered glass crucible with Et2O
(5 × 5 mL), to afford purified DV51. Yield with three times
column purification: 30 mg (33%). 1H NMR (DMSO-d6,
500 MHz, due to the existence of two diastereoisomeric pairs
of enantiomers, each having twenty six diastereotopic protons
in the aromatic region, proton signals cannot be fully assigned
and integrations are not given): δ 9.64 (s), 9.51 (s), 9.41–9.38
(m), 9.30–9.17 (m), 9.08–8.99 (m), 8.81 (s), 8.75 (s), 8.57–8.55
(m), 8.44–8.11 (m), 8.02–7.54 (m), 7.34–7.31 (m), 7.02 (d), 6.94
(d). 13C{1H} NMR (DMSO-d6, 125 MHz): δ 165.55, 165.18,
163.14, 161.54, 159.69, 159.34, 158.13, 157.58, 156.76, 154.98,
154.33, 154.17, 153.44, 152.82, 150.08, 150.01, 149.20, 148.85,
140.27, 139.92, 139.49, 139.04, 138.24, 137.48, 137.40, 137.08,
137.03, 136.97, 136.42, 136.19, 135.74, 135.63, 134.89, 134.63,
134.26, 132.24, 131.06, 130.68, 130.64, 130.07, 129.52, 129.33,
129.05, 128.40, 128.34, 128.24, 128.18, 128.02, 127.04, 126.91,
126.56, 126.29, 126.06, 125.81, 125.27, 125.02, 124.82, 123.50,
122.78, 122.37, 120.79, 120.52, 119.47, 119.18. FTIR vmax
(cm−1) 2098.2 (s), 1976.7 (m), 1714.4 (m), 1600.6 (m), 1538.9
(m), 1475.3 (m), 1403.9 (m), 1375.0 (s), 1255.4 (m), 1230.4 (s),
1124.3 (m), 1020.2 (m), 981.6 (m), 896.7 (m), 808.0 (m), 781.0
(s), 759.8 (s), 734.8 (s), 702.0 (s), 601.7 (m), 524.5 (m). HRMS
(ESI−): calculated for C56H36N6O4RuS2 (DV44) [M − H]−
1021.1235, observed 1021.1255.
Scheme 1 Synthetic route followed for the preparation of the novel pyridine-
quinoline hybrid ligand LP1.
Scheme 2 Synthesis of Ru(II) photosensitizers DV42 and DV44.
Results and discussion
analyzed via UV-vis and 1H NMR spectroscopy. Purified and
Synthesis, NMR, and MS studies of ruthenium dyes
The synthetic approach followed for the preparation of the half-deprotonated DV51 (Fig. 2), free of S-coordinated NCS
bidentate pyridine-quinoline hybrid ligand LP1 is presented in linkage isomers, was obtained from DV44 in 33% isolated
Scheme 1. LP1 was synthesized by a Suzuki coupling reaction yield by Sephadex LH-20 column chromatography. DV42 and
of 6-bromo-4-phenyl-2-pyridin-2-ylquinoline (1) with (4-vinyl- DV51 were characterized by 1H, 13C, COSY, and HSQC NMR
phenyl)boronic acid (2), followed by a Heck coupling reaction spectroscopy. Additional characterization was performed
of the resulting 4-phenyl-2-pyridin-2-yl-6-(4-vinyl-phenyl)-qui- including UV-vis spectroscopy, FTIR spectroscopy, Raman
noline (3) with 9-bromoanthracene (4). LP1 was obtained in spectroscopy, and HRMS.
43% total isolated yield and characterized by 1H, 13C, COSY,
Due to the unsymmetrical nature of LP1, there are two possi-
and HSQC NMR spectroscopy, as well as with FTIR and ble diastereoisomeric pairs of enantiomers for DV42 and
Elemental Analysis. Complexes DV42 and DV44, both with cis DV51 (as well as for DV44). Indeed, the existence of two
geometry, were prepared by straightforward experimental pro- diastereoisomers for both DV42 and DV51 is clear in their
cedures (Scheme 2), both in 73% isolated yields. These reac- respective NMR spectra. For example, two well-resolved
tions were monitored by removing small aliquots, which were doublets at δ 7.02 and 6.94 ppm, for the major and the minor
6586 | Dalton Trans., 2013, 42, 6582–6591
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
1
isomers respectively, were recorded in the H NMR spectra of sensitized by DV51 present better absorbance than those sensi-
DV51. Integration of these two peaks, which correspond to the tized by DV42, while both show slightly lower absorbance than
vinylic protons next to the phenyl ring of coordinated LP1, Z907-sensitized films. The absorption edges of the sensitized
suggests a 62/38 ratio for the two diastereoisomeric pairs of films were found to be 715 and 725 nm, for DV51 and DV42
DV51. Although the existence of two diastereoisomers for respectively, by the tangential method (Fig. 4b). Thus, the
DV42 and DV51 results in fifty two diastereotopic aromatic absorption of both DV42- and DV51-sensitized films extends
protons for each, making the complete NMR peak assignment well into the red spectral range, shifted by 30–40 nm relative to
impossible, this does not impose any problem in their func- the absorption of the Z907 dye,11 in accordance with our antici-
tion as efficient photosensitizers in DSCs. Equally important, pation for a red-shift for DV42 and DV51 due to their extended
the molecular ions for DV42 and DV44 (the precursor of DV51) aromatic conjugation.
were detected by negative mode high resolution ESI at m/z
Vibrational (infrared, Raman, and resonance Raman)
spectroscopy
975.1068 and 1021.1255 respectively (requiring 975.1088 and
1021.1235) indisputably proving the structure of these novel
Ru(II) complexes.
The infrared spectra of DV42 and DV51 were recorded in the
4000–400 cm−1 region. These spectra are very rich in the
fingerprint region (1600–400 cm−1) due to the various atomic
vibrations of the two coordinated bidentate ligands LP1 and
H2dcbpy. Noteworthily, the presence of a strong absorption
band at 2098 cm−1 [v(CvN) stretching], along with the
absence of such a band at about 2080 cm−1, in the infrared
spectra of DV51, confirms solely N-coordination for the ambi-
dentate NCS ligands.7b,12
Resonant Raman spectra of dyes DV42 and DV51 are shown
in Fig. 5. The two strong bipyridine modes, ν(CvN) and
ν(CvC), appear at 1475 cm−1 and 1540 cm−1, at the same fre-
quencies observed for the Z907 dye,13 although reversed in
intensity. A new ν(CvC) mode is observed at 1603 cm−1 and
dominates over the 1613 ν(CvC) cm−1 one, which is resolved
as a satellite peak in the spectra, after the fitting analysis. This
last mode is present in Z907 spectra too. Furthermore, charac-
teristic Raman peaks of the quinoline hybrid ligand C–C
stretching modes14 at 1358 cm−1 and several C–H modes above
2600 cm−1 from the extended aromatic conjugation unit are
observed.
UV-visible studies
The UV-vis spectra of the dye solutions (Fig. 4a) show low fre-
quency MLCT transitions (λmax) at 588 nm for DV42 and
552 nm for DV51; this shift is mainly due to the solvatochro-
mic effect. Extinction coefficients (εm) equal to 9438 and
8896 M−1 cm−1 for DV42 and DV51, respectively, were calculated.
In comparison to the Z907 dye (Fig. 3), with λmax = 518 nm and
εm = 10 180 M−1 cm−1, the absorption of the new dyes is
slightly lower and shifts well to the red. The UV-vis spectra of
the sensitized TiO2 films show a maximum at about 550 nm
for both DV42 and DV51. These maxima are red-shifted by
30 nm in relation to the corresponding maximum of the films
sensitized with the structurally related Z907 dye (520 nm),
which was used as a reference. Even though the thickness of
the TiO2 films cannot be precisely controlled, the films
Upon grafting the dyes onto TiO2, the corresponding reso-
nant Raman spectra are enriched with the TiO2 anatase
modes at frequencies below 650 cm−1. The strongest mode
Fig. 4 Absorption spectra of dyes DV42 (DMF), DV51 (EtOH), and Z907 (EtOH)
in solution (a), and the corresponding spectra of the same dyes chemisorbed
onto transparent TiO2 films (b).
Fig. 5 Resonant Raman spectra of dyes DV42 and DV51 in the powder form
and resonant Raman spectra of dyes DV42, DV51, and Z907 after sensitization
of thick TiO2 films. Raman modes discussed in the text are marked in the figure.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6582–6591 | 6587

View Article Online
Paper
Dalton Transactions
observed is the E1g at 143 cm−1. The weak Raman peak of the
protonated carboxyl group at about 1720 cm−1 is even more
reduced in intensity, while in the case of DV51 it vanishes,
indicating chemisorption of the dye on the semiconducting
surface.7f,g,13 Moreover, weak shifts and changes in the relative
intensities of the C–H and C–C modes in the 1250–1320 cm−1
region are observed after fitting analysis, in agreement with
literature findings regarding related dyes.7g,13
Table 2 First oxidation potentials, optical bandgap, and excited-state oxidation
potentials of Ru(II) complexes DV42 and DV51
Sensitizer
Eox a (V vs. NHE)
Eg b (eV)
E*oxc (V vs. NHE)
DV42
DV51
+1.04
+1.14
+1.72
+1.90
−0.68
−0.76
a Electrochemical data. b Derived from the optical spectra of the dyes
in solution (Fig. 4b). c Calculated.
Contact angle studies
Contact angle measurements provide information on the
chemical nature of the sensitized photoelectrode surface and,
amongst others, can be used as a preliminary diagnostic
method for determining the efficient and optimum duration of
TiO2 dye sensitization.11,15 The surface of the non-sensitized
TiO2 photoelectrodes is very hydrophilic and rough. As a result,
the corresponding contact angle, measured for 10 μL water dro-
plets on the various films, is very low with θ = 11°. Following
sensitization, the films show higher contact angles (Table 1),
55° for the DV51 and 27° for the DV42 films, but remain
hydrophilic with values that are commonly observed upon sen-
sitization with the N3 dye.15,16 In this respect, DV51- and
DV42-sensitized films differ significantly from the Z907-
sensitized TiO2 films, which present strong hydrophobic char-
acter exhibiting a measured contact angle of θ = 125°; this
contact angle is in agreement with that reported in the litera-
ture.11 Therefore, the increased steric bulk of the novel pyri-
dine-quinoline polyaromatic ligand does not result in a
significantly enhanced hydrophobicity, as one would antici-
pate based on the ligand’s extended aromatic system.
corresponding ruthenium(II) ion. Both Eox values are analo-
gous to the previously measured Eox values for the benchmark
Z907 dye (+1.01 V vs. NHE).21 Given that neither DV42 nor
DV51 emit when photoexcited in the visible, their LUMO
energy levels could not be estimated from the intersection of
the corresponding superimposed UV-vis and emission spec-
tra.7g Alternatively, the dye excited-state oxidation potentials
(E*ox)22 were estimated at −0.68 and −0.76 V for DV42 and
DV51, respectively, by using the values of the absorption
threshold (Eg) determined from the optical spectra. E*ox values
of both dyes are more negative than the conduction band edge
of TiO2 (−0.5 V vs. NHE)23 permitting efficient electron injec-
tion from the excited dye molecules into the semiconductor
conduction band.
Photovoltaic studies
The photovoltaic performance of the sensitized TiO2 photo-
electrodes was examined following their incorporation into
sandwich-type liquid DSCs. The J–V characteristics were
obtained under 1 sun (AM 1.5 G) illumination (Fig. 6) and the
corresponding derived electric parameters are summarized in
Table 3. The overall power conversion efficiency (η) determined
for the DV42 dye-based cells was as high as 3.24%. DV51-
based cells showed 4.36% power conversion efficiency
Electrochemical studies
The cyclic voltammograms of both DV42 and DV51 dyes
present quasi-reversible waves with a large peak separation
between the anodic (Eap) and cathodic (Ecp) peak potentials
(ΔEp = Eap − Ecp ∼ 0.26–0.28 V), attributed to the oxidation of Ru(II)
to Ru(III). The first oxidation potentials (Eox) were estimated to
accompanied with a short-circuit current (Jsc) of 9.30 mA cm−2
,
−
be +1.04 and +1.14 V vs. NHE,17,18 lying well below the I−/I3
redox potential (+0.31 V vs. NHE)17 and allowing easy regener-
ation for both dyes. The 0.1 V difference in the oxidation
potential of the two dyes (Table 2) may be attributed to the
contribution of the NCS vs. the Cl ligands on the HOMO levels
of the respective complexes, as proposed in the literature for
several other ruthenium(II)-based dyes.19 Chloride ligands,
having an increased electron-donating nature in comparison
with thiocyanates,20 promote the easier oxidation of the
Table 1 Measured contact angles for 10 μL water droplets on DV42-, DV51-,
and Z907-sensitized TiO2 films
Surface
θ (°)
TiO2
11
27
55
2
3
3
7
Fig. 6 J–V characteristics of DSCs using optimum TiO2 electrodes (employing a
scattering layer) sensitized by DV42, DV51, and Z907 dyes under illumination
of 1 sun (AM 1.5 G).
TiO2-DV42
TiO2-DV51
TiO2-Z907
125
6588 | Dalton Trans., 2013, 42, 6582–6591
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Table 3 Performance parameters of DSCs based on titania films sensitized by
DV42, DV51, and Z907 dyes
Sensitizer
Jsc (mA cm−2
)
Voc (V)
FF
η (%)
DV42
DV51
Z907
7.85
9.30
14.15
0.63
0.68
0.81
0.66
0.69
0.68
3.24
4.36
7.82
an open-circuit potential (Voc) of 0.68 V, and a filling factor
(FF) of 0.69. Thus, all photovoltaic parameters are improved in
DV51-sensitized cells in comparison with the DV42-sensitized
cells.
In view of the fact that electron collection is similar for
photoelectrodes consisting of the same TiO2 type, the
Jsc increase in the DV51-based cells can be attributed to
improved light harvesting in comparison to the DV42-
sensitized cells. Consistent with this hypothesis, simple TiO2
films (without any scattering layer) sensitized by the DV51 dye
present a higher optical density (A = 0.88 at MLCT) when com-
pared with those sensitized by DV42 (A = 0.83 at the respective
MLCT, Fig. 4b). Besides improved light harvesting, the current
increase could also be realized because DV51 either enhances
electron injection (having an excited-state oxidation potential
80 mV higher than that of the DV42 dye)22 or is regenerated
more efficiently by the redox electrolyte. An improved DV51
regeneration could originate from the highly efficient electron
transfer from the iodine ions to the NCS ligands, which are
known to be excellent electron acceptors,24 despite the fact
that the driving force for DV51 is 100 mV smaller than that of
DV42 (see the first oxidation potentials in Table 2).
The 0.05 V Voc increase in the DV51-sensitized cells can be
attributed to a small shift of the TiO2 conduction band. Given
that DV42 leaves two protons on the TiO2 surface upon adsorp-
tion, in comparison to one proton for DV51, the semiconduc-
tor may be charged more positively in the case of DV42.
Suppressed recombination between electrons injected in the
semiconductor and triiodides, in the case of DV51-sensitized
cells, seems to be an even more possible reason for the
improved Voc. This suppressed recombination may originate
from the increased hydrophobicity of the DV51-sensitized
cells, when compared with their DV42 counterparts. The use
of hydrophobic dyes in order to increase Voc is a well-known
concept in DSCs literature;25 furthermore, Grätzel and co-
workers have observed a 50 mV difference between two dyes
that differ about 9° in the contact angles of the respective
films.11 In order to more accurately elucidate the observed
Fig. 7 IPCE spectra of DSCs based on DV42, DV51, and Z907 dyes.
cells in relation with those sensitized by the archetypal Z907
dye (Fig. 6) can be ascribed to the more hydrophobic character
of Z907, preventing electrons from recombining with triiodides
at the TiO2 surface.
To better understand the DSC performance of DV42 and
DV51, the corresponding IPCE (incident photon-to-current
efficiency) spectra were recorded (Fig. 7). Slight differences
were observed concerning the shape of the action spectra pro-
files for the Z907 and DV51 dyes, presenting IPCE values over
85% and close to 60% at about 530 nm, respectively. This is in
agreement with their Jsc values reported in Table 3. In contrast,
DV42-sensitized cells presented a significant red shift of about
30 nm, attaining a maximum IPCE value of 47% at about
560 nm. This shift is not consonant with the absorbance
characteristics of DV42 shown in Fig. 4a (note that the trans-
mittance/reflectance of the films was recorded on transparent
films); the shift can be attributed to the stronger interaction of
the DV42 dye with the scattering TiO2 layer employed in the
IPCE measurement. By integrating the product of the incident
photon flux density and the IPCE values over the 400 to
800 nm wavelength regime, the calculated short-circuit photo-
currents closely match the experimental Jsc values for all cells
(within a maximum deviation of 10%).
Conclusions
differences and correlate the photovoltaic performance with Two novel, heteroleptic ruthenium(II) sensitizers (DV42 and
the electron dynamics in DV42- vs. DV51-sensitized solar cells, DV51), bearing an unsymmetrical pyridine-quinoline hybrid
comprehensive studies are currently underway and will be ligand with extended π-conjugation, were synthesized, exten-
reported in due course.
sively characterized, and utilized in liquid electrolyte dye-sensi-
Finally, comparative experiments performed on identical tized solar cells. The absorption of both DV42- and DV51-
DSCs sensitized by the widely utilized standard Z907 dye, sensitized TiO2 films is red-shifted by 30–40 nm relative to the
which attains power conversion efficiencies over 10% in the lit- absorption of TiO2 films sensitized with the benchmark Z907
erature,26 show that the efficiency of the cells obtained from ruthenium(II) dye. This red shift is attributed to the extended
DV51 is about 56% of that attained by Z907 (Table 3). Again, aromatic conjugation of DV42 and DV51. Contact angle
the reduced Jsc and Voc values of DV42- and DV51-sensitized measurements of DV42- and DV51-sensitized TiO2 films
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6582–6591 | 6589

View Article Online
Paper
Dalton Transactions
suggest that these films are hydrophilic with contact angle
values that are commonly observed upon sensitization with
the standard N3 ruthenium(II) dye. The excited-state oxidation
potentials of DV42 and DV51 lie well above the TiO2 conduc-
tion band, permitting efficient electron injection from the
excited dye molecules into the semiconductor conduction
band. Moreover, both novel ruthenium(II) dyes have first oxi-
dation potentials that lie below the I−/I3− redox potential allow-
ing easy regeneration. Dye-sensitized solar cells integrating
DV42- or DV51-sensitized TiO2 photoelectrodes afford power
conversion efficiencies of 3.24 or 4.36% respectively. These
power conversion efficiencies are up to 56% of the correspond-
ing efficiencies attained by TiO2 photoelectrodes sensitized by
the standard Z907 ruthenium(II) dye under similar experimen-
tal conditions.
4 (a) M. K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-
Baker, E. Muller, P. Liska, N. Vlachopoulos and M. Grätzel,
J. Am. Chem. Soc., 1993, 115, 6382; (b) M. K. Nazeeruddin,
F. De Angelis, S. Fantacci, A. Selloni, G. Viscardi, P. Liska,
S. Ito, B. Takeru and M. Grätzel, J. Am. Chem. Soc., 2005,
127, 16835.
5 C. Y. Chen, S. J. Wu, C. G. Wu, J. G. Chen and K. C. Ho,
Angew. Chem., Int. Ed., 2006, 45, 5822.
6 M. Grätzel, DSC-IC 3, Nara, Japan, April 23–25, 2009.
7 (a) K. Chryssou, V. J. Catalano, R. Kurtaran and P. Falaras,
Inorg. Chim. Acta, 2002, 328, 204; (b) K. Chryssou,
T. Stergiopoulos and P. Falaras, Polyhedron, 2002, 21, 2773;
(c) T. Stergiopoulos, S. Karakostas and P. Falaras, J. Photo-
chem. Photobiol., A, 2004, 163, 331; (d) A. I. Philippopoulos,
A. Terzis, C. P. Raptopoulou, V. J. Catalano and P. Falaras,
Eur. J. Inorg. Chem., 2007, 5633; (e) J. Faiz,
A. I. Philippopoulos, A. G. Kontos, P. Falaras and
Z. Pikramenou, Adv. Funct. Mater., 2007, 17, 54;
(f) G. Konti, E. Chatzivasiloglou, V. Likodimos,
G. Kantonis, A. G. Kontos, A. I. Philippopoulos and
P. Falaras, Photochem. Photobiol. Sci., 2009, 8, 726;
(g) G. C. Vougioukalakis, T. Stergiopoulos, G. Kantonis,
A. G. Kontos, K. Papadopoulos, A. Stublla, P. G. Potvin and
P. Falaras, J. Photochem. Photobiol., A., 2010, 214, 22.
8 C. S. Karthikeyan, H. Wietasch and M. Thelakkat,
Adv. Mater., 2007, 19, 1091.
Acknowledgements
This publication is based on work supported in part by the
European Community’s Seventh Framework Programme (FP7/
2007–2013) under grant agreement no. 246124 (Sensitizer Acti-
vated Nanostructured Solar Cells – SANS) and by the Greek
Ministry of Development under grant agreement PENED
03ED118 (Organic Solar Cells). The Foundation for Education
and European Culture is acknowledged for providing a scholar-
ship to G.C.V. Prof. Joannis K. Kallitsis and Dr Andreas Stefo-
9 H. Snaith, Energy Environ. Sci., 2012, 5, 6513.
poulos are thankfully acknowledged for providing 4-phenyl-2- 10 S. P. Economopoulos, A. K. Andreopoulou, V. G. Gregoriou
pyridin-2-yl-6-(4-vinyl-phenyl)-quinoline. Mr Naoum Vaenas is and J. K. Kallitsis, Chem. Mater., 2005, 17, 1063.
acknowledged for carrying out a series of SEM measurements 11 M. Wang, S.-J. Moon, D. Zhou, F. Le Formal, N.-L. Cevey-
and adsorption–desorption experiments.
Ha, R. Humphry-Baker, C. Gratzel, P. Wang,
S. M. Zakeeruddin and M. Gratzel, Adv. Funct. Mater., 2010,
20, 1821.
12 P. Maruthamuthu and S. Anandan, Sol. Energy Mater. Sol.
Cells, 1999, 59, 199.
Notes and references
1 This type of cell was reported for the first time in 1991: 13 V. Likodimos, T. Stergiopoulos, P. Falaras, R. Harikisun,
B. O’Regan and M. Grätzel, Nature, 1991, 353, 737.
J. Desilvestro and G. Tulloch, J. Phys. Chem. C, 2009, 113,
2 (a) A. Hagfeldt and M. Grätzel, Chem. Rev., 1995, 95, 49;
9412.
(b) R. Argazzi, N. Y. M. Iha, H. Zabri, F. Odobel and 14 N. Puviarasan, V. Arjunan and S. Mohan, Turk. J. Chem.,
C. A. Bignozzi, Coord. Chem. Rev., 2004, 248, 1299; 2004, 28, 53.
(c) M. K. Nazeeruddin, S. M. Zakeeruddin, J.-J. Lagref, 15 S. B. Ambade, R. S. Mane, A. V. Ghule, M. G. Takwale,
P. Liska, P. Comte, C. Barolo, G. Viscardi, K. Schenk and
M. Grätzel, Coord. Chem. Rev., 2004, 248, 1317;
A. Abhyankar, B. W. Chod and S.-H. Han, Scr. Mater., 2009,
61, 12.
(d) A. S. Polo, M. K. Itokazu and N. Y. M. Iha, Coord. Chem. 16 E. Stathatos, P. Lianos, V. Jovanovski and B. Orel, J. Photo-
Rev., 2004, 248, 1343; (e) G. J. Meyer, Inorg. Chem., 2005, 44, chem. Photobiol., A, 2005, 169, 57.
6852; (f) N. Robertson, Angew. Chem., Int. Ed., 2006, 45, 17 J. Bisquert, ChemPhysChem, 2011, 12, 1633.
2338; (g) P. Xie and F. Guo, Curr. Org. Chem., 2007, 11, 18 K. R. Justin Thomas, P. Singh, A. Baheti, Y.-C. Hsu,
1272; (h) T. W. Hamann, R. A. Jensen, A. B. F. Martinson, K.-C. Ho and J. T’suen Lin, Dyes Pigm., 2011, 91, 33.
H. van Ryswyk and J. T. Hupp, Energy Environ. Sci., 2008, 1, 19 A. J. Hallett and J. E. Jones, Dalton Trans., 2011, 40, 3871.
66; (i) M. Grätzel, Acc. Chem. Res., 2009, 42, 1788; 20 M.
( j) G. C. Vougioukalakis, A. I. Philippopoulos,
T. Stergiopoulos and P. Falaras, Coord. Chem. Rev., 2011,
255, 2602.
K.
Nazeeruddin,
P.
Pechy,
T.
Renouard,
S. M. Zakeeruddin, R. Humphry-Baker, P. Comte, P. Liska,
L. Cevey, E. Costa, V. Shklover, L. Spiccia, G. B. Deacon,
C. A. Bignozzi and M. Grätzel, J. Am. Chem. Soc., 2001, 123,
1613.
3 K. Hara, T. Sato, R. Katoh, A. Furube, Y. Ohga, A. Shinpo,
S. Suga, K. Sayama, H. Sugihara and H. Arakawa, J. Phys. 21 F.-T. Kong, S.-Y. Dai and K.-J. Wang, Chin. J. Chem., 2007,
Chem. B, 2003, 107, 597.
25, 168.
6590 | Dalton Trans., 2013, 42, 6582–6591
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
22 A. Listorti, B. O’Regan and J. R. Durrant, Chem. Mater., 25 (a) A. Fornelli, M. Plannells, M. A. Sarmentero,
2011, 23, 3381.
E. Martinez-Ferreo, B. C. O’Regan, P. Ballester and
E. Palomares, J. Mater. Chem., 2008, 18, 1652;
(b) H. Choi, C. Baik, S. O. Kang, J. Ko, M.-S. Kang,
M. K. Nazeeruddin and M. Grätzel, Angew. Chem., Int.
Ed., 2008, 47, 327.
23 S. Haid, M. Marszalek, A. Mishra, M. Wielopolski,
J. Teuscher, J. E. Moser, R. Humphry-Baker,
S. M. Zakeeruddin, M. Grätzel and P. Bäuerle, Adv. Funct.
Mater., 2012, 22, 1291.
24 C.-H. Hu, A. M. Asaduzzaman and G. Schreckenbach, 26 L.-L. Li, C.-W. Chang, H.-H. Wu, J.-W. Shiu, P.-T. Wu and
J. Phys. Chem. C, 2010, 114, 15165.
E. W.-G. Diau, J. Mater. Chem., 2012, 22, 6267.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 6582–6591 | 6591