Towards compatibility between ruthenium sensitizers and cobalt electrolytes in dye-sensitized solar cells

Article abstract of DOI:10.1002/anie.201304608

Ruthenium and Co: Ruthenium(II) complexes remain prime candidates for dye-sensitized solar applications; however, current ruthenium sensitizers are not compatible with cobalt(II/III) electrolytes. Herein, the effect of surface insulation on device efficiency is studied by comparing two cyclometalated tris-heteroleptic ruthenium(II) complexes. This approach demonstrates a general principle that leads to unprecedented efficiency for a ruthenium(II) sensitizer used in combination with a cobalt electrolyte. Copyright

Full text of DOI:10.1002/anie.201304608

Angewandte
Chemie
Communications
Dye-Sensitized Solar Cells
Ruthenium and Co: Ruthenium(II) com-
plexes remain prime candidates for dye-
sensitized solar applications; however,
current ruthenium sensitizers are not
compatible with cobalt(II/III) electrolytes.
Herein, the effect of surface insulation on
device efficiency is studied by comparing
two cyclometalated tris-heteroleptic ruth-
enium(II) complexes. This approach
demonstrates a general principle that
leads to unprecedented efficiency for
a ruthenium(II) sensitizer used in com-
bination with a cobalt electrolyte.
L. E. Polander, A. Yella, B. F. E. Curchod,
N. Ashari Astani, J. Teuscher, R. Scopelliti,
P. Gao, S. Mathew, J.-E. Moser,
I. Tavernelli, U. Rothlisberger, M. Grꢀtzel,
M. K. Nazeeruddin,
J. Frey*
&&&& — &&&&
Towards Compatibility between
Ruthenium Sensitizers and Cobalt
Electrolytes in Dye-Sensitized Solar Cells
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
DOI: 10.1002/anie.201304608
Dye-Sensitized Solar Cells
Towards Compatibility between Ruthenium Sensitizers and Cobalt
Electrolytes in Dye-Sensitized Solar Cells**
Lauren E. Polander, Aswani Yella, Basile F. E. Curchod, Negar Ashari Astani, Joꢀl Teuscher,
Rosario Scopelliti, Peng Gao, Simon Mathew, Jacques-E. Moser, Ivano Tavernelli,
Ursula Rothlisberger, Michael Grꢁtzel, Md. Khaja Nazeeruddin, and Julien Frey*
Dye-sensitized solar cells (DSCs)[1] are low-cost alternatives
to conventional silicon technologies for solar energy con-
version. A typical DSC is composed of a chromophore that is
anchored to a mesostructured anode of anatase titania (TiO2).
Upon absorption of photons, electrons are injected from the
excited state of the dye into the TiO2 conduction band.
Subsequently, the ground state form of the sensitizer is
regenerated by reductive electron transfer from the electro-
lyte. The electrons injected into the semiconductor flow
towards the counter electrode where the redox mediator is in
turn reduced.[2] Under simulated AM 1.5 G illumination
(100 mWcmꢀ2), the power conversion efficiency of a cell (h) is
defined as the product of the generated photocurrent density
(JSC), the open-circuit photovoltage (VOC),[3] and the fill factor
(FF), as follows: h = JSC VOC FF.[4]
contrast to the iodide/triiodide couple (Iꢀ/I3 ), the electro-
chemical potential of these species can be adjusted through
substitution and/or the modification of the ligand skele-
ton.[11–15] This tuning minimizes energy losses in the device by
optimizing the driving force for regeneration, which allows for
considerable improvement of the open-circuit voltage relative
to iodine electrolytes.
ꢀ
Current ruthenium(II) sensitizers are not designed to
perform with cobalt electrolytes. Among the few examples
reported in the literature,[11,16–20] only limited performances
have been achieved. The highest performing dye (Z907)
yields efficiencies of up to 6.5% with cobalt[18] vs. 8.5% with
iodine,[21] unless the coadsorbent is specifically engineered.[22]
Molecular dynamic simulations suggest close contact inter-
actions between the cobalt(III) species and the anchored
sensitizer(s), which causes undesired recombination of the
electrons injected into the TiO2 to the redox mediator.[23]
Similar issues have been addressed for organic dyes through
the addition of peripheral bulky substituents. This strategy
prolongs the electron lifetime in the semiconductor by
preventing the electrolyte from accessing the surface.[12]
Implementation of a similar design principle could
improve compatibility between ruthenium(II) sensitizers
and cobalt electrolytes. Cyclometalated tris-heteroleptic
complexes have the desirable photoelectrochemical proper-
ties[24] to yield high-efficiency DSCs,[25] while presenting
versatility towards NCS-free sensitizers. Herein, we report
a design that takes advantage of this multifunctional structure
to insulate the surface from the electrolyte, tune the energy
levels and the light-harvesting properties of the complex, and
provide an anchor to the TiO2. Owing to this rational design,
DSCs exhibiting comparable efficiencies with both cobalt and
iodine redox mediators were obtained.
In addition to their long-term stability, sensitizers of the
[Ru(2,2’-bipyridine)2(NCS)2] type have attained power con-
version efficiencies of up to 12% with iodine electrolytes.[5–9]
The recent surge of organic dyes to this level of performance
can be largely explained by the development of alternative
redox mediators, in particular cobalt(II/III) complexes.[10] In
[*] Dr. L. E. Polander, Dr. A. Yella, Dr. J. Teuscher, R. Scopelliti,
Dr. P. Gao, Dr. S. Mathew, Prof. M. Grꢀtzel, Dr. M. K. Nazeeruddin,
Dr. J. Frey
Laboratory of Photonics and Interfaces, Institute of Chemical
Science and Engineering, ꢁcole Polytechnique Fꢂdꢂrale
de Lausanne, 1015 Lausanne (Switzerland)
E-mail: julien.frey@gmail.com
B. F. E. Curchod, N. Ashari Astani, Dr. I. Tavernelli,
Prof. U. Rothlisberger
Laboratory of Computational Chemistry and Biochemistry, Institute
of Chemical Science and Engineering, ꢁcole Polytechnique Fꢂdꢂrale
de Lausanne, 1015 Lausanne (Switzerland)
To improve compatibility of ruthenium(II) sensitizers
with cobalt redox mediators, one could design a tris-hetero-
leptic complex that maintains favorable highest occupied
molecular orbital (HOMO) energetics while also enabling
interfacial control at the TiO2 surface. However, thus far the
literature does not contain a ligand that fulfills both of these
criteria.[26,27] We found 2’,6’-dimethoxy-2,3’-bipyridine to be
convenient for tuning the HOMO energy level. The alkoxy
substituents are also advantageous as a substitution point to
insulate the TiO2 surface by elongation of the chains. To
validate this concept, both methoxy- and dodecyloxy-deriv-
atives were synthesized. The other two ligands used to
complete the coordination sphere of the ruthenium center
have two functions: 2,2’-bis(5-hexylthiophen-2-yl)-2,2’-bipyr-
Prof. J.-E. Moser
Group for Photochemical Dynamics, Institute of Chemical Science
and Engineering, ꢁcole Polytechnique Fꢂdꢂrale de Lausanne
1015 Lausanne (Switzerland)
[**] We acknowledge financial support for this work from Solvay SA and
the European Community’s Seventh Framework Programme (FP7/
2007-2013) under grant agreement no. 281063 of the Powerweave
project. J.T. thanks the Swiss National Science Foundation for
financial support (200020_143908). Support from the Swiss
National Science Foundation (200020-130082) and the NCCR-
MUST interdisciplinary research program is also gratefully
acknowledged. J.F. is grateful to Dr. E. Baranoff (University of
Birmingham) for his invaluable mentoring.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201304608.
2
www.angewandte.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
ꢀ
Scheme 1. Synthesis of complexes 3a and 3b. Charged complexes were isolated as PF6 salts. a) 2’,6’-dimethoxy-2,3’-bipyridine or 2’,6’-
bis(dodecyloxy)-2,3’-bipyridine, NaOH, KPF6 in MeCN (1a: 56%, 1b: 43%); b) 2,2’-bis(5-hexylthiophen-2-yl)-2,2’-bipyridine, 4,4’-dimethylester-2,2’-
bipyridine in EtOH (2a: 48%; 2b: 45%); c) DMF, H2O, Et3N (3:1:1) (3a: 77%; 3b: 95%).
idine enhances the absorptivity of the complex, compared to
an unsubstituted 2,2’-bipyridine ligand,[7] whereas 2,2’-bipyr-
idine-4,4’-dicarboxylic acid serves as an electron-accepting
ligand and anchor to the TiO2.
Complexes 1–3 were prepared by following a straightfor-
ward procedure developed by Berlinguette et al.[19] As shown
in Scheme 1, with [{Ru(C6H6)Cl2}2] as the ruthenium source,
the cyclometalating ligand is introduced first to yield pre-
cursors 1a and 1b. The complex is subsequently reacted with
stoichiometric amounts of the two different bipyridine
ligands, in a one-pot procedure that gives the desired
complexes, 2a or 2b, in yields close to the theoretical limit.
Finally, sensitizers 3a and 3b are obtained as neutral
complexes after saponification. Unlike most ruthenium(II)
sensitizers, which are isolated using size-exclusion chroma-
tography, complexes 1–3 can be purified using conventional
Figure 1. Crystal structure of 2a·2CH2Cl2. Thermal ellipsoids are drawn
at the 50% probability level. The counter ion, hexyl chains, hydrogen
atoms, and solvent molecules are omitted for clarity. Additional
information can be found in Table S1. N blue, O red, S yellow, Ru tur-
quoise.
chromatography over alumina or silica and by using precip-
itation techniques. Details on the synthesis and character-
ization of these compounds are given in the Supporting
Information.
ligand, the thiophenes lie in the 2,2’-bipyridine plane (torsions
angles < 18), thus ensuring good electronic communication.
The optical and electrochemical properties of 3a and 3b
were determined in dichloromethane solution (Figure 2; see
also Figures S10 and S11). Both complexes show virtually
identical properties, as summarized in Table S3. The UV/Vis
spectrum of 3b shows broad and intense absorptions below
350 nm that are dominated by p–p* transitions. At longer
wavelengths, the spectrum reveals two strong bands centered
Owing to their tris-heteroleptic nature, complexes 2 and 3
can potentially adopt two conformations with cyclometala-
tion either in trans to the dicarboxy-bipyridine or in trans to
the dithiophenyl-bipyridine ligand. The NMR data indicate
that a single isomer is produced. To confirm the arrangement
of the ligands around the metal center, single crystals of 2a
were grown by slow diffusion of hexane into a dichloro-
methane solution. The crystal structure is shown in Figure 1,
with selected bond lengths and angles presented in the
Supporting Information, Table S2. As expected, the complex
adopts a distorted octahedral geometry around the ruthenium
center. The dimethoxy ester bipyridine ligand, located trans to
the cyclometalated ring, experiences the electron donating
character of the C24–Ru bond, which manifests in an
elongation of the Ru–N1 bond. The NMR analysis from
which we conclude that complexes 2 and 3 adopt this same
conformation (Figures S7 and S8) is consistent with the
observed trans effect, as detailed in the Supporting Informa-
tion. The methoxy substituents of the cyclometalated ring are
arranged within the plane of the ligand to avoid lone pair
interactions with the pyridyl nitrogen. At the light-harvesting
Figure 2. a) Steady-state absorption (c) and normalized emission
spectra (a, lexc =580 nm), and b) differential pulse voltammogram
(c) with a ferrocene internal standard (a) in dichloromethane of
3b. Scan rate of 100 mVsꢀ1 with 10 mV pulses.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
at 418 nm (23400m cmꢀ1) and 580 nm (19600m cmꢀ1), which
are characteristic of metal-to-ligand charge-transfer transi-
tions. These assignments are further supported by time-
dependent density functional theory calculations for 3a (LR-
TDDFT/M06; Figures S12–S21). Full details on the quantum
chemical calculations can be found in the Supporting
Information.
Furthermore, the energy levels of complexes 3a and 3b
are adequately poised for use in DSCs. The differential pulse
voltammogram of complex 3b exhibits quasi-reversible one-
electron oxidation at + 0.86 V (EHOMO = ꢀ4.91 eV) and one-
electron reduction at ꢀ1.35 V vs. NHE.[28] The anodic peak
potential is ascribed to oxidation of the metal center to
ruthenium(III), whereas the cathodic peak is assigned to
a ligand-centered reduction. According to (U)DFT/M06
calculations, the vertical ionization energy of 3a is calculated
at 5.04 eV, whereas the adiabatic ionization energy lies at
4.96 eV. Both values are in good agreement with the
estimated HOMO energy and corroborate the oxidation of
the ruthenium center.[29] As observed in Figure S13, the spin
Figure 3. a,b) PIA spectra after 470 nm excitation, and (c,d) TA decay
traces following excitation at 520 nm (probe at 750 nm) of 8.0 mm
thick TiO2 films coated with 3a (a, c) and 3b (b, d), respectively.
Decays in acetonitrile (g) and in presence of [CoII/III(phen)3]2/3+
electrolyte (c) were fitted with a single exponential component
DA(t)/A0 exp[ꢀ(t/t)]. 7 ns (fwhm) laser pulse of 50 mJcmꢀ2
.
+
density of the geometry optimized 3aC appears only slightly
(kreg+krec) = 96% (where ki = 1/ti), which is paramount for
efficient DSCs.
The incident photon-to-current efficiency (IPCE) spectra
using a [CoII/III(phen)3]2/3+ electrolyte are shown in Figure 4a.
Complex 3b was also investigated using Iꢀ/I3ꢀ for comparison
purposes. The three IPCEs have a similar shape in the range
delocalized over the cyclometalated pyridine.
+
+
These ES+/S values allow regeneration of 3aC and 3bC
from iodine (E0 =+ 0.35 V vs. NHE), as well as a variety of
cobalt-based electrolytes with more positive electrochemical
potentials.[13] In this study, a tris(1,10-phenanthroline) cobalt-
(II/III) ([CoII/III(phen)3]2/3+, E0 =+ 0.62 V vs. NHE) electro-
lyte is chosen to maximize the VOC while maintaining
sufficient driving force for regeneration. The excited state
oxidation potential of the dyes (ES+/S*) is estimated using the
zero–zero transition energy (E0-0) measured at the intersec-
tion of the absorption and normalized emission spectra. The
ES+/S* values obtained for 3a and 3b (ca. ꢀ0.90 V vs. NHE)
provide a priori ample driving force for electron injection into
the TiO2 conduction band (EC = ca. ꢀ0.5 V vs. NHE).[30]
Provided that electron injection is not an issue, the
efficiency of TiO2 films sensitized with 3a and 3b depends on
two main charge-transfer reactions: 1) back-recombination,
which corresponds to electron transfer from the electron in
the TiO2 to the oxidized dye, and 2) dye-regeneration, which
corresponds to reductive electron transfer from the electro-
lyte to the oxidized sensitizer. Their respective time constants
(trec and treg) can be quantified using transient absorption
(TA) decay measurements. The oxidized forms of the dyes
have absorption spectra that differ sharply from their neutral
counterparts (Figure 3a,b). Following excitation at 470 nm,
the photoinduced absorption (PIA) signal that appears at
740 nm is characteristic of the formation of a ruthenium(III)
species,[31] whose TA decay is used to derive the correspond-
ing time constants. The decays were fitted with a single
exponential component DA(t) / A0exp[-(t/t)], where A0 is the
pre-exponential factor and t is the characteristic time
constant.[32] The fits are represented as solid lines on Fig-
ure 3c,d. With an inert electrolyte, trec was estimated at 280
and 373 ms for 3a and 3b, respectively. In contrast, the
presence of [CoII/III(phen)3]2/3+ gives rise to accelerated TA
decays, with treg values of 9.7 ms and 17 ms for 3a and 3b,
respectively. These trec and treg constants indicate nearly
Figure 4. a) Photocurrent action spectra in mesoscopic solar cells, and
b) J–V characteristics measured under simulated AM 1.5 G full sun
illumination (100 mWcmꢀ2) for devices employing 3a and 3b. Film
thickness: (4.0+4.0) mm. Cobalt electrolyte: 0.25m [Co(phen)3](TFSI)2,
0.05m [Co(phen)3](TFSI)3, 0.25m TBP, 0.1m LiTFSI in acetonitrile;
iodine electrolyte: 1.0m PMII, 0.03m I2, 0.5m TBP, 0.1m GuNCS,
0.05m LiI. 3a+cobalt (a), 3b+cobalt (c), 3b+iodine (g).
GuNCS=guanidinium thiocyanate, phen=1,10-phenanthroline,
PMII=1-propyl-3-methylimidazolium iodide, TBP=4-tert-butylpyridine,
TFSI=trifluoromethanesulfonimide.
of 380–770 nm, but with significantly different intensities.
Similar to other ruthenium(II) sensitizers,[18,19,22] the photo-
current action spectra is limited to ca. 50% with 3a. The
presence of C12 alkoxy chains on the cyclometalated ligand
results in an increase of up to 70%, as exemplified by 3b.
Notably, this is lower than what can be obtained when iodine
is used as redox mediator. The solar-to-electricity conversion
efficiencies were evaluated by recording the J–V character-
istics under simulated AM 1.5 G illumination (100 mWcmꢀ2).
Results are depicted in Figure 4b and photovoltaic parame-
ters are reported in Table 1. As expected, the JSC values follow
quantitative regeneration quantum yields, Freg = kreg
/
4
www.angewandte.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
Table 1: Detailed photovoltaic parameters of devices employing 3a and
3b under simulated AM 1.5 G illumination (100 mWcmꢀ2).
improved in cobalt-based DSCs. Upon increasing the steric
properties of the cyclometalating ring, notable improvements
in photon-to-current conversion and J–V characteristics are
observed. Transient absorption decay measurements indicate
that the kinetics of recombination and regeneration remain
constant, despite the modifications of the sensitizers. How-
ever, through charge extraction measurements, the device
efficiency is directly correlated to increased surface protec-
tion. Physical insulation of the TiO2 surface prolongs the
electron lifetime by preventing abnormal recombination of
injected electrons into the cobalt(III) species. This approach
demonstrates a general principle that leads to unprecedented
efficiency for a ruthenium(II) sensitizer used in combination
with a cobalt electrolyte.
Dye
Electrolyte
J
SC [mAcmꢀ2
]
VOC [mV]
FF
h [%]
3a
3b
3b
cobalt
cobalt
iodine
8.3
13.2
16.3
714
837
715
0.79
0.78
0.75
4.7
8.6
8.7
the trend observed in IPCE intensities. The measured VOC
values show large differences between 3a and 3b. The lower
VOC obtained with iodine in the case of 3b highlights the
benefit of using a cobalt electrolyte.
Relative shifts in IPCE intensities and J–V characteristics
can be rationalized based on transient photovoltage decays
and charge extraction measurements. While maintaining
constant electrolyte, variations in VOC can often be attributed
to shifts in the conduction band edge of TiO2.[33] In the
presence of [CoII/III(phen)3]2/3+, we observe negligible varia-
tion in the conduction band edge (Figure 5a), which indicates
Received: May 28, 2013
Published online: && &&, &&&&
Keywords: cobalt · dye-sensitized solar cells ·
.
energy conversion · ruthenium · sensitizers
[1] B. OꢀRegan, M. Grꢁtzel, Nature 1991, 353, 737 – 740.
[2] A. Hagfeldt, G. Boschloo, L. Sun, L. Kloo, H. Pettersson, Chem.
Rev. 2010, 110, 6595 – 6663.
[3] The VOC is defined as the energy difference between the Fermi
level (EF) of the TiO2 and the HOMO level of the redox
mediator, as follows: VOC = (EFꢀEHOMO)/e, where e is the
elementary charge.
[4] This equation is applicable under 100 mWcmꢀ2 irradiation with
JSC in mAcmꢀ2 and VOC in V.
[5] M. K. Nazeeruddin, F. De Angelis, S. Fantacci, A. Selloni, G.
Viscardi, P. Liska, S. Ito, B. Takeru, M. Grꢁtzel, J. Am. Chem.
Soc. 2005, 127, 16835 – 16847.
Figure 5. a) Charge density of TiO2 (dn), and b) electron lifetime (tn)
*
against the open-circuit photovoltage (VOC). 3a+cobalt ( ), 3b+cobalt
*
( ).
[6] Y. Chiba, A. Islam, Y. Watanabe, R. Komiya, N. Koide, L. Han,
Jpn. J. Appl. Phys. 2006, 45, L638 – L640.
[7] F. Gao, Y. Wang, D. Shi, J. Zhang, M. Wang, X. Jing, R.
Humphry-Baker, P. Wang, S. M. Zakeeruddin, M. Grꢁtzel, J.
Am. Chem. Soc. 2008, 130, 10720 – 10728.
[8] C.-Y. Chen, M. Wang, J.-Y. Li, N. Pootrakulchote, L. Alibabaei,
C.-H. Ngoc-le, J.-D. Decoppet, J.-H. Tsai, C. Grꢁtzel, C.-G. Wu
et al., ACS Nano 2009, 3, 3103 – 3109.
[9] Q. Yu, Y. Wang, Z. Yi, N. Zu, J. Zhang, M. Zhang, P. Wang, ACS
Nano 2010, 4, 6032 – 6038.
[10] 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, M.
Grꢁtzel, Science 2011, 334, 629 – 634.
[11] S. A. Sapp, C. M. Elliott, C. Contado, S. Caramori, C. A.
Bignozzi, J. Am. Chem. Soc. 2002, 124, 11215 – 11222.
[12] S. M. Feldt, E. A. Gibson, E. Gabrielsson, L. Sun, G. Boschloo,
A. Hagfeldt, J. Am. Chem. Soc. 2010, 132, 16714 – 16724.
[13] S. M. Feldt, G. Wang, G. Boschloo, A. Hagfeldt, J. Phys. Chem. C
2011, 115, 21500 – 21507.
[14] J.-H. Yum, E. Baranoff, F. Kessler, T. Moehl, S. Ahmad, T.
Bessho, A. Marchioro, E. Ghadiri, J.-E. Moser, C. Yi et al., Nat.
Commun. 2012, 3, 631.
[15] M. K. Kashif, J. C. Axelson, N. W. Duffy, C. M. Forsyth, C. J.
Chang, J. R. Long, L. Spiccia, U. Bach, J. Am. Chem. Soc. 2012,
134, 16646 – 16653.
that the TiO2 experiences a similar dipole shift with 3a and
3b. Therefore, differences in VOC can be attributed to the
relative electron lifetimes in TiO2 (tn); the electron lifetime of
3b is prolonged by ca. 1.5 orders of magnitude relative to 3a
(Figure 5b). The recombination of electrons injected into the
semiconductor can occur by reductive electron transfer to the
electrolyte and/or to the oxidized form of the sensitizer.
Given that the time constants for recombination of injected
electrons with 3aC and 3bC are similar, the presence of C12
alkoxy chains clearly insulates the TiO2 surface by preventing
close proximity with the redox mediator. This effect has
a major impact on both the JSC and VOC photovoltaic
parameters, as shown in Table 1. Overall, the JSC and VOC
obtained for 3a (8.3 mAcmꢀ2 and 714 mV, respectively)
contribute to a power conversion efficiency of 4.7%. As
a result of the increased electron lifetime, the values
measured for 3b rise to 13.2 mAcmꢀ2 and 837 mV, respec-
tively; DSCs with this sensitizer reach efficiencies of up to
8.6% in the presence of a cobalt electrolyte. Notably, this
result closely matches that obtained with iodine (8.7%),
which has not been thus far reported for a ruthenium(II)
sensitizer.
+
+
[16] H. Nusbaumer, S. M. Zakeeruddin, J.-E. Moser, M. Grꢁtzel,
Chem. Eur. J. 2003, 9, 3756 – 3763.
In summary, we have shown that the efficiency of
cyclometalated ruthenium(II) sensitizers can be greatly
[17] B. M. Klahr, T. W. Hamann, J. Phys. Chem. C 2009, 113, 14040 –
14045.
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
[18] Y. Liu, J. R. Jennings, Y. Huang, Q. Wang, S. M. Zakeeruddin, M.
Grꢁtzel, J. Phys. Chem. C 2011, 115, 18847 – 18855.
[19] P. G. Bomben, T. J. Gordon, E. Schott, C. P. Berlinguette, Angew.
Chem. 2011, 123, 10870 – 10873; Angew. Chem. Int. Ed. 2011, 50,
10682 – 10685.
[20] Y. Xie, T. W. Hamann, J. Phys. Chem. Lett. 2013, 4, 328 – 332.
[21] S. M. Zakeeruddin, M. K. Nazeeruddin, R. Humphry-Baker, P.
Pꢂchy, P. Quagliotto, C. Barolo, G. Viscardi, M. Grꢁtzel,
Langmuir 2002, 18, 952 – 954.
[22] Y. Liu, J. R. Jennings, X. Wang, Q. Wang, Phys. Chem. Chem.
Phys. 2013, 15, 6170 – 6174.
[23] E. Mosconi, J.-H. Yum, F. Kessler, C. J. Gomez-Garcia, C.
Zuccaccia, A. Cinti, M. K. Nazeeruddin, M. Grꢁtzel, F. De An-
gelis, J. Am. Chem. Soc. 2012, 134, 19438 – 19453.
[26] P. G. Bomben, B. D. Koivisto, C. P. Berlinguette, Inorg. Chem.
2010, 49, 4960 – 4971.
[27] P. G. Bomben, K. D. Thꢂriault, C. P. Berlinguette, Eur. J. Inorg.
Chem. 2011, 1806 – 1814.
[28] N. G. Connelly, W. E. Geiger, Chem. Rev. 1996, 96, 877 – 910.
[29] The vertical ionization corresponds to the energy difference
between the cationic and the neutral species at the molecular
geometry of the neutral compound; the adiabatic ionization
energy corresponds to the energy difference between the
optimized cationic and optimized neutral species.
[30] A. Hagfeldt, M. Grꢁtzel, Chem. Rev. 1995, 95, 49 – 68.
[31] S. Ardo, G. J. Meyer, Chem. Soc. Rev. 2009, 38, 115 – 164.
[32] A. Y. Anderson, P. R. F. Barnes, J. R. Durrant, B. C. OꢀRegan, J.
Phys. Chem. C 2011, 115, 2439 – 2447, and references therein.
[33] P. R. F. Barnes, K. Miettunen, X. Li, A. Y. Anderson, T. Bessho,
M. Grꢁtzel, B. C. OꢀRegan, Adv. Mater. 2013, 25, 1881 – 1922.
[24] P. G. Bomben, K. C. D. Robson, P. A. Sedach, C. P. Berlinguette,
Inorg. Chem. 2009, 48, 9631 – 9643.
[25] T. Bessho, E. Yoneda, J.-H. Yum, M. Guglielmi, I. Tavernelli, H.
Imai, U. Rothlisberger, M. K. Nazeeruddin, M. Grꢁtzel, J. Am.
Chem. Soc. 2009, 131, 5930 – 5934.
6
www.angewandte.org
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 6
These are not the final page numbers!