Novel ruthenium sensitizers with a dianionic tridentate ligand for dye-sensitized solar cells: The relationship between the solar cell performances and the electron-withdrawing ability of substituents on the ligand

Article abstract of DOI:10.1039/c3dt52873a

Five novel ruthenium sensitizers (TUS sensitizers) with a dianionic tridentate ligand (pyridine-2,6-dicarboxyamidato and its derivatives) have been synthesized for application to dye-sensitized solar cells (DSCs). These TUS sensitizers have much larger molar absorption coefficients in the wavelength range below 600 nm compared with those of Black dye which is a structural analog and a highly efficient ruthenium sensitizer. The energy levels of HOMOs and LUMOs of TUS sensitizers shifted to the positive direction with increasing the electron-withdrawing ability of the substituents on the dianionic tridentate ligand. The energy levels of HOMO and LUMO showed linear correlation with respect to the Hammett constant of the substituents. The DSCs with TUS sensitizers showed much lower performances than that of Black dye. Both inferior adsorptivity on the TiO₂ surface and unfavorable energy levels of HOMOs and LUMOs for the effective electron transfer reactions in the DSCs are considered to be the main reasons for the much lower performances of TUS sensitizers. The conversion efficiency of the DSC with a TUS sensitizer increased with increasing the electron-withdrawing ability of the substituents on the dianionic tridentate ligand. The observed linear relationship between the conversion efficiency and the driving force of the reduction process of the oxidized form of dyes by I^- suggests that the dye regeneration process is a rate-determining step in the DSCs with TUS sensitizers. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c3dt52873a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Novel ruthenium sensitizers with a dianionic
tridentate ligand for dye-sensitized solar cells:
the relationship between the solar cell
performances and the electron-withdrawing
ability of substituents on the ligand†
Cite this: Dalton Trans., 2014, 43,
8026
Hironobu Ozawa, Shunsuke Honda, Daichi Katano, Takahito Sugiura and
Hironori Arakawa*
Five novel ruthenium sensitizers (TUS sensitizers) with a dianionic tridentate ligand (pyridine-2,6-dicarb-
oxyamidato and its derivatives) have been synthesized for application to dye-sensitized solar cells (DSCs).
These TUS sensitizers have much larger molar absorption coefficients in the wavelength range below
600 nm compared with those of Black dye which is a structural analog and a highly efficient ruthenium
sensitizer. The energy levels of HOMOs and LUMOs of TUS sensitizers shifted to the positive direction
with increasing the electron-withdrawing ability of the substituents on the dianionic tridentate ligand. The
energy levels of HOMO and LUMO showed linear correlation with respect to the Hammett constant of
the substituents. The DSCs with TUS sensitizers showed much lower performances than that of Black dye.
Both inferior adsorptivity on the TiO2 surface and unfavorable energy levels of HOMOs and LUMOs for
the effective electron transfer reactions in the DSCs are considered to be the main reasons for the much
lower performances of TUS sensitizers. The conversion efficiency of the DSC with a TUS sensitizer
increased with increasing the electron-withdrawing ability of the substituents on the dianionic tridentate
ligand. The observed linear relationship between the conversion efficiency and the driving force of
the reduction process of the oxidized form of dyes by I− suggests that the dye regeneration process is a
rate-determining step in the DSCs with TUS sensitizers.
Received 12th October 2013,
Accepted 4th March 2014
DOI: 10.1039/c3dt52873a
www.rsc.org/dalton
tcterpy = 4,4′,4″-tricarboxy-2,2′:6′,2″-terpyridine), Fig. 1} is one
of the most famous and highly efficient ruthenium sensitizers,
Introduction
In the last two decades, significant attention has been focused which has a broad absorption band extending into the near IR
to the development of dye-sensitized solar cells (DSCs) due to region and exhibits more than 11% conversion efficiency.4 For
their relatively higher light-to-electrical energy conversion further improvement of the conversion efficiency of DSCs,
efficiency with lower production cost.1 The conversion development of extremely efficient ruthenium sensitizers
efficiency of DSCs recently exceeded 12% by employing highly having larger molar absorption coefficients in the near IR
efficient sensitizers.2 Number of sensitizers, especially ruthe- region as well as in the visible region is required because the
nium-complex sensitizers, have thus far been synthesized to near IR light is not utilized effectively even for highly efficient
achieve high-performance DSCs because the photo- and elec- ruthenium sensitizers such as Black dye.
trochemical properties of dyes are reflected directly in the con-
On the other hand, long-term stability is also an important
version efficiency of the DSCs.3 Among them, Black dye requirement for the practical application of DSCs. Some
{(TBA)3[Ru(Htcterpy)(NCS)3] (TBA
=
tetrabutylammonium, possibilities which decrease the conversion efficiency of DSCs
have been thus far pointed out. For example, it is reported
that isothiocyanate (NCS) ligands, which are often employed
in efficient ruthenium sensitizers as ancillary ligands to
Department of Industrial Chemistry, Faculty of Engineering, Tokyo University of Science,
adjust the energy level of HOMO, undergo gradually ligand-
substitution and isomerization reactions under the irradia-
tion conditions.5 In addition, it is also reported that
NCS ligands undergo sulfur atom release, resulting in the
12-1, Ichigaya-Funagawara, Shinjuku, Tokyo 162-0826, Japan.
E-mail: h.arakawa@ci.kagu.tus.ac.jp; Fax: (+81) 3 5261 4631; Tel: (+81) 3 5228 831
†Electronic supplementary information (ESI) available. See DOI:
10.1039/c3dt52873a
8026 | Dalton Trans., 2014, 43, 8026–8036
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 1 Structures of TUS sensitizers and Black dye.
formation of a cyano (CN) complex under the irradiation con- of the substituents, or to know the suitable energy level of
ditions.6 Since promotion of such reactions decreases the con- HOMO of the ruthenium sensitizer for the effective reduction
version efficiency of DSCs, NCS ligands are considered to be a by I−.8j,11 In the case of pyridine-2,6-dicarboxyamidato deriva-
weak point for ruthenium sensitizers to achieve enough stabi- tive ligands (the ligands of TUS sensitizers), the donor ability
lity for the practical application of DSCs. Moreover, synthetic of these ligands is also considered to be strong, which would
yields of ruthenium sensitizers having multiple NCS ligands greatly increase the energy levels of HOMOs of TUS sensitizers.
are generally lower because structural isomers (S-bonded Actually, the energy level of HOMO of Ru(terpy)(L) (L = N,N′-bis-
compounds) must be formed simultaneously as byproducts of (4-phenyl)pyridine-2,6-dicarboxyamidato), which is a structural
the target N-bonded compound.7 Since S-bonded compounds analog of TUS sensitizers, was reported to be 0.2 V vs. SCE.12
show inferior solar cell performances compared with N- This value is much higher for the effective reduction by I−;
bonded ones, multiple times of separations of S-bonded therefore, a substituent with a relatively strong electron-with-
isomers must be required to obtain higher conversion drawing ability is considered to be crucial to adjust the donor
efficiency. These procedures might raise the production cost of ability of the pyridine-2,6-dicarboxyamidato derivative ligand,
DSCs; therefore, it is better to replace NCS ligands with other and the energy levels of HOMOs of TUS sensitizers. Here we
ligands which provide enough stability, and a higher synthetic report the syntheses, and photo- and electrochemical properties
yield. In this context, various kinds of bidentate and tridentate of five novel TUS sensitizers with different substituents on the
ligands, which can bind to the ruthenium atom stronger than dianionic tridentate ligand. In addition, the effects of the elec-
monodentate NCS ligands, have been explored to solve the tron-withdrawing ability of substituents on the phenyl rings of
above mentioned problems, and several ruthenium sensitizers the dianionic tridentate ligands on the solar cell performances
having such a multidentate ligand instead of NCS ligands have of DSCs have been investigated. This study revealed that the
been reported.8 We have also reported such ruthenium sensi- conversion efficiency of the DSCs with these TUS sensitizers
tizers having a β-diketonato derivative ligand instead of two improves with increasing electron-withdrawing ability of the
NCS ligands.9
substituents on the phenyl rings of the dianionic tridentate
In this study, five novel ruthenium complexes (TUS-29– ligands.
TUS-33, Fig. 1) having a dianionic tridentate ligand (pyridine-
2,6-dicarboxyamidato and its derivatives) instead of three NCS
ligands have been synthesized. It is expected that synthetic
Results and discussion
Syntheses
yields of these ruthenium sensitizers are improved largely
because structural isomers do not exist. It is noted here that a
ruthenium sensitizer with a dianionic tridentate ligand is rela- Three novel tridentate ligands were synthesized by the reaction
tively rare10 compared to those with a monoanionic tridentate of pyridine-2,6-dicarbonyl chloride and two equivalents of
ligand. Generally, the donor ability of an anionic multidentate 4-substituted aniline (Fig. 2). In 1H NMR spectra of these triden-
ligand is stronger than that of an NCS ligand. Therefore, one tate ligands in DMSO, characteristic signals corresponding to
of the most important points in the utilization of an anionic the amide protons (–CONH–) were observed in the chemical
multidentate ligand, instead of NCS ligands, is turning off the shift range between 10.8 and 11.5 ppm. TUS sensitizers were
donor ability of the anionic multidentate ligand, which affects synthesized by the reaction of Ru(tcterpy)Cl3 and the pyridine-
directly the energy level of HOMO of the ruthenium sensitizer, 2,6-dicarboxyamide derivative in the presence of N(C2H5)3 in
by introducing a substituent with a relatively strong electron- DMF. In this reaction, a small amount of TBAOH was added to
withdrawing ability. In this regard, systematic studies on the dissolve Ru(tcterpy)Cl3. The crude product was purified on a
precise control of the donor ability of an anionic multidentate Sephadex LH-20 column using a mixed solvent of C2H5OH and
ligand by changing the substituents have been carried out to H2O (1 : 1, v/v) containing 0.05 M TBAOH as an eluent. Novel
know the suitable strength of the electron-withdrawing ability TUS sensitizers (TUS-29–TUS-33, except for TUS-31) could be
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8026–8036 | 8027

View Article Online
Paper
Dalton Transactions
Fig. 2 Synthetic schemes of TUS sensitizers.
obtained by a simple procedure with relatively higher synthetic
yields. All TUS sensitizers were successfully characterized by
1H NMR spectroscopy and elemental analysis. In addition, the
monovalent cation (proton adduct) of each neutral complex
was successfully observed by adding a small amount of acetic
1
acid as a proton source in ESI-TOF MS measurements. In H
NMR spectra of TUS sensitizers in DMSO, characteristic
signals corresponding to the amide protons of the tridentate
ligands disappeared, clearly indicating that pyridine-2,6-dicar-
boxyamide derivatives bind to the ruthenium(II) atom, not O
atoms but N atoms, via dissociation of amide protons. Such a
binding mode of the pyridine-2,6-dicarboxyamide derivative to
the ruthenium(II) atom was confirmed previously by X-ray crys-
tallographic analysis for structural analogues.12 In the case of
TUS-29, information on the binding mode of pyridine-2,6-
dicarboxyamide could not be obtained from the 1H NMR spec-
Fig. 3 Molar absorptivity spectra of TUS sensitizers and Black dye in
trum due to its relatively lower solubility (the 1H NMR spec- DMSO.
trum was obtained using a basic mixed solution of CD3OD
and NaOD). However, pyridine-2,6-dicarboxyamide seems to
serve as a dianionic tridentate ligand as its other derivatives
because TUS-29 showed similar photo- and electrochemical
properties to the other TUS sensitizers (vide infra).
(dianionic tridentate ligand) display weak and broad absorp-
tions in the wavelength region between 600 and 800 nm. From
the TD-DFT calculations, these absorptions are assigned to
MLCT transitions with shoulder absorptions of spin-forbidden
3MLCT transitions in the longer wavelength range.10b From the
Photo- and electrochemical studies
Molar absorptivity spectra of TUS sensitizers and Black dye in structural similarity between these ruthenium sensitizers and
DMSO are shown in Fig. 3. TUS sensitizers displayed relatively TUS ones, the observed broad absorptions for TUS sensitizers
strong two absorptions in the visible region at around 420 and would be assignable to MLCT transitions with shoulder
520 nm, assignable to the mixture of MLCT and LLCT tran- absorptions of spin-forbidden 3MLCT transitions. The onset
sitions, and the MLCT transition, respectively. Molar absorp- wavelengths of the absorption spectra of TUS-29 and TUS-31
tion coefficients of these two absorptions of TUS sensitizers exceeded 1000 nm. Absorption onsets of TUS sensitizers
were about two times larger than those of Black dye although (TUS-30–TUS-33) shifted to a shorter wavelength region with
molar absorption coefficients of TUS sensitizers in the wave- increasing the electron-withdrawing ability of the substituents
length region between 600 and 700 nm were smaller than on the dianionic tridentate ligand. These results suggest that
those of Black dye. The most interesting feature of the absorp- the electron-withdrawing ability of the substituents on the di-
tion spectra of TUS sensitizers is broad absorptions in the anionic tridentate ligand affects the energy level of HOMO
longer wavelength range extending into the near IR region. and/or LUMO of TUS sensitizers.
Although molar absorption coefficients of these bands were
The electrochemical properties of TUS sensitizers were
not so large (ca. 2500 cm−1 M−1), it is expected that DSCs with investigated by cyclic voltammetry in DMSO. All TUS sensi-
TUS sensitizers would show higher incident photon-to-current tizers displayed quasi-reversible oxidation waves in the poten-
conversion efficiency (IPCE) in this wavelength region than tial range between 0.3 and 0.7 V vs. SCE (Table 1). These
that with Black dye. It is recently reported that ruthenium sen- oxidation waves are assigned to oxidations of the Ru(II)/Ru(III)
sitizers with tcterpy and pyridine-2,6-dipyrazolato derivatives couples. As summarized in Table 1, the oxidation potentials of
8028 | Dalton Trans., 2014, 43, 8026–8036
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Table 1 Electrochemical properties of TUS sensitizers and Black dyea
Ru(III) couple and the amidato-ligand-centered oxidation,
respectively.12 From the structural similarity between Ru(terpy)(L)
and TUS sensitizers, the second quasi-reversible oxidation
observed for TUS-31 seems to be the amidato-ligand-centered
oxidation. It is also reported that these two oxidation poten-
tials shift in the negative direction with increasing the elec-
tron-donating ability of the substituents on the phenyl units.12
Since TUS-31 has quite strong electron-donating substituents
(–NMe2) on the phenyl units, the oxidation potential of the
amidato-ligand-centered oxidation seems to shift largely in the
negative direction compared with the other TUS sensitizers.
Therefore, only TUS-31 displayed the second oxidation in the
potential range below 1.0 V vs. SCE.
EHOMO
[V vs. SCE]
ELUMO
[V vs. SCE]
Sensitizer
E0–0 b [V]
TUS-29
TUS-30
TUS-31
TUS-32
TUS-33
Black dyec
0.30
0.40
0.30
0.48
0.67
0.66
1.23
1.29
1.24
1.33
1.43
1.61
−0.93
−0.89
−0.94
−0.85
−0.76
−0.95
a All potentials are given vs. SCE. Oxidation potentials of TUS
sensitizers were measured in DMSO containing 0.1 M TBAClO4.
b E0–0 was estimated from the onset wavelength of the absorption
spectrum. c Data from ref. 13.
Phosphorescence from the excited state of each TUS sensi-
tizer in DMSO was not observed in the measurable wavelength
range of our fluorophotometer (λ ≤ 1010 nm); therefore, the
lowest transition energy (E0–0) was estimated from the onset
wavelength of the absorption spectrum (Table 1). E0–0
increased with increasing the electron-withdrawing ability of
the substituents on the dianionic tridentate ligand because
the absorption onset shifted gradually to the shorter wave-
length region as mentioned above. Consequently, a linear
relationship between the energy levels of LUMOs of TUS sensi-
tizers (TUS-30–TUS-33) and the Hammett constants was also
observed (Fig. S1†).
TUS sensitizers (TUS-30–TUS-33) shifted in the positive direc-
tion with increasing the electron-withdrawing ability of the
substituents on the dianionic tridentate ligand. A linear
relationship between oxidation potentials of TUS sensitizers
and Hammett constants (σp; N(CH3)2 = −0.83, H = 0, CF3
=
0.54; σm; F = 0.34)14 of the substituents on the dianionic tri-
dentate ligand was observed (Fig. 4), clearly indicating that the
energy levels of HOMOs of TUS sensitizers can be tuned by
changing the electron-withdrawing ability of the substituents
on the phenyl rings of the pyridine-2,6-diamidato ligand. Only
in the case of TUS-31, the second oxidation (quasi-reversible)
and the third one (irreversible) were observed at 0.5 and 0.7 V
vs. SCE, respectively (Fig. S2†). Irreversible oxidation was also
observed at around 0.7 V vs. SCE for L-NMe2 (the ligand of
TUS-31). In addition, the oxidation potential of N,N-dimethyl-
aniline was reported at around 0.8 V vs. SCE.15 Therefore,
the third irreversible oxidation observed for TUS-31 must be
assigned to the oxidation of N,N-dimethylaniline units of the
L-NMe2 ligand. On the other hand, it is reported that Ru(terpy)
(L) (L = N,N′-bis(4-methoxyphenyl)pyridine-2,6-dicarboxyami-
dato) displays two quasi-reversible oxidations at 0.16 and 0.94 V
vs. SCE, and these two oxidations are assigned to the Ru(II)/
On the other hand, oxidation potentials of Ru(II)/Ru(III)
couples of TUS sensitizers (HOMOs) are relatively close to that
−
of the I−/I3 redox couple (ca. 0.2 V vs. SCE) except for TUS-33.
Therefore, effective reductions of the oxidized forms of TUS
sensitizers by I− are not expected in the DSCs with TUS sensi-
tizers (TUS-29–TUS-32). In these cases, dye regeneration by I−
might be a rate-determining step in the DSCs. The energy
levels of LUMOs of TUS-29–TUS-31 were high enough for the
effective electron injection into the conduction band of TiO2,
while those of TUS-32 and TUS-33 were found to be relatively
close to the conduction band energy of TiO2 (ca. 0.7 V vs. SCE).
In the case of TUS-32 and TUS-33, electron injection from the
excited state of dyes into the conduction band of TiO2 would
be a rate-determining step in the DSCs. Further discussions
will be carried out in the following section.
DFT calculations
Frontier molecular orbitals of TUS sensitizers are shown in
Fig. 5. Higher energy HOMOs (HOMO and HOMO−1) of TUS
sensitizers, except for TUS-31, are localized mainly in the Ru(II)
atom and two amidato N atoms of the dianionic tridentate
ligand (Fig. S4–S8†). These results explain reasonably the fact
that the energy levels of HOMOs of TUS sensitizers are affected
strongly by the electron-withdrawing ability of the substituents
on the phenyl rings of the dianionic tridentate ligand as men-
tioned above. In the case of TUS-31, HOMO is localized in N,N-
dimethylamino units of the dianionic tridentate ligand.
HOMO−1 and HOMO−2 are also localized in N,N-dimethyl-
amino units, and are populated partly in the Ru(II) atom and
two amidato N atoms. HOMO−3 is populated largely in the
Ru(II) atom and two amidato N atoms (Fig. S6†). The energy
Fig. 4 Relationship between the energy levels of HOMOs (EHOMO) of
TUS sensitizers and the Hammett constants of the substituents (2σ) on
dianionic tridentate ligands.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8026–8036 | 8029

View Article Online
Paper
Dalton Transactions
Fig. 5 Frontier molecular orbitals (HOMO and LUMO) of fully optimized structures of TUS sensitizers in acetonitrile.
levels of HOMO and HOMO−1 are much higher than that of showed much lower conversion efficiency than that of Black
HOMO−2 (Fig. S3†); however, the first oxidation of TUS-31 is dye. These poor performances are attributed mainly to the
assigned to the Ru(II)/Ru(III) couple. Therefore, HOMO and quite smaller Jsc values. IPCE values of the DSCs with all TUS
HOMO−1 do not participate in the redox reaction at 0.3 V vs. sensitizers were quite lower in the whole visible region com-
SCE even though they have higher energy. It seems probable pared with that of Black dye (Fig. 6), even though the molar
that relatively larger overpotential is required to oxidize the absorption coefficients of all TUS sensitizers are much larger
N,N-dimethylaminophenyl units because the observed oxidation in the wavelength range below 600 nm. In addition, the near
wave (the third oxidation wave) is irreversible. The validity of IR light was not converted into the photocurrent at all in the
this interpretation is supported by the fact that N,N-dimethyl- DSCs with TUS sensitizers although they have characteristic
aniline displays an irreversible oxidation wave, and forms broad absorptions in this wavelength range. These much lower
several kinds of oxidized products via relatively complicated IPCE values of the DSCs with TUS sensitizers would be due to
processes.15a On the other hand, lower energy LUMOs (LUMO, the unfavorable energy levels of HOMOs and/or LUMOs as
LUMO+1, and LUMO+2) of TUS sensitizers are delocalized described above. One of the other possible interpretations for
mainly in the whole tcterpy ligand with some population in the quite poor performances of TUS sensitizers is that the
the Ru(II) atom and two amidato N atoms.
amount of dye adsorption in all TUS sensitizers is much
smaller than that of Black dye (Table 2). In general, the Jsc
value and the conversion efficiency increase with increasing
Solar cell performances of the DSCs
Solar cell performances of the DSCs with TUS sensitizers or
Black dye are summarized in Table 2. All TUS sensitizers
Table 2 Solar cell performances of DSCs with TUS sensitizers or Black
dye using a standard electrolytea
The amount of
Jsc
dye adsorption
Sensitizer
[mA cm−2
]
Voc [V]
FF
η [%]
[×10−7 mol cm−2
]
TUS-29
TUS-30
TUS-31
TUS-32
TUS-33
Black dye
6.30
5.39
3.34
8.81
7.24
21.75
0.59
0.51
0.47
0.53
0.58
0.67
0.69
0.69
0.68
0.67
0.72
0.71
2.55
1.88
1.07
3.12
3.03
10.45
0.6
0.8
0.8
0.6
0.5
2.1
a The electrolyte was an acetonitrile solution containing 0.05 M I2,
0.1 M LiI, 0.6 M DMPImI, and 0.3 M TBP (TiO2 film thickness: ca.
32 μm, active area: 0.25 cm2). Irradiation was carried out using a solar
simulator (AM 1.5, 100 mW cm−2).
Fig. 6 IPCE spectra of the DSCs with TUS sensitizers or Black dye using
a standard electrolyte.
8030 | Dalton Trans., 2014, 43, 8026–8036
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
the amount of dye adsorption until the TiO2 surface is covered
fully by the adsorbed dyes. Consequently, the dependence of
the solar cell performances of the DSCs with TUS sensitizers
on the amount of dye adsorption has been investigated. In the
case of TUS-29, the amount of dye adsorption increased with
prolonging the dye-loading time, and finally it reached 1.6 ×
10−7 mol cm−2 after 45 h immersion. However, this value is
still smaller than that of Black dye, indicating that the adsorp-
tivity of TUS sensitizers on the TiO2 surface is inferior to that
of Black dye. In this case, the Jsc value and the conversion
efficiency decreased gradually with increasing the amount of
dye adsorption against the expectation (data not shown). The
reason for the smaller amounts of dye adsorption is still
unclear.
Table 3 Solar cell performances of DSCs with TUS sensitizers or Black
dye using a TBP-free electrolytea
The amount of
Jsc
dye adsorption
Sensitizer
[mA cm−2
]
Voc [V]
FF
η [%]
[×10−7 mol cm−2
]
TUS-29
TUS-30
TUS-31
TUS-32
TUS-33
Black dye
9.06
3.63
2.73
5.07
10.36
24.60
0.47
0.42
0.36
0.47
0.45
0.50
0.68
0.64
0.59
0.63
0.64
0.60
2.85
0.98
0.58
1.51
3.01
7.44
0.5
0.7
0.6
0.6
0.6
2.0
a The electrolyte was an acetonitrile solution containing 0.05 M I2,
0.1 M LiI, and 0.6 M DMPImI (TiO2 film thickness: ca. 32 μm, active
area: 0.26 cm2). Irradiation was carried out using a solar simulator
(AM 1.5, 100 mW cm−2).
On the other hand, the conversion efficiency of the DSC
with a TUS sensitizer increased with increasing the electron-
withdrawing ability of the substituents on the dianionic triden-
tate ligand, although it was saturated in the case of TUS-33. As
Solar cell performances of the DSCs with TUS sensitizers
or Black dye using TBP-free electrolytes are summarized in
shown in Fig. 7, the energy gap between the LUMO and the Table 3. Jsc values of the DSCs with TUS-29 and TUS-33 were
conduction band of TiO2 (ΔG1) decreased with increasing the
improved by using TBP-free electrolytes; however, those of
electron-withdrawing ability of the substituents. While, the
TUS-30, TUS-31 and TUS-32 decreased against the expectation.
energy gap between the HOMO and the redox potential of the The Voc value of each DSC was decreased by removing TBP,
I−/I3− couple (ΔG2) increased with increasing the electron-with-
which was caused mainly by the positive shift of the conduc-
drawing ability of the substituents. Therefore, the rate-deter-
tion band energy of TiO2. As a result, the conversion efficiency
mining step in the DSC with a TUS sensitizer (TUS-30, TUS-31 of the DSCs with TUS sensitizers was decreased by using TBP-
and TUS-32) under this condition seems to be the reduction
free electrolytes in the case of TUS-30, TUS-31 and TUS-32,
while that of TUS-29 and TUS-33 was unchanged. IPCE values
process of the oxidized form of the TUS sensitizer by I−. In the
case of TUS-33, the rate-determining step is considered to be increased only in the case of the DSCs with TUS-29 and
the electron injection process from the excited state of dyes to
the conduction band of TiO2. In order to ensure these con-
TUS-33 (Fig. 8), even though the driving force of the electron
injection process (or ΔG1) was enlarged in each case. These
siderations and to improve the conversion efficiency of the results support the above mentioned considerations that ΔG1
DSCs with TUS sensitizers, further investigations were carried
out using TBP-free electrolytes. It is noted here that TBP
is not sufficient in the DSC with TUS-33, and that ΔG1 is
sufficient in the DSCs with TUS-30, TUS-31 and TUS-32 when
(TBP = 4-tert-butylpyridine) is quite often used as one of the the standard electrolytes were used. On the other hand, the
components of the electrolyte of DSCs, and is known to raise
the conduction band energy of TiO2. Therefore, the conduction
decrease of Jsc value and conversion efficiency of the DSCs
with TUS-30, TUS-31 and TUS-32 suggests that there is another
band energy of TiO2 can be lowered by removing the TBP from reason for the poor performances of these TUS sensitizers
the electrolyte.16
besides both the unfavorable energy levels of HOMOs and the
Fig. 7 HOMO–LUMO energy diagrams of TUS sensitizers and Black dye.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8026–8036 | 8031

View Article Online
Paper
Dalton Transactions
Fig. 8 IPCE spectra of the DSCs with TUS sensitizers or Black dye using
a TBP-free electrolyte.
Fig. 10 Relationship between the conversion efficiency of the DSC
with a TUS sensitizer and ΔG2.
therefore, the Jsc value and conversion efficiency were
decreased by removing TBP. For comparison, effective electron
injection occurred by removing TBP in the case of TUS-33.
Therefore, the Jsc value was improved, and the conversion
efficiency was not decreased by using a TBP-free electrolyte in
the case of TUS-33. However, the enhanced backward electron
−
transfer reaction from the conduction band of TiO2 to the I3
would be an additional reason for the inferior performances of
all TUS sensitizers.
Among five TUS sensitizers, TUS-29 showed relatively
higher conversion efficiency compared with the other four TUS
sensitizers. In addition, the shape of the IPCE spectrum
obtained from the DSC with TUS-29 was different from those
Fig. 9 Electron lifetimes as a function of Voc of the DSCs with TUS sen- of the other TUS sensitizers. Especially, IPCE values of TUS-29
sitizers and Black dye.
in the wavelength range between 450 and 750 nm were rela-
tively larger than those of TUS-31, which has almost the same
HOMO–LUMO energy levels and absorption coefficients in this
much smaller amounts of dye adsorption as mentioned above. wavelength range. Since only TUS-29 does not have phenyl
As shown in Fig. 9, the electron lifetime in the TiO2 photoelec- rings on the dianionic tridentate ligand, this structural differ-
trode of the DSC with each TUS sensitizer was found to be ence seems to be one reason for the relatively higher perform-
shorter than that of Black dye, suggesting that the backward ance of TUS-29. For example, the access of I− to the Ru center
electron transfer reactions from the conduction band of TiO2 of TUS-29 might occur more easily due to the absence of bulky
−
to the oxidized form of the dye or to I3 in the electrolyte were phenyl rings. In such a case, reduction of the oxidized form of
promoted. In the case of TUS sensitizers, the amounts of dye TUS-29 seems to proceed more effectively, which increases the
adsorption were much smaller than that of Black dye (Table 3), conversion efficiency of the DSC. On the other hand, the con-
−
which promotes the access of I3 to the TiO2 surface, resulting version efficiency of the DSC with TUS sensitizers (TUS-30–
in the enhancement of the backward electron transfer reac- TUS-33) increased with increasing the electron-withdrawing
tion. Therefore, the dominant pathway of the backward elec- ability of the substituents on the dianionic tridentate ligand.
−
tron transfer reactions seems to be electron transfer to I3 in With increasing the electron-withdrawing ability of the substi-
the electrolyte in the case of TUS sensitizers. By removing the tuents, ΔG1 decreased gradually while ΔG2 increased largely
−
TBP from the electrolyte, the approach of I3 to the TiO2 (Fig. 7). As shown in Fig. 10, a linear relationship between the
surface might be further enhanced because TBP adsorbed conversion efficiency and the ΔG2 was observed in the DSCs
onto the TiO2 surface uncovered by dyes. The backward elec- with TUS sensitizers. Therefore, the rate-determining step
−
tron transfer reaction from the conduction band of TiO2 to I3
in the DSC with a TUS sensitizer under this condition is
might be further enhanced by removing TBP from the electro- considered to be not the electron injection process but
lyte. In addition, the rate-determining step is not an electron the reduction process of the oxidized form of a TUS sensitizer
injection process in the case of TUS-30, TUS-31 and TUS-32; by I−.
8032 | Dalton Trans., 2014, 43, 8026–8036
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
wire counter electrode, and a Ag wire reference electrode.
The supporting electrolyte was 0.1 M LiClO4. Potentials were
Conclusions
Five novel ruthenium sensitizers (TUS-29–TUS-33) having a calibrated with respect to the redox potential of Fc/Fc+
dianionic tridentate ligand (pyridine-2,6-dicarboxyamidato (380 mV vs. SCE).
and its derivatives) instead of three NCS ligands have been syn-
thesized. The molar absorption coefficients of TUS sensitizers
DFT calculations
were much larger than those of Black dye in the wavelength MO calculation was carried out using the DFT methods
region below 600 nm. In addition, TUS sensitizers have charac- implemented in the DMol3 code package in Materials Studio
teristic broad absorption bands in the wavelength range 5.5 (Accelrys Inc.). Ground-state geometry optimization was
between 750 and 950 nm. The energy levels of HOMOs and performed using the generalized gradient corrected (GGA)
LUMOs of TUS sensitizers were observed in the potential range function by Perdew, Burke, and Ernzerhof (PBE). Calculations
between 0.30 and 0.67 V, and −0.93 and −0.76 V vs. SCE, were performed using the double numerical plus polarization
respectively. These HOMO and LUMO energy levels of TUS sen- (DNP) basis set. The core electrons were treated with DFT
sitizers shifted in the positive direction with increasing the semi-core pseudopotentials (DSPPs).
electron-withdrawing ability of the substituents on the dianio-
nic tridentate ligand. The energy levels of HOMO and LUMO
Synthesis
of TUS sensitizers showed a linear correlation with respect to Three novel ligands (L–R) were prepared by following a general
the Hammett constant of the substituents of the dianionic tri- procedure with a synthetic yield of nearly 90%. The color
dentate ligand.
of N,N′-bis(4-dimethylaminophenyl)pyridine-2,6-dicarboxamide
The DSCs with TUS sensitizers showed much lower conver- (L-NMe2) is light green while that of the other two ligands is
sion efficiency than that of Black dye although TUS sensitizers white.
have superior light absorption properties in the wavelength
N,N′-Bis(4-dimethylaminophenyl)pyridine-2,6-dicarboxy-
range below 600 nm compared with Black dye. One of the amide (L-NMe2). Pyridine-2,6-dicarbonyl chloride (4 mmol,
possible interpretations is that the amounts of dye adsorption 0.82 g) was dissolved in 50 mL of acetonitrile. N,N′-Dimethyl-
of TUS sensitizers were much smaller than that of Black dye. p-phenylenediamine (13 mmol, 1.8 g) was added to the solu-
The other possible interpretation is that the energy levels of tion, and then this mixture was refluxed for 3 h. After the reac-
HOMO and LUMO are not suitable for the effective electron tion mixture was condensed to about 20 mL, a light green
injection and/or for the effective dye regeneration by I−. On the precipitate was filtered, washed with acetonitrile, and dried
other hand, a linear relationship between the conversion in vacuo; yield 1.4 g (86%). 1H NMR (400 MHz, DMSO): δ =
efficiency and the driving force for the reduction of TUS sensi- 10.82 (s, 2H, NH), 8.34 (d, J = 7.5 Hz, 2H, pyridine), 8.26 (t, J =
tizers by I− was observed. This result suggests that the 7.8 Hz, 1H, pyridine), 7.68 (d, J = 9.5 Hz, 4H, phenyl), 7.80
reduction process of the oxidized form of dye by I− is the rate- (d, J = 10.1 Hz, 4H, phenyl), 2.92 (s, 12H, CH3).
determining step in the DSCs with TUS sensitizers. This study
N,N′-Bis(4-trifluoromethylphenyl)pyridine-2,6-dicarboxyamide
provides important insights into the redox-potential engineer- (L-CF3). 1H NMR (400 MHz, DMSO): δ = 11.28 (s, 2H, NH),
ing of ruthenium sensitizers for the efficiency improvement of 8.46 (d, J = 7.8 Hz, 2H, pyridine), 8.36 (t, J = 7.7 Hz, 1H, pyri-
the DSCs.
dine), 8.20 (d, J = 8.9 Hz, 4H, phenyl), 7.84 (d, J = 8.6 Hz, 4H,
phenyl).
N,N′-Bis(4-trifluoromethyl-2,3,5,6-tetrafluorophenyl)pyridine-
2,6-dicarboxyamide (L-CF3F4). 1H NMR (400 MHz, DMSO): δ =
11.54 (s, 2H, NH), 8.46 (d, J = 8.0 Hz, 2H, pyridine), 8.37 (t, J =
7.8 Hz, 1H, pyridine).
Experimental section
Materials and general measurements
Black dye3c and N,N′-diphenylpyridine-2,6-dicarboxyamide17
Five novel TUS sensitizers were prepared by following a
were prepared according to the literature. All solvents and general procedure.
reagents were of the highest quality available and were used as
received.
Ru(tcterpy)(pyridine-2,6-dicarboxyamidato) (TUS-29). Ru-
(tcterpy)Cl3 (0.5 mmol, 0.29 g) and pyridine-2,6-dicarboxy-
Elemental analysis was carried out on a Perkin Elmer amide (0.5 mmol, 0.08 g) were suspended in 25 mL of DMF.
2400II elemental analyzer using acetanilide as a standard 5 mL of 0.1 M TBAOH solution was added to this mixture to
1
material. H NMR spectra were acquired on a Bruker BioSpin dissolve the starting materials. N(C2H5)3 (7.2 mmol, 0.73 g)
AVANCE 400M spectrometer, where chemical shifts in CD3OD was further added to this mixture, and then refluxed overnight.
and DMSO were referenced to the internal standard tetra- After most of the solvent was evaporated, the resulting residue
methylsilane. ESI-TOF (electrospray ionization time-of-flight) was once dissolved in a minimum amount of 0.1 M TBAOH
mass spectra were recorded on a Bruker Micromass focus solution. Undissolved materials were filtered, and the pH
spectrometer. UV-visible absorption spectra were recorded on value of this filtrate was adjusted to 2.3 by adding 0.5 M HCl.
a JASCO V-670 UV-Vis spectrophotometer. Electrochemical The crude product was filtered, and purified by column
measurements were carried out in DMSO at a sweep rate of chromatography (Sephadex LH-20) using a mixed solution of
50 mV s−1 using a glassy carbon disk working electrode, a Pt EtOH and H2O (1 : 1, v/v) containing 0.05 M TBAOH as an
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8026–8036 | 8033

View Article Online
Paper
Dalton Transactions
eluent. The main band was collected, and 0.5 M HCl was using a KEYENCE VHX-200 digital microscope. The TiO2
added until a reddish brown precipitate formed. This reddish photoelectrodes were immersed in a 1-propanol solution of
brown precipitate was filtered, washed with H2O, and dried 0.2 mM Black dye and 20 mM DCA,19 or in a mixed solvent of
in vacuo; 0.18 g (58%). 1H NMR (400 MHz, CD3OD + NaOD): 1-propanol and DMSO (1 : 1, v/v) containing 0.2 mM of each
δ = 9.07 (s, 2H, terpyridine), 8.89 (s, 2H, terpyridine), 8.22 (d, TUS sensitizer and 20 mM DCA for 3 h at room temperature to
J = 7.8 Hz, 2H, pyridine), 8.18 (t, J = 7.6 Hz, 1H, pyridine), 7.72 adsorb dyes onto the TiO2 surface.
(d, J = 5.8 Hz, 2H, terpyridine), 7.39 (d, J = 6.6 Hz, 2H, ter-
Photoelectrochemical measurements were performed in a
pyridine). ESI-TOF MS (positive ion, CH3OH): 631.01 m/z two-electrode sandwich cell configuration made up of the dye-
({[M] + H}+). Anal. calcd for ([M]·5H2O): C, 41.73; H, 3.64; adsorbed TiO2 electrode, a platinum-sputtering counter elec-
N, 11.68. Found: C, 42.15; H, 3.35; N, 11.69%.
trode, a spacer film (50 μm), and an electrolyte solution (0.05 M
Ru(tcterpy)(L) (TUS-30). Yield: 41%. 1H NMR (400 MHz, I2, 0.1 M LiI, and 0.6 M DMPImI (1,2-dimethyl-3-propylimi-
DMSO): δ = 8.89 (s, 2H, terpyridine), 8.82 (s, 2H, terpyridine), dazolium iodide) in acetonitrile). TUS sensitizers and Black
8.34 (t, J = 8.4 Hz, 1H, pyridine), 8.25 (d, J = 8.4 Hz, 2H, pyri- dye were desorbed from the TiO2 film by immersing in 0.05 M
dine), 7.93 (d, J = 6.1 Hz, 2H, terpyridine), 7.80 (d, J = 5.8 Hz, NaOH solution, and the amount of dye adsorption was
2H, terpyridine), 6.54–6.47 (m, 6H, phenyl), 5.35 (d, J = 7.0 Hz, estimated from the absorption spectrum of the resulting
4H, phenyl). ESI-TOF MS (positive ion, CH3OH): 783.08 m/z solution.
({[M] + H}+). Anal. calcd for ([M]·0.25TBA·3H2O): C, 54.94;
The photocurrent–voltage (I–V) characteristics of the DSCs
H, 4.39; N, 9.77. Found: C, 54.48; H, 4.27; N, 9.88%.
were measured on a Keithley 2400 source meter under
Ru(tcterpy)(L-NMe2) (TUS-31). Yield: 17%. 1H NMR irradiation of AM 1.5, 100 mW cm−2 (1 sun) supplied by a
(400 MHz, DMSO): δ = 8.79 (s, 2H, terpyridine), 8.67 (s, 2H, ter- solar simulator (Yamashita Denso, YSS-150A). The incident
pyridine), 8.24 (t, J = 7.9 Hz, 1H, pyridine), 8.15 (d, J = 8.0 Hz, light intensity was calibrated with a grating spectroradiometer
2H, pyridine), 7.83 (d, J = 5.5 Hz, 2H, terpyridine), 7.64 (d, J = LS-100 (EKO Instruments) and a Si photodiode (Bunkoh
5.5 Hz, 2H, terpyridine), 5.90 (d, J = 8.5 Hz, 4H, phenyl), 5.21 Keiki). The incident monochromatic photon-to-current conver-
(d, J = 9.1 Hz, 4H, phenyl). The peaks of N(CH3)2 could not be sion efficiency (IPCE) was measured on a PEC-S10 (Peccell
observed due to the overlap of the solvent peak. ESI-TOF MS Technologies).
(positive ion, CH3OH): 869.16 m/z ({[M] + H}+). Anal. calcd
for ([M]·6H2O): C, 50.46; H, 4.75; N, 11.48. Found: C, 50.34;
H, 4.82; N, 11.05%.
Acknowledgements
Ru(tcterpy)(L-CF3) (TUS-32). Yield: 57%. 1H NMR (400 MHz,
DMSO): δ = 8.89 (s, 2H, terpyridine), 8.82 (s, 2H, terpyridine), This work was supported by the New Energy and Industrial
8.38 (t, J = 7.8 Hz, 1H, pyridine), 8.32 (d, J = 8.0 Hz, 2H, pyri- Technology Development Organization (NEDO) of Japan.
dine), 7.93 (d, J = 5.5 Hz, 2H, terpyridine), 7.81 (d, J = 6.0 Hz,
2H, terpyridine), 6.91 (d, J = 8.0 Hz, 4H, phenyl), 5.61 (d, J =
9.0 Hz, 4H, phenyl). ESI-TOF MS (positive ion, CH3OH): 919.03
References
m/z ({[M] + H}+). Anal. calcd for ([M]·0.33TBA·2H2O): C, 51.45;
H, 3.69; N, 8.58. Found: C, 51.20; H, 3.64; N, 8.63%.
Ru(tcterpy)(L-CF3F4) (TUS-33). Yield: 45%. 1H NMR (400 MHz,
DMSO): δ = 9.23 (s, 2H, terpyridine), 9.16 (s, 2H, terpyridine),
8.66–8.64 (m, 3H, pyridine), 8.08 (d, J = 5.5 Hz, 2H, ter-
pyridine), 8.01 (d, J = 6.0 Hz, 2H, terpyridine). ESI-TOF MS
(positive ion, CH3OH): 1062.93 m/z ({[M] + H}+). Anal. calcd
for ([M]·3H2O): C, 41.99; H, 1.81; N, 7.53. Found: C, 42.07;
H, 1.87; N, 7.60%.
1 (a) B. O’Regan and M. Grätzel, Nature, 1991, 353, 737–
740; (b) M. Grätzel, Acc. Chem. Res., 2009, 42, 1788–
1798; (c) A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo
and H. Pettersson, Chem. Rev., 2010, 110, 6595–6663;
(d) J.-H. Yum, E. Baranoff, S. Wenger, M. K. Nazeeruddin
and M. Grätzel, Energy Environ. Sci., 2011, 4, 842–
857.
2 (a) M. Grätzel, DSC-IC 3, 2009; (b) A. Yella, H.-W. Lee,
H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin,
E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin and M. Grätzel,
Science, 2011, 334, 629–634.
3 (a) M. K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-
Baker, E. Müller, P. Liska, N. Vlachopoulos and
M. Grätzel, J. Am. Chem. Soc., 1993, 115, 6382–6390;
(b) M. K. Nazeeruddin, S. M. Zakeeruddin, R. Humphry-
Baker, M. Jirousek, P. Liska, N. Vlachopoulos, V. Shklover,
C.-H. Fischer and M. Grätzel, Inorg. Chem., 1999, 38, 6298–
6305; (c) M. K. Nazeeruddin, P. Pechy, T. Renouard,
S. M. Zakeeruddin, R. Humphyry-Baker, P. Comte, P. Liska,
L. Cevey, E. Costa, V. Shklover, L. Spiccia, G. B. Deacon and
C. A. Bignozzi, J. Am. Chem. Soc., 2001, 123, 1613–
Preparation of TiO2 electrodes and DSC
TiO2 paste was prepared using titanium isopropoxide.18 Nano-
crystalline TiO2 photoelectrodes were prepared by screen print-
ing the TiO2 paste on fluorine-doped SnO2 conducting glasses
(FTO, Nippon Sheet Glass Co., 10 Ω per square). TiO2 films
were composed of five layers (the bottom and the second layer:
20 nm TiO2 particles, the third layer: a 95 : 5 mixture of 20 nm
and 100 nm particles, the fourth layer: an 8 : 2 mixture of
20 nm and 100 nm particles, and the fifth layer: 400 nm TiO2
particles; film thickness: approximately 32 μm). TiO2 photo-
electrodes were calcinated at 520 °C after every layer was
coated. The active areas of these TiO2 films were determined
8034 | Dalton Trans., 2014, 43, 8026–8036
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
1624; (d) G. C. Vougioukalakis, A. I. Philippopoulos,
T. Stergiopoulos and P. Falaras, Coord. Chem. Rev., 2011,
255, 2602–2621; (e) A. Reynal and E. Palomares,
Eur. J. Inorg. Chem., 2011, 4509–4526; (f) A. Abbotto and
H. Kawaguchi, Y. Okuyama and H. Arakawa, Chem. Lett.,
2011, 40, 658–660; (f) H. Ozawa, H. Kawaguchi,
Y. Okuyama and H. Arakawa, Ambio, 2012, 41, 149–
150.
N. Manfredi, Dalton Trans., 2011, 40, 12421–12438; 10 (a) C.-C. Chou, K.-L. Wu, Y. Chi, W.-P. Hu, S. J. Yu,
(g) K. C. D. Robson, P. G. Bomben and C. P. Berlinguette,
Dalton Trans., 2012, 41, 7814–7829.
G.-H. Lee, C.-L. Lin and P.-T. Chou, Angew. Chem., Int. Ed.,
2011, 50, 2054–2058; (b) K.-L. Wu, C.-H. Li, Y. Chi,
J. N. Clifford, L. Cabau, E. Palomares, Y.-M. Cheng,
H.-A. Pan and P.-T. Chou, J. Am. Chem. Soc., 2012, 134,
7488–7496; (c) T. Funaki, H. Funakoshi, N. Onozawa-
Komatsuzaki, K. Kasuga, K. Sayama and H. Sugihara,
Chem. Lett., 2012, 41, 647–649.
4 (a) Y. Chiba, A. Islam, Y. Watanabe, R. Komiya, N. Koide
and L. Han, Jpn. J. Appl. Phys., 2006, 45, L638–L640;
(b) R. Yoneya-Ogura, S. Nakane, M. Morooka, M. Orihashi,
Y. Suzuki and K. Noda, Appl. Phys. Lett., 2009, 94, 073308;
(c) L. Han, A. Islam, C. Malapaka, B. Chiranjeevi, S. Zhang,
X. Yang and M. Yanagida, Energy Environ. Sci., 2012, 5, 11 (a) A. Islam, F. A. Chowdhury, Y. Chiba, R. Komiya,
6057–6060.
N. Fuke, N. Ikeda and L. Han, Chem. Lett., 2005, 34, 344–
345; (b) A. Islam, F. A. Chowdhury, Y. Chiba, R. Komiya,
N. Fuke, N. Ikeda, K. Nozaki and L. Han, Chem. Mater.,
2006, 18, 5178–5185; (c) A. Reynal, A. Forneil, E. Martinez-
Ferrero, A. Sánchez-Díaz, A. Vidal-Ferran, B. C. O’Regan
and E. Palomares, J. Am. Chem. Soc., 2008, 130, 13558–
13567; (d) S. Gao, A. Islam, Y. Numata and L. Han,
Appl. Phys. Express, 2010, 3, 062301; (e) P. G. Bomben,
K. D. Thériault and C. P. Berlinguette, Eur. J. Inorg. Chem.,
2011, 1806–1814; (f) A. Islam, S. P. Singh and L. Han,
Int. J. Photoenergy, 2011, 204639; (g) S. P. Singh, A. Islam,
M. Yanagida and L. Han, Int. J. Photoenergy, 2011, 520848;
(h) W.-S. Han, J.-K. Han, H.-Y. Kim, M. J. Choi, Y.-S. Kang,
C. Pac and S. O. Kang, Inorg. Chem., 2011, 50, 3271–
3280.
5 (a) H. T. Nguyen, H. M. Ta and T. Lund, Sol. Energy Mater.
Sol. Cells, 2007, 91, 1934–1942; (b) P. T. Nguyen, R. Degn,
H. T. Nguyen and T. Lund, Sol. Energy Mater. Sol. Cells,
2009, 93, 1939–1945.
6 O. Kohle, M. Grätzel, A. F. Meyer and T. B. Meyer, Adv.
Mater., 1997, 9, 904–906.
7 M. K. Nazeeruddin and M. Grätzel, J. Photochem. Photobiol.,
A, 2001, 145, 79–86.
8 (a) S. H. Wadman, J. M. Kroon, K. Bakker, M. Lutz,
A. L. Spek, G. P. M. Klink and G. Koten, Chem. Commun.,
2007, 1907–1909; (b) K. Chen, Y.-H. Hong, Y. Chi,
W.-H. Liu, B.-S. Chen and P.-T. Chou, J. Mater. Chem., 2009,
19, 5329–5335; (c) T. Funaki, M. Yanagida, N. Onozawa-
Komatsuzaki, K. Kasuga, Y. Kawanishi, M. Kurashige,
K. Sayama and H. Sugihara, Inorg. Chem. Commun., 2009, 12 M. Dasgupta, S. Nag, G. Das, M. Nethaji and
12, 842–845; (d) B.-S. Chen, K. Chen, Y.-H. Hong, S. Bhattacharya, Polyhedron, 2008, 27, 139–150.
W.-H. Liu, T.-H. Li, C.-H. Lai, P.-T. Chou, Y. Chi and 13 (a) H. Ozawa, S. Oura, R. Shimizu and H. Arakawa, Chem.
G.-H. Lee, Chem. Commun., 2009, 5844–5846; (e) T. Bessho,
E. Yoneda, J.-H. Yum, M. Guglielmi, I. Tavernelli, H. Imai,
U. Rothlisberger, M. K. Nazeeruddin and M. Grätzel, J. Am.
Lett., 2012, 41, 1406–1408; (b) H. Ozawa, Y. Yamamoto,
K. Fukushima, S. Yamashita and H. Arakawa, Chem. Lett.,
2013, 42, 897–899.
Chem. Soc., 2009, 131, 5930–5934; (f) W.-C. Chang, 14 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91,
H.-S. Chen, T.-Y. Li, N.-M. Hsu, Y. S. Tingare, C.-Y. Li, 165–195.
Y.-C. Liu, C. Su and W.-R. Li, Angew. Chem., Int. Ed., 2010, 15 (a) R. Hand and R. F. Nelson, J. Electrochem. Soc., 1970,
49, 8161–8164; (g) P. G. Bomben, T. J. Gordon, E. Schott
and C. P. Berlinguette, Angew. Chem., Int. Ed., 2011, 50,
10682–10685; (h) H. Kisserwan and T. H. Ghaddar, Dalton
117, 1353; (b) T. L. Macdonald, W. G. Gutheim,
R. B. Martin and F. P. Guengerich, Biochemistry, 1989, 28,
2071.
Trans., 2011, 40, 3877–3884; (i) K. C. D. Robson, 16 (a) K. Hara, Y. Dan-oh, C. Kasada, Y. Ohga, A. Shinpo,
B. D. Koivisto, A. Yella, B. Sporinova, M. K. Nazeeruddin,
T. Baumgartner, M. Grätzel and C. P. Berlinguette, Inorg.
Chem., 2011, 50, 5494–5508; ( j) T. Funaki, H. Funakoshi,
O. Kitao, N. Onozawa-Komatsuzaki, K. Kasuga, K. Sayama
and H. Sugihara, Angew. Chem., Int. Ed., 2012, 51, 7528–
7531.
9 (a) Y. Takahashi, H. Arakawa, H. Sugihara, K. Hara,
A. Islam, R. Katoh, Y. Tachibana and M. Yanagida, Inorg.
Chim. Acta, 2000, 310, 169–174; (b) A. Islam, H. Sugihara
and H. Arakawa, J. Photochem. Photobiol., A, 2003, 158, 131–
138; (c) H. Sugihara, S. Sano, T. Yamaguchi, M. Yanagida,
T. Sato, Y. Abe, Y. Nagao and H. Arakawa, J. Photochem.
Photobiol., A, 2004, 166, 81–90; (d) T. Yamaguchi,
N. Shibayama, M. Nakade, Y. Abe and H. Arakawa, Proc.
SPIE-Int. Soc. Opt. Eng., 2007, 6656, 665609; (e) H. Ozawa,
S. Suga, K. Sayama and H. Arakawa, Langmuir, 2004, 20,
4205–4210; (b) Z.-S. Wang, K. Hara, Y. Dan-oh, C. Kasada,
A. Shinpo, S. Suga, H. Arakawa and H. Sugihara, J. Phys.
Chem. B, 2005, 109, 3907–3914; (c) P. Balraju, P. Suresh,
M. Kumar, M. S. Roy and G. D. Sharma, J. Photochem.
Photobiol., A, 2009, 206, 53–63; (d) T. Marinado,
D. P. Hagberg, M. Hedlund, T. Edvinsson, E. M. J. Johansson,
G. Boschloo, H. Rensmo, T. Brinck, L. Sun and A. Hagfeldt,
Phys. Chem. Chem. Phys., 2009, 11, 133–141; (e) R. Katoh,
M. Kasuya, S. Kodate, A. Furube, N. Fuke and N. Koide,
J. Phys. Chem. C, 2009, 113, 20738–20744; (f) S. Qu, W. Wu,
J. Hua, C. Kong, Y. Long and H. Tian, J. Phys. Chem. C,
2010, 114, 1343–1349; (g) H. Long, D. Zhou, M. Zhang,
C. Peng, S. Uchida and P. Wang, J. Phys. Chem. C,
2011, 115, 14408–14414; (h) H. Ozawa, H. Kawaguchi,
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 8026–8036 | 8035

View Article Online
Paper
Dalton Transactions
Y. Okuyama and H. Arakawa, Eur. J. Inorg. Chem., 2013, 18 Z.-S. Wang, H. Kawauchi, T. Kashima and H. Arakawa,
5187–5195. Coord. Chem. Rev., 2004, 248, 1381–1389.
17 J. F. Malone, C. M. Murray, G. M. Dolan, R. Docherty and 19 H. Ozawa, M. Awa, T. Ono and H. Arakawa, Chem. –Asian J.,
A. J. Lavery, Chem. Mater., 1997, 9, 2983–2989. 2012, 7, 156–162.
8036 | Dalton Trans., 2014, 43, 8026–8036
This journal is © The Royal Society of Chemistry 2014