Dye molecular structure device open-circuit voltage correlation in Ru(II) sensitizers with heteroleptic tridentate chelates for dye-sensitized solar cells

Article abstract of DOI:10.1021/ja300828f

Dicarboxyterpyridine chelates with π-conjugated pendant groups attached at the 5- or 6-position of the terminal pyridyl unit were synthesized. Together with 2,6-bis(5-pyrazolyl)pyridine, these were used successfully to prepare a series of novel heteroleptic, bis-tridentate Ru(II) sensitizers, denoted as TF-11-14. These dyes show excellent performance in dye-sensitized solar cells (DSCs) under AM1.5G simulated sunlight at a light intensity of 100 mW cm ^-2 in comparison with a reference device containing [Ru(Htctpy)(NCS)₃][TBA]₃ (N749), where H₃tctpy and TBA are 4,4′,4″-tricarboxy-2,2′:6′,2″- terpyridine and tetra-n-butylammonium cation, respectively. In particular, the sensitizer TF-12 gave a short-circuit photocurrent of 19.0 mA cm^-2, an open-circuit voltage (V_OC) of 0.71 V, and a fill factor of 0.68, affording an overall conversion efficiency of 9.21%. The increased conjugation conferred to the TF dyes by the addition of the π-conjugated pendant groups increases both their light-harvesting and photovoltaic energy conversion capability in comparison with N749. Detailed recombination processes in these devices were probed by various spectroscopic and dynamics measurements, and a clear correlation between the device V_OC and the cell electron lifetime was established. In agreement with several other recent studies, the results demonstrate that high efficiencies can also be achieved with Ru(II) sensitizers that do not contain thiocyanate ancillaries. This bis-tridentate, dual-carboxy anchor configuration thus serves as a prototype for future omnibearing design of highly efficient Ru(II) sensitizers suited for use in DSCs.

Full text of DOI:10.1021/ja300828f

Article
pubs.acs.org/JACS
Dye Molecular Structure Device Open-Circuit Voltage Correlation in
Ru(II) Sensitizers with Heteroleptic Tridentate Chelates for Dye-
Sensitized Solar Cells
Kuan-Lin Wu,†,⊥ Cheng-Hsuan Li, Yun Chi,* John N. Clifford, Lydia Cabau, Emilio Palomares,*
†,⊥
,†
‡
‡
,‡,§
∥
∥
,∥
Yi-Ming Cheng, Hsiao-An Pan, and Pi-Tai Chou*
†
Department of Chemistry and Low Carbon Energy Research Center, National Tsing Hua University, Hsinchu 30013, Taiwan
Institute of Chemical Research of Catalonia (ICIQ), Avda. Països Catalans 16, 43007 Tarragona, Spain
ICREA, Avda. Lluís Companys 28, Barcelona E-08030, Spain
‡
§
∥
Department of Chemistry and Center for Emerging Material and Advanced Devices, National Taiwan University, Taipei 10617,
Taiwan
*
S Supporting Information
ABSTRACT: Dicarboxyterpyridine chelates with π-conju-
gated pendant groups attached at the 5- or 6-position of the
terminal pyridyl unit were synthesized. Together with 2,6-
bis(5-pyrazolyl)pyridine, these were used successfully to
prepare a series of novel heteroleptic, bis-tridentate Ru(II)
sensitizers, denoted as TF-11−14. These dyes show excellent
performance in dye-sensitized solar cells (DSCs) under
AM1.5G simulated sunlight at a light intensity of 100 mW
−
2
cm in comparison with a reference device containing
Ru(Htctpy)(NCS) ][TBA] (N749), where H tctpy and
[
3
3
3
TBA are 4,4′,4″-tricarboxy-2,2′:6′,2″-terpyridine and tetra-n-butylammonium cation, respectively. In particular, the sensitizer
−2
TF-12 gave a short-circuit photocurrent of 19.0 mA cm , an open-circuit voltage (VOC) of 0.71 V, and a fill factor of 0.68,
affording an overall conversion efficiency of 9.21%. The increased conjugation conferred to the TF dyes by the addition of the π-
conjugated pendant groups increases both their light-harvesting and photovoltaic energy conversion capability in comparison
with N749. Detailed recombination processes in these devices were probed by various spectroscopic and dynamics
measurements, and a clear correlation between the device VOC and the cell electron lifetime was established. In agreement with
several other recent studies, the results demonstrate that high efficiencies can also be achieved with Ru(II) sensitizers that do not
contain thiocyanate ancillaries. This bis-tridentate, dual-carboxy anchor configuration thus serves as a prototype for future
omnibearing design of highly efficient Ru(II) sensitizers suited for use in DSCs.
ye-sensitized solar cells (DSCs) have received significant
class of dye, are known as [Ru(H dcbpy) (NCS) ] (N3) and
2 2 2
D
attention as potential low-cost alternatives to conven-
tional semiconductor-based solar cells made from silicon. In
a DSC, a suitable sensitizer/dye is deposited on a porous film
[Ru(Hdcbpy) (NCS) ][TBA] (N719), where H dcbpy and
2 2 2 2
1
−7
TBA are 4,4′-dicarboxy-2,2′-bipyridine and tetra-n-butylammo-
nium cation, respectively. Another much studied dye is the
panchromatic sensitizer [Ru(Htctpy)(NCS) ][TBA] (N749),
made of a nanocrystalline metal oxide such as TiO . Upon
2
3
3
illumination with solar radiation, the photoinjected electrons
diffuse through the interconnected mesoporous particles and
are then extracted at the anode, while the oxidized sensitizer is
rapidly regenerated by the I /I redox couple in the electrolyte
solution. The oxidized I diffuses to the cathode and is
where H tctpy is 4,4′,4″-tricarboxy-2,2′:6′,2″-terpyridine. This
3
dye is notable for its ability to harvest solar radiation extending
15
−
−
into the near-IR region. Importantly, all of these Ru(II)
3
−
sensitizers possess thiocyanate ancillaries, whose intimate
3
−
−
−
8,9
involvement in the regeneration process with the I /I
3
reduced back to I to complete the regenerative process.
redox couple has been confirmed experimentally and also
using DFT calculations.
Among the fruitful progress made for DSCs, the innovative
design of the dye sensitizer is of pivotal importance in achieving
high efficiency. Typical sensitizers include both pure organic
A significant departure from the above molecular design was
1
6,17
recently reported by van Koten and co-workers,
who
(
i.e., metal-free) and organometallic (i.e., metal-containing)
1
0−13
demonstrated that cyclometalating chelates are capable of
dyes,
for which the Ru(II)-based sensitizers are superior to
most others and are capable of producing certified cells with
1
4
impressive efficiencies of up to ∼11.7%. The most studied
Received: January 25, 2012
Published: April 16, 2012
Ru(II) dyes, which may be considered as the templates for this
©
2012 American Chemical Society
7488
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496

Journal of the American Chemical Society
Article
preserving the light-harvesting properties of a class of bis-
these novel dye sensitizers to serve as sensitizers in stable, high-
efficiency DSCs is moreover discussed.
tridentate Ru(II) complexes. Subsequently, Grat
̈
zel and co-
workers announced the design of tris-bidentate [Ru-
+
(
H dcbpy) (dfppy)] , where dfppy is 4,6-difluorophenylpyr-
2 2
EXPERIMENTAL SECTION
■
idinato, which in the absence of thiocyanate can still afford a
high conversion efficiency (η) of 10.1%, suggesting that the
almost ever present thiocyanate is not a complete necessity in
General Procedures. All of the reactions were performed under
an argon atmosphere, and solvents were distilled from appropriate
drying agents prior to use. Commercially available reagents were used
without further purification, unless otherwise stated. All reactions were
monitored using precoated thin-layer chromatography plates (0.20
mm with fluorescent indicator UV254). Mass spectra were obtained
on a JEOL SX-102A instrument operated in either electron impact
(EI) or fast atom bombardment (FAB) mode. H and C NMR
spectra were recorded on a Varian Mercury-400 or INOVA-500
instrument. Elemental analysis was carried out with a Heraeus CHN-O
Rapid Elementary Analyzer.
18
the design of efficient Ru(II) sensitizers. Subsequent studies
demonstrating the advantages of using cyclometalating chelates
in the assembly of thiocyanate-free Ru(II) complexes for DSC
applications were independently reported by Berlinguette and
1
9,20
1
13
co-workers.
A high η of 8.02% has been documented for
both tricarboxyterpyridine and dicarboxybipyridine Ru(II)
21
complexes that are devoid of thiocyanate ligands.
Complementary to the aforementioned progress, our
research team has been involved in the functionalization of
Ru(II) sensitizers with either bidentate or tridentate azolate
Synthesis of TF-11. Ligand L1 (375 mg, 0.69 mmol) and
RuCl ·3H O (198 mg, 0.76 mmol) in 15 mL of absolute ethanol was
3
2
refluxed for 4 h. The mixture was cooled to room temperature (RT)
and filtered. The precipitate was washed with ethanol and dried in
vacuum to give a brown solid. This brown solid (100 mg, 0.133 mmol)
was then mixed with 2,6-bis(3-trifluoromethyl-1H-pyrazol-5-yl)-
pyridine (45.4 mg, 0.133 mmol) and KOAc (65.3 mg, 0.67 mmol)
in 15 mL of toluene and refluxed for 6 h. After cooling to RT, the
solvent was removed by rotary evaporation. The residue was dissolved
2
2−25
chelates.
It is believed that the azolic group may be more
reactive than the traditional heterocyclic cyclometalates because
of the greater N−H acidity. Moreover, the stronger electron-
withdrawing character of azolate is also capable of increasing
the oxidation potential of Ru(II) sensitizers containing these
ligands, which is essential for the facile regeneration of the
−
−
in CH Cl , washed with water, and dried over anhydrous MgSO .
2
2
4
oxidized sensitizers by the I /I redox couple. Indeed, only
3
After removal of the solvent, the residue was purified by silica gel
column chromatography, eluting with ethyl acetate. The obtained
brown solid was then dissolved in a mixture of 10 mL of acetone and
with cyclometalate chelates possessing sufficient electron-
deficient character has it been possible to achieve satisfactorily
19
high η under typical device operation conditions.
0
.57 mL of 0.5 N NaOH. The resulting solution was brought to reflux
In yet another approach, we recently reported a class of
tricarboxyterpyridine Ru(II) complexes such as TF-1 and TF-2
for 3 h and then cooled to RT and concentrated. The residue was
dissolved in excess water and titrated with 2 N HCl solution to reach
(
Scheme 1), which contain as ancillary ligands 2,6-bis(3-
pH 3. The precipitate was collected and washed with water, CH
and acetone in sequence to yield the pure product TF-11 (42.9 mg,
3%).
Selected Spectral Data for TF-11. MS (FAB, 102Ru): m/z 935 [M
2 2
Cl ,
3
Scheme 1. Structures of the Ru(II) Sensitizers TF-1, TF-2,
PRT-11, and PRT-12
+
1
+
H] . H NMR (500 MHz, DMSO-d , 298 K): δ 9.30 (s, 1H), 9.18
6
(
s, 1H), 9.11 (s, 1H), 8.89 (d, JHH = 8.0 Hz, 1H), 8.20 (t, JHH = 8.5 Hz,
1
7
2
0
H), 8.14 (d, JHH = 6.5 Hz, 1H), 8.09 (d, JHH = 8.5 Hz, 2H), 7.69−
.66 (m, 2H), 7.29 (s, 2H), 7.21 (s, 2H), 6.83 (d, JHH = 4.0 Hz, 1H),
.72 (t, JHH = 6.0 Hz, 2H), 1.54−1.51 (m, 2H), 1.24−1.12 (m, 6H),
.84 (t, JHH = 7.0 Hz, 3H). 19F NMR (376 MHz, DMSO-d , 298 K): δ
6
−
58.48 (s, 6F).
Synthesis of TF-12. With the procedure described for TF-11, the
reaction of L2 (175 mg, 0.32 mmol), RuCl ·3H O, (92 mg, 0.35
3
2
mmol), 2,6-bis(3-trifluoromethyl-1H-pyrazol-5-yl)pyridine (46 mg,
.13 mmol), and 4-ethylmorpholine (0.04 mL, 0.35 mmol) afforded,
0
after NaOH hydrolysis and subsequent acidification, TF-12 as a brown
solid product (99 mg, 36%).
Selected Spectral Data for TF-12. MS (FAB, 102Ru): m/z 934 [M
+
1
+
H] . H NMR (400 MHz, DMSO-d , 298 K): δ 9.21 (s, 1H), 9.15
6
(
s, 1H), 9.00 (d, JHH = 8.0 Hz, 1H), 8.97 (s, 1H), 7.97 (t, JHH = 8.0 Hz,
trifluoromethylpyrazol-5-yl)pyridine and its derivative bearing a
26
1H), 7.89 (t, JHH = 8.0 Hz, 1H), 7.73 (d, JHH = 8.0 Hz, 2H), 7.62 (d,
JHH = 8.0 Hz, 1H), 7.32 (d, JHH = 8.0 Hz, 1H), 7.11 (s, 2H), 6.78 (d,
JHH = 6 Hz, 1H), 6.11 (d, JHH = 3.2 Hz, 1H), 5.38 (d, JHH = 3.2 Hz,
1
(
π-conjugated pendant group, respectively. These sensitizers
allow for excellent DSCs with η values of up to 10.7%.
Moreover, we also speculated that the π-conjugated pendant
group at the ancillary site could alternatively be placed on the
terpyridine ligand, rendering sensitizers showing comparable or
perhaps even better efficiencies in DSC devices. The desired
dicarboxyterpyridine chelates and the associated Ru(II)
complexes such as PRT-11 and PRT-12 (Scheme 1) were
H), 2.57 (t, JHH = 7.6 Hz, 2H), 1.50 (quin, J = 7.6 Hz, 2H), 1.31
HH
19
m, 6H), 0.87 (t, JHH = 6.8 Hz, 3H). F NMR (376 MHz, DMSO-d6,
2
98 K): δ −58.30 (s, 6F). Anal. Calcd for C H F N O RuS·2H O:
40
30
6
8
4
2
C, 49.54; N, 11.55; H, 3.53. Found: C, 49.64; N, 11.27; H, 3.51.
Synthesis of TF-13. With the procedure described for TF-11, the
reaction of L3 (137 mg, 0.23 mmol), RuCl ·3H O, (65 mg, 0.25
mmol), 2,6-bis(3-trifluoromethyl-1H-pyrazol-5-yl)pyridine (71 mg,
.20 mmol), and KOAc (100 mg, 1.02 mmol) afforded, after NaOH
3
2
27
successfully synthesized in our previous approach. Efforts
have thus switched to combining both the dicarboxyterpyridine
and dianionic 2,6-bis(3-trifluoromethylpyrazol-5-yl)pyridine
ancillary ligands in the assembly of a new class of bis-tridentate
Ru(II) sensitizers that contain no thiocyanate ligands. In this
study, we report the synthesis and structural and spectroscopic
characterization of four such novel Ru(II) complexes as well as
their performance and properties in DSCs. The prospects for
0
hydrolysis and subsequent acidification, TF-13 as a brown solid
product (142 mg, 63%).
Selected Spectral Data for TF-13. MS (FAB, 102Ru): m/z 993 [M
+
1
+
H] . H NMR (400 MHz, DMSO-d , 298 K): δ 9.31 (s, 1H), 9.15
6
(s, 1H), 9.09 (s, 1H), 8.77 (d, JHH = 8.4 Hz, 1H), 8.22 (t, JHH = 7.2 Hz,
1H), 8.12 (d, JHH = 7.6 Hz, 2H), 7.76 (d, JHH = 5.6 Hz, 1H), 7.72 (s,
2H), 7.68 (d, JHH = 6.0 Hz, 1H), 7.30 (s, 2H), 4.16 (m, 2H), 4.10 (m,
7
489
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496

Journal of the American Chemical Society
Article
2
6
H), 2.54 (t, JHH = 7.2 Hz, 2H), 1.46−1.44 (m, 2H), 1.22−1.18 (m,
Photovoltaic Characterization. Photovoltaic measurements were
recorded with a Newport Oriel class A solar simulator (model 91159)
equipped with a class A 150 W xenon light source powered by a
Newport power supply (model 69907). The light output (area: 2 in. ×
2 in.) was calibrated to AM 1.5 using a Newport Oriel correction filter
to reduce the spectral mismatch in the 350−750 nm region to less
than 4%. The power output of the lamp was measured to 1 Sun (100
H), 0.82 (t, JHH = 6.8 Hz, 3H). 19F NMR (376 MHz, DMSO-d , 298
6
K): δ −55.64 (s, 6F). Anal. Calcd for C H F N O RuS·2H O: C,
42
32
6
8
6
2
49.08; N, 10.90; H, 3.53. Found: C, 49.09; N, 10.89; H, 3.86.
Synthesis of TF-14. With the procedure described for TF-11,
complex TF-14 (124 mg) was obtained in 69% yield from the
sequential reaction of ligand L4, RuCl ·3H O, and 2,6-bis(3-
trifluoromethyl-1H-pyrazol-5-yl)pyridine followed by NaOH hydrol-
ysis and acidification.
3
2
−2
mW cm ) using a certified Si reference cell (SRC−1000−TC−QZ,
VLSI standard, S/N 10510-0031). The current−voltage characteristics
of each cell were obtained by applying an external potential bias to the
cell and measuring the generated photocurrent with a Keithley digital
source meter (model 2400). The incident photon-to-current
conversion efficiency (IPCE) was measured as a function of excitation
wavelength using the incident light from a 300 W xenon lamp (model
6258, Newport Oriel), which was focused through an Oriel
Cornerstone 260 1/4 m monochromator (model 74100) onto the
photovoltaic cell under measurement. The light intensities were
calibrated with a Newport 818 UV detector. Photovoltaic performance
Selected Spectral Data for TF-14. MS (FAB, 102Ru): m/z 993 [M
+
1
+
H] . H NMR (400 MHz, DMSO-d , 298 K): δ 9.21 (s, 1H), 9.16
6
(
s, 1H), 9.02 (d, JHH = 8.0 Hz, 1H), 8.98 (s, 1H), 7.97 (q, JHH = 7.2
Hz, 2H), 7.83 (m, 2H), 7.63 (d, JHH = 4.8 Hz, 1H), 7.29 (d, JHH = 7.2
Hz, 1H), 7.16 (s, 2H), 6.82 (d, JHH = 6.0 Hz, 1H), 4.06 (s, 2H), 3.68
(
0
−
1
s, 2H), 2.39−2.30 (m, 2H), 1.41−1.39 (m, 2H), 1.39−1.25 (m, 6H),
19
.91−0.82 (m, 3H). F NMR (376 MHz, DMSO-d , 298 K): δ
6
58.24 (s, 6F). Anal. Calcd for C H F N O RuS·2H O: C, 49.08; N,
42 32 6 8 6 2
0.90; H, 3.53. Found: C, 49.41; N, 10.59; H, 3.69.
2
DFT Calculations. All of the calculations were performed using the
was measured using a metal mask with an aperture area of 0.36 cm .
28
Gaussian 09 program. For ground-state geometry optimization of
TF-11 and TF-12, density functional theory (DFT) with the B3LYP
Transient Photovoltage and Charge Extraction Measure-
ments. Transient photovoltage (TPV) and charge extraction (CE)
29,30
2
measurements were carried out on optimized 0.16 cm DSC devices
hybrid functional was used.
The basis set consisted of the 6-
1G(d) set for H, C, N, O, and S atoms and the LANL2DZ basis set
for the Ru atom, and a relativistic effective core potential (ECP) was
41
31
using a system similar to that employed by O’Regan et al.
3
Electrical Impedance Spectroscopy Measurements. Electrical
impedance spectroscopy (EIS) experiments were carried out using a
PARSTAT 2273 electrochemical workstation (AMETEK Princeton
32−34
applied to the inner core electrons of Ru.
the absorption spectrum, the vertical excitation from the optimized
ground-state structure was calculated at the time-dependent DFT
For the simulation of
Applied Research, Oak Ridge, TN) with a frequency range of 0.05−
6
1
0 Hz and a potential modulation of 10 mV.
(
TD-DFT) level using the same hybrid functional (B3LYP) and basis
Transient Absorption Spectroscopy Measurements. TAS
set (LANL2DZ for Ru, 6-31G(d) for ligand atoms). All of the
calculations included solvation effects [with N,N-dimethylformamide
2
measurements were carried out on 1 cm DSC devices using a system
4
2
similar to that used by Durrant and co-workers.
Stability Test. The photoanodes of the devices employed in this
(
DMF) as the solvent] using the conductor-like polarizable continuum
35−38
model (C-PCM).
study were composed of a 15 μm transparent TiO thin film and a 5
Device Fabrication. The fluorine-doped tin oxide (FTO) glass
2
μm thick layer of 400 nm TiO particles. A 370 nm cutoff long-pass
used as current collector (3.2 mm thickness, sheet resistance of 9 Ω/
cm , Pilkington) was first cleaned for 30 min in a detergent solution
using an ultrasonic bath and then rinsed with water and ethanol. After
treatment in a UV−O system for 15 min (PSD series UV−ozone
cleaning, Novascan Technologies, Inc.), the FTO glass plates were
immersed in a 40 mM aqueous TiCl solution at 70 °C for 30 min and
then washed with water and ethanol. The photoanodes composed of
nanocrystalline TiO were prepared using literature procedures. For
optimized DSC devices, TiO electrodes with a thickness of 15 μm
were deposited onto transparent conducting glass over which a
scattering layer with a thickness of 7 μm containing 400 nm TiO2
particles (PST-400, JGC Catalysts and Chemicals, Japan) was screen-
printed (0.16 cm active area). For transient absorption spectroscopy
2
2
filter film was attached on the cell surface during illumination. The cell
was irradiated under a Suntest CPS+ lamp (ATLAS GmbH, 100 mW
−
2
cm ) during visible-light soaking at 60 °C. The electrolyte consisted
of 1.0 M 1,3-dimethylimidazolium iodide (DMII), 0.15 M iodine, 0.1
M guanidinium thiocyanate (GNCS), 0.1 M LiI, and 0.5 M N-butyl-
3
4
1
H-benzimidazole (NBB) in 3-methoxypropionitrile (MPN).
3
9
2
RESULTS AND DISCUSSION
Sensitizer Synthesis and Characterization. The chem-
ical structures of the novel TF Ru(II) sensitizers are shown in
Scheme 2. The synthesis of these Ru(II) complexes required
2
■
2
2
(
TAS) measurements, 7 μm thick TiO electrodes (1 cm active area)
2
Scheme 2. Structures of Ru(II) sensitizers TF-11−14
were used. The TiO film thickness was measured by α-step IQ surface
2
profile (KLA Tencor). The TiO electrodes were heated under an air
2
flow at 325 °C for 30 min and then at 375 °C for 5 min, 450 °C for 15
min, and 500 °C for 30 min. The TiO electrodes were treated with a
2
4
0 mM aqueous solution of TiCl , sintered at 70 °C for 30 min, and
4
then washed with water and ethanol. The electrodes were heated again
at 500 °C for 30 min and left to cool to 80 °C before being dipped
into the dye solution (0.3 mM in ethanol with 15 mM
chenodeoxycholic acid as a coadsorbent) for 18 h at 25 °C. Next,
the Pt counter electrodes were spin-coated onto the FTO plates using
a solution of H PtCl (2 mg of Pt in 1 mL isopropyl alcohol) and
2
6
sintered at 400 °C for 15 min. The dye-sensitized TiO electrodes
2
were assembled with the Pt counter electrodes by inserting a hot-melt
Surlyn film (Meltonix 1170-25, 25 μm, Solaronix) as spacer between
the electrodes and then heated at 130 °C. The electrolyte consisted of
0
.6 M 1,2-dimethyl-3-propylimidazolium iodide (DMPII), 0.05 M of
the use of dicarboxyterpyridine chelates with either a 2-
hexylthiophene or 2-hexyl-3,4-ethylenedioxythiophene (2-
hexyl-EDOT) functional group attached at the 5- or 6-position
of one terminal pyridyl unit. Their synthesis was achieved in
reasonable yield from the Pd-catalyzed coupling of diethyl 6-
bromo-2,2′-bipyridine-4,4′-dicarboxylate and the corresponding
iodine, 0.5 M tert-butylpyridine (TBP), and 0.1 M of LiI in
acetonitrile, which was then injected into the cell through a drilled
hole at the counter electrode. Finally, the hole was sealed using a hot-
melt Surlyn film and a cover glass. To reduce scattered light from the
edge of the glass electrodes of the dyed TiO layer, all of the devices
were covered with a light-shading mask with a size of 0.6 cm × 0.6
cm.
2
40
43
tri-n-butyltin reagents (Scheme 3). Diethyl 6-bromo-2,2′-
7
490
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496

Journal of the American Chemical Society
Article
Scheme 3. Simplified Synthetic Protocol for the
Dicarboxyterpyridine Chelates
bipyridine- 4,4′-dicarboxylate was made using a modified
procedure that was originally intended for double bromination
44
of diethyl 2,2′-bipyridine-4,4′-dicarboxylate, while each of the
tri-n-butyltin reagents were prepared using literature methods.
After this, the addition of the dicarboxyterpyridine ligand and
2
,6-bis(3-trifluoromethyl-1H-pyrazol-5-yl)pyridine to the
RuCl ·H O reagent resulted in the formation of the
3
2
ethoxycarbonyl derivatives, and subsequent NaOH-catalyzed
ester hydrolysis afforded the Ru(II) sensitizers TF-11−14 in
high yield. The identities of all of the chelating ligands and the
1
Ru(II) sensitizers were verified by routine mass analysis, H and
1
9
F NMR spectroscopies, and/or elemental analysis. Relevant
synthetic procedures and spectroscopic data for the chromo-
phoric ligands and their intermediates are all given in the
Supporting Information.
The absorption spectra of these TF dyes in DMF solution
are shown in Figure 1a, together with that of N749 for
comparison. Their pertinent photophysical and electrochemical
properties are summarized in Table 1. In contrast to N749 with
its broad absorption band centered around 600 nm, the TF
complexes show narrower and more intense absorption bands
centered at 510 nm. These bands are mainly assigned to the
metal-to-ligand charge-transfer (MLCT) transition to the
dicarboxyterpyridine ligand. Furthermore, for TF-11 and TF-
1
3 there exist intense signals at 363 and 389 nm, respectively,
that are assigned to the intraligand π−π* transition associated
with the added thiophene and EDOT pendant groups. This
assignment is further indicated by the lack of any similar
transitions for TF-12 and TF-14, for which the respective
thiophene and EDOT pendant groups are believed to adopt a
vertical orientation relative to the terpyridine chelate because of
the excessive steric congestion exerted by the adjacent 2,6-
bis(5-pyrazolyl)pyridine chelate.
Figure 1. (a) Absorption spectra of the TF dyes and N749 in DMF.
Further salient differences with respect to the absorption
spectrum of N749 are that all of the TF dyes exhibit broad
absorption at longer wavelengths with peak maxima centered at
(
b) The absorption spectra of TF-11 and TF-12 and the
corresponding optically active electronic transitions with the associated
frontier orbitals (oscillator strength >0.01; red and blue bars are for
TF-11 and TF-12, respectively, while occupied and unoccupied
orbitals are shown in magenta and green, respectively). (c) Absorption
6
85−712 nm and a shoulder extending to 800 nm, which are
3
probably attributable to the spin-forbidden MLCT transition.
Notably, the intense MLCT band at 606 nm in N749 is
substantially suppressed in all of the TF-11−14 dyes, which
could be due to the lack of thiocyanate. These spectral
assignments are further supported by theoretical calculations.
With TF-11 and TF-12 as the paradigm, Figure 1b depicts the
selected optically active electronic transitions obtained from
TD-DFT C-PCM calculations in DMF as the medium. The
corresponding frontier orbitals for some of the major
transitions are also depicted in Figure 1b. The calculated
energy levels, oscillator strengths, and orbital transition analyses
are listed in Tables S1 and S2 and Figure S1 in the Supporting
Information. According to these calculations, the spin density
propagation for the lower-lying electronic transitions (>550
spectra for all samples taken on a 6 μm mesoporous TiO thin film.
2
nm) is mainly derived from the Ru(II) d orbital and bonding
π
orbitals of the pyrazolates to the antibonding orbitals of the
dicarboxyterpyridine chelate. A localized intraligand π−π*
transition appears, as evidenced by the frontier orbital analyses
and high oscillator strength, in the 350−470 nm region,
corresponding to the S and/or higher excited states. Special
6
attention was paid to the absorption bands in the higher-energy
370−430 nm region, for which the frontier orbital analyses
indicate that the electron density is in part distributed around
the thiophene fragment (see the electron−hole densities at 424
and 416 nm for TF-11 and TF-12 respectively, in Figure 1b),
7
491
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496

Journal of the American Chemical Society
Article
Table 1. Photophysical and Electrochemical Data for TF-
1
1−14 and N749
E°
E
E°*
c
ox
(V)
0−0
b
λabs/nm (ε/L mol− cm−1
1
)
a
(V)
(V)
dye
TF-11 363 (35047), 512 (8483), 712 (1456)
0.93
1.68
1.71
1.70
−0.75
−0.80
−0.80
TF-12 319 (34790), 513 (11568), 704 (2228) 0.91
TF-13 319 (44183), 389 (43643), 506 (9918), 0.90
688 (2073)
TF-14 322 (31275), 514 (9817), 685 (2024)
N749 338(24519), 398 (8847), 598 (7330)
0.88
0.89
1.66
1.66
−0.78
−0.77
a
Absorption and emission spectra were measured in DMF solution.
b
E0−0 was determined from the intersection of the absorption peak and
c
the tangent of the emission peak in DMF. E°* was calculated as E° −
E
ox
.
−0
0
manifesting the contribution of the pendant groups to the
harvesting of solar energy and hence the overall DSC
performance (see below).
Cyclic voltammetry was conducted to examine whether the
oxidation potentials of this series of TF sensitizers match the
redox potential of the electrolyte. This would ensure that there
is enough driving energy for dye regeneration to occur and also
verify that the excited-state oxidation potentials, E°*, are more
ox
negative than the conduction band of TiO for efficient electron
2
injection. All of the oxidation potentials were measured in DMF
−1
with 0.1 M TBAPF at a scan rate of 50 mV s (vs NHE). It
6
was calibrated with ferrocene/ferrocenium as the reference and
converted to NHE by addition of 0.63 V. As shown in Table 1,
their electrochemical oxidation potentials E° appeared in the
ox
range 0.93−0.88 V vs NHE, which is notably larger than that of
−
−
the I /I redox couple (ca. 0.4 V vs NHE). Alternatively, the
3
Figure 2. (a) Incident photon-to-current conversion efficiency (IPCE)
spectra and (b) photocurrent density−voltage curves of DSC devices
sensitized with TF dyes and the reference N749.
excited-state oxidation potentials E°*, which ranged from
ox
−
0.75 to −0.80 V, are sufficiently more negative than the
conduction-band edge of the TiO electrode (ca. −0.5 V vs
2
NHE).
Table 2. Performance of DSCs Measured under AM1.5G
Device Characterization. The performance of these
sensitizers in DSC devices was examined. The composition of
the electrolyte (0.6 M DMPII, 0.05 M iodine, 0.5 M TBP, and
a
One-Sun Irradiation
JSC
VOC
(V)
dye loading
−8
dye
(mA cm−2)
FF
η (%)
(10 mol cm−2)c
0
.1 M LiI in acetonitrile) was essentially identical to that
15
employed previously for the N749 sensitizer, except that we
TF-11
TF-12
17.7
19.0
21.5
18.7
17.7
16.1
0.710
0.710
0.710
0.680
0.650
0.750
0.699
0.681
0.683
0.668
0.697
0.694
8.80
9.21
7.53
5.70
reduced the I concentration from 0.1 to 0.05 M with the aim
2
b
of suppressing charge recombination and boosting the device
efficiency. The IPCEs of these DSC dyes are shown in Figure
10.4
TF-13
TF-14
N749
8.53
8.05
8.40
3.99
3.85
3.78
2
a. The steep rise of the IPCE action spectra for all of the TF
complexes started at ∼810 nm, matching well with their
absorption spectra on TiO thin films (see Figure 1c). Excellent
a
2
Except for those with specific remarks, all of the devices were
fabricated using a 15 + 7 μm TiO anode with a 4 mm × 4 mm
IPCE performances were observed in the visible wavelength
region (450−700 nm). It is notable that except for TF-12,
which exhibits a maximum IPCE of 60% between 600 and 720
nm, the other TF sensitizers show inferior IPCE values of
2
working area, an electrolyte that consists of 0.6 M DMPII, 0.05 M I
.5 M TBP, and 0.1 M LiI in acetonitrile. Also, all of the as-prepared
,
2
0
devices were covered by a 6 mm × 6 mm shadow mask for the
b
performance measurement. Measured in the absence of the shadow
∼
50% in this region. For a comparison, DSCs fabricated using
N749 showed a maximum IPCE at 450−610 nm and a steadily
c
mask. The dye loading on 6 μm TiO films was determined by
2
desorbing the dye into a 0.1 M TBAOH solution in 1:1 (v/v) MeOH/
H O and then performing the UV/vis spectral analysis.
2
decreasing IPCE at longer wavelengths and beyond 900 nm
(
see Figure 2a). For the EDOT-functionalized sensitizers TF-
3 and TF-14, although greater extinction coefficients were
observed, both still exhibited lower short-circuit current density
JSC), open-circuit voltage (VOC), and overall conversion
1
47
observed previously, the presence of EDOT groups in the
(
sensitizers can lower VOC because of an acceleration of
efficiency (η) values. It is believed that the inferior efficiencies
are due to the greater molecular sizes, which then prohibit the
recombination between electrons in TiO and the oxidized
electrolyte.
2
effective penetration into the interior of the TiO electrode, as
shown by the slightly reduced dye loading (Table 2). Relevant
Figure 2b shows the photocurrent density−voltage curves of
2
the DSC devices recorded under AM1.5G simulated sunlight at
−2
studies of the trade-off between geometric enlargement and cell
a light intensity of 100 mW cm ; the data is detailed in Table
4
5,46
efficiency have been well-documented.
Furthermore, as we
2. N749 showed J , VOC, fill factor (FF), and η values of 16.1
SC
7
492
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496

Journal of the American Chemical Society
Article
−2
mA cm , 750 mV, 0.694, and 8.40%, respectively. It should be
noted that the conversion efficiency is lower than those
48
49,50
reported by Wang (10.5%) and Han (11.1%).
However,
the TF-sensitized solar cells were endowed with better
−2
efficiencies. For instance, TF-11 gave JSC = 17.7 mA cm ,
VOC = 710 mV, and FF = 0.699, corresponding to an overall
conversion efficiency of η = 8.80%, while the TF-12 sensitizer
−2
gave the best performance data, JSC = 19.0 mA cm , V = 710
OC
mV, FF = 0.681, and η = 9.21%. All these TF sensitizers
−2
showed higher current densities (17.7−19.0 mA cm ) than
−2
N749 (16.1 mA·cm ), reflecting the increased absorption and
light-harvesting efficiency upon introduction of the π-
conjugated pendant groups.
To investigate the recombination processes occurring in
these devices, TPV, CE, EIS, and TAS measurements were
carried out. CE is a tool to measure electron densities in
functioning devices, whereas TPV and EIS are used to measure
the lifetimes of TiO electrons in devices under operational
2
conditions and their recombination with oxidized species
present in the electrolyte. TAS on the other hand can be
used to measure the charge recombination reaction between
photoinjected electrons in TiO and dye cations in the absence
2
of electrolyte and also the regeneration of these dye cations in
the presence of electrolyte.
Differences in VOC between cells can generally be explained
either by shifts in the TiO conduction-band edge (manifested
2
by the shift of the exponential distribution of experimental data
measured by CE) and/or by TiO recombination lifetimes
2
(
investigated via TPV measurements). The CE and TPV data
for devices composed of the TF sensitizers as well as for N749
are shown in Figure 3. The DSC devices showed negligible
differences in charge densities (Figure 3a) as measured by CE,
indicating that the conduction-band potential is similar in all
cases. Therefore, shifts in the Fermi level cannot explain the
differences in voltage observed in these devices. On the other
hand, electron lifetimes measured at identical electron densities
using TPV measurements (Figure 3b) fit well with the cell
voltages listed in Table 2, where longer electron lifetimes
correspond to larger device voltages. From Figure 3b, the
electron lifetimes follow the order N749 > TF-12 > TF-11 >
TF-13 > TF-14.
Figure 3. (a) TiO electron density as a function of voltage deduced
from CE measurements and (b) recombination lifetime τ as a function
2
of TiO electron density deduced from TPV measurements for DSC
2
devices containing TF sensitizers and the reference N749. The cell
voltage was induced via illumination from a series of LEDs.
both the TPV and EIS measurements indicate that devices
based on the EDOT-functionalized sensitizers (TF-13 and TF-
To complement the TPV measurements, EIS measurements
were also conducted to investigate the electron lifetimes in
these devices. Figure 4 shows the Nyquist plots measured
under dark conditions at a forward bias of 0.7 V. Two
semicircles from left to right in the Nyquist plot are visible and
represent the impedances of the charge transfer on the Pt
1
4) have shorter lifetimes than their thiophene-functionalized
counterparts (TF-11 and TF-12). We also observed this in a
counter electrode (R , smaller circles) and the charge
Pt
recombination at the TiO /dye/electrolyte interface (R , larger
2
r
5
1
circles). The radii of the second series of semicircles indicates
Rr to decrease in the order N749 > TF-12 > TF-11 > TF-13 >
TF-14. A smaller R value in theory means faster charge
r
recombination between electrons in TiO2 and electron
acceptors in the electrolyte and thus a shorter TiO electron
2
lifetime. Therefore, the trends observed for the electron
lifetimes extracted from the TPV and EIS measurements are
identical and furthermore are consistent with the trend in VOC
values measured for these devices. We note that a correlation of
the electron lifetime with the cell VOC has also been observed
5
2−54
for DSC devices based on organic donor−acceptor dyes
55
and thiocyanate-containing Ru(II) sensitizers, but we present
such a correlation for the first time here for a series of
thiocyanate-free Ru(II) sensitizers. Finally, we point out that
Figure 4. Electrochemical impedance spectra measured under dark
conditions at a forward bias of 0.7 V for DSC devices containing TF
sensitizers and the reference N749.
7
493
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496

Journal of the American Chemical Society
Article
previous study involving organic dyes that contained similar
functional groups.
TAS measurements investigating the charge recombination
dynamics in DSC devices were performed for the reference dye
N749 and for TF-12 in the presence and absence of electrolyte
difference in regeneration to the rather different molecular
structures of the TF dyes relative to N749. The bulky ancillary
tridentate terpyridine ligand in the TF series would be expected
to hinder the interaction between iodide and Ru(II) where the
cation is centered. For N749, the thiocyanate groups do not
present such an obstacle. Moreover, as already mentioned, the
thiocyanate groups take an active part in regeneration. In any
case, the slower regeneration did not appear to have any
significant effect on device performance, as the TF dyes showed
outstanding efficiencies as demonstrated in Table 2.
47
(
Figure 5). The dynamics for the other TF dyes can be found
58
The insets in Figure 5 show the TAS kinetics for N749 and
TF-12 devices under AM1.5G one-sun illumination (provided
by a series of LEDs) under open-circuit conditions along with
TPV decays recorded under almost identical conditions
following excitation with a laser pulse. It is clear that the
decay of the long-lived TAS signal and the TPV decay
correspond rather well. It is therefore reasonable to conclude
that the long-lived signal in the TAS dynamics is that of
electrons in the TiO2.
The long-term stability of a device is a crucial parameter for
its practical application in DSCs. The device based on TF-12
was fabricated by using a double-layered TiO film (15 + 5 μm)
2
with a low-volatility electrolyte based on MPN as the solvent.
After a 1000 h testing period at 60 °C under one-sun light
soaking, the photovoltaic parameters J , VOC, and FF of the
SC
TF-12-based cell changed only slightly from the initial values
(
Figure 6). As a result, η retained 98% of its initial value; it
Figure 5. Transient absorption kinetics of (a) N749 and (b) TF-12
DSC devices in the presence (red) and absence (black) of redox
electrolyte. The insets show comparisons of the TAS kinetics (red)
and TPV decay (blue), both recorded under AM1.5G one-sun
illumination. All of the TAS and TPV transients were recorded using
laser excitation pulses of 628 nm (N749) or 500 nm (TF-12). The
TAS kinetics was recorded at 800 nm.
Figure 6. Evolution of the solar cell parameters of TF-12 measured
under AM1.5G sunlight soaking at 60 °C. A 405 nm cutoff long-pass
filter was put on the cell surface during illumination. The efficiency
decline was measured to be 2.3%. The electrolyte was 1.0 M DMII,
in the Supporting Information. In the absence of electrolyte,
both N749 and TF-12 showed multiexponential behavior very
similar to that observed previously in studies involving Ru(II)
0
.15 M iodine, 0.1 M GNCS, 0.1 M LiI, and 0.5 M NBB in MPN.
42
polypyridine complexes. In the presence of electrolyte, the
dynamics becomes biphasic as the dye cations are regenerated
by the iodide in the electrolyte, resulting in the appearance of a
•−
56
should be noted that a small drop of 30 mV in VOC is balanced
by an increase in JSC due perhaps to the long-term re-
equilibrium and hence slight redistribution of dyes adsorbed on
long-lived signal assigned to electrons in TiO and/or I2
.
TiO
2
to avoid, in part, aggregation.
2
The dynamics for N749 are similar to those measured
57
CONCLUSION
previously, indicating efficient regeneration. For TF-12,
however, the loss of the cation signal due to regeneration is
not nearly as rapid. This behavior was the same for all of the TF
dyes measured and indicates that regeneration in the case of the
TF dyes is not as efficient as for N749. This difference cannot
be due to reaction driving force, as electrochemical measure-
ments on the TF dyes indicated ground-state potentials similar
to that of N749 (0.8 V vs NHE). We tentatively attribute the
■
To sum up, we have reported a new series of bis-tridentate
Ru(II) sensitizers (TF-11−14) in which a π-conjugated
thiophene pendant group has been strategically added onto
the dicarboxyterpyridine chelating ligand at the 5- or 6-position
of the terminal pyridyl unit. This pendant group serves as an
antenna to harvest solar energy, increasing the light-harvesting
efficiency as demonstrated by absorption and IPCE spectra and
7
494
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496

Journal of the American Chemical Society
Article
verified using DFT calculations. In addition, DSC devices
showed high overall conversion efficiencies. Devices based on
(12) Mishra, A.; Fischer, M. K. R.; Bauerle, P. Angew. Chem., Int. Ed.
2
009, 48, 2474.
−2
(13) Ning, Z.; Fu, Y.; Tian, H. Energy Environ. Sci. 2010, 3, 1170.
TF-12 reached a short-circuit photocurrent of 19.0 mA cm ,
an open-circuit voltage of 0.71 V, and a fill factor of 0.68,
affording an overall conversion efficiency of 9.21%, which is
higher than that of the reference N749 device. A clear
correlation between the device VOC and the cell electron
lifetime under working conditions was also observed for these
sensitizers using both electrical impedance and transient
photovoltage measurements. The designated bis-tridentate
configuration firmly stabilizes the corresponding Ru(II)
sensitizers, as evidenced by the fact that a TF-12 device
retained ≥98% of its initial efficiency after a 1000 h testing
period at 60 °C under one-sun light soaking. We believe that
the degradation even over this time span could be avoided by
improving the sealing process, showing the great prospects of
these designs for use in dye-sensitized solar cells.
(
(
14) Reynal, A.; Palomares, E. Eur. J. Inorg. Chem. 2011, 4509.
15) Nazeeruddin, M. K.; Pechy, P.; Renouard, T.; Zakeeruddin, S.
M.; Humphry-Baker, R.; Comte, P.; Liska, P.; Cevey, L.; Costa, E.;
Shklover, V.; Spiccia, L.; Deacon, G. B.; Bignozzi, C. A.; Gratzel, M. J.
Am. Chem. Soc. 2001, 123, 1613.
16) Wadman, S. H.; Kroon, J. M.; Bakker, K.; Lutz, M.; Spek, A. L.;
̈
(
van Klink, G. P. M.; van Koten, G. Chem. Commun. 2007, 1907.
(17) Wadman, S. H.; Kroon, J. M.; Bakker, K.; Havenith, R. W. A.;
van Klink, G. P. M.; van Koten, G. Organometallics 2010, 29, 1569.
(18) Bessho, T.; Yoneda, E.; Yum, J.-H.; Guglielmi, M.; Tavernelli, I.;
Imai, H.; Rothlisberger, U.; Nazeeruddin, M. K.; Grat
Chem. Soc. 2009, 131, 5930.
19) Bomben, P. G.; Gordon, T. J.; Schott, E.; Berlinguette, C. P.
Angew. Chem., Int. Ed. 2011, 50, 10682.
20) Bomben, P. G.; Theriault, K. D.; Berlinguette, C. P. Eur. J. Inorg.
Chem. 2011, 1806.
21) Robson, K. C. D.; Koivisto, B. D.; Yella, A.; Sporinova, B.;
Nazeeruddin, M. K.; Baumgartner, T.; Gratzel, M.; Berlinguette, C. P.
Inorg. Chem. 2011, 50, 5494.
̈
zel, M. J. Am.
(
(
(
ASSOCIATED CONTENT
Supporting Information
■
̈
*
S
(
(
22) Chi, Y.; Chou, P.-T. Chem. Soc. Rev. 2007, 36, 1421.
23) Chen, B.-S.; Chen, K.; Hong, Y.-H.; Liu, W.-H.; Li, T.-H.; Lai,
Synthetic procedures and the associated spectral data for all of
the chromophoric ligands, figures showing the absorption
spectra and selected electronic transitions of TF-11 and TF-12
obtained using the TD-DFT C-PCM calculations in DMF, and
figures containing TAS and TPV decays. This material is
available free of charge via the Internet at http://pubs.acs.org.
C.-H.; Chou, P.-T.; Chi, Y.; Lee, G.-H. Chem. Commun. 2009, 5844.
24) Chen, K.; Hong, Y.-H.; Chi, Y.; Liu, W.-H.; Chen, B.-S.; Chou,
P.-T. J. Mater. Chem. 2009, 19, 5329.
25) Wu, K.-L.; Hsu, H.-C.; Chen, K.; Chi, Y.; Chung, M.-W.; Liu,
W.-H.; Chou, P.-T. Chem. Commun. 2010, 46, 5124.
26) Chou, C.-C.; Wu, K.-L.; Chi, Y.; Hu, W.-P.; Yu, S. J.; Lee, G.-H.;
Lin, C.-L.; Chou, P.-T. Angew. Chem., Int. Ed. 2011, 50, 2054.
27) Yang, S.-H.; Wu, K.-L.; Chi, Y.; Cheng, Y.-M.; Chou, P.-T.
Angew. Chem., Int. Ed. 2011, 50, 8270.
28) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
(
(
(
AUTHOR INFORMATION
Corresponding Author
ychi@mx.nthu.edu.tw; epalomares@iciq.es; chop@ntu.edu.tw
■
(
(
Author Contributions
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision B.01; Gaussian, Inc.: Wallingford, CT, 2009.
⊥
K.-L.W. and C.-H.L. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Science Council of
Taiwan. We are also grateful to the National Center for High-
Performance Computing for computer time and facilities. E.P.
is grateful for the financial support from ICIQ, ICREA, and the
Spanish MICINN Projects CTQ2010-18859 and CONSOL-
IDER CDS-0007 HOPE-2007 and also thanks the EU for the
ERCstg Polydot and the Catalan Government for the 2009-
SGR-207 Project. J.N.C. thanks the MICINN for the Juan de la
Cierva Fellowship.
(
(
(
(
29) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
30) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
31) Hariharan, P. C.; Pople, J. A. Mol. Phys. 1974, 27, 209.
32) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270.
REFERENCES
1) Peter, L. M. J. Phys. Chem. Lett. 2011, 2, 1861.
2) Clifford, J. N.; Martinez-Ferrero, E.; Viterisi, A.; Palomares, E.
■
(
(
(33) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
(34) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284.
(35) Andzelm, J.; Kolmel, C.; Klamt, A. J. Chem. Phys. 1995, 103,
9312.
(36) Houjou, H.; Inoue, Y.; Sakurai, M. J. Am. Chem. Soc. 1998, 120,
4459.
(37) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 2 1993,
799.
(38) Mineva, T.; Russo, N. Int. J. Quantum Chem. 1997, 61, 665.
Chem. Soc. Rev. 2011, 40, 1635.
3) Yum, J.-H.; Chen, P.; Gra
ChemSusChem 2008, 1, 699.
4) Hamann, T. W.; Jensen, R. A.; Martinson, A. B. F.; Van Ryswyk,
H.; Hupp, J. T. Energy Environ. Sci. 2008, 1, 66.
5) Goncalves, L. M.; de Zea Bermudez, V.; Ribeiro, H. A.; Mendes,
A. M. Energy Environ. Sci. 2008, 1, 655.
(
̈
tzel, M.; Nazeruddin, M. K.
(
(
(
(
(
6) Anandan, S. Sol. Energy Mater. Sol. Cells 2007, 91, 843.
7) Gratzel, M. Inorg. Chem. 2005, 44, 6841.
8) Hagfeldt, A.; Boschloo, G.; Sun, L.; Kloo, L.; Pettersson, H.
(39) Ito, S.; Murakami, T. N.; Comte, P.; Liska, P.; Grat
Nazeeruddin, M. K.; Gratzel, M. Thin Solid Films 2008, 516, 4613.
(40) Ito, S.; Nazeeruddin, K.; Liska, P.; Comte, P.; Charvet, R.;
Pechy, P.; Jirousek, M.; Kay, A.; Zakeeruddin, S. M.; Gratzel, M. Prog.
Photovoltaics. 2006, 14, 589.
̈
zel, C.;
̈
̈
Chem. Rev. 2010, 110, 6595.
̈
(
(
(
̈
9) Gratzel, M. Acc. Chem. Res. 2009, 42, 1788.
10) Robertson, N. Angew. Chem., Int. Ed. 2006, 45, 2338.
11) Ooyama, Y.; Harima, Y. Eur. J. Org. Chem. 2009, 2903.
(41) O’Regan, B. C.; Bakker, K.; Kroeze, J.; Smit, H.; Sommeling, P.;
Durrant, J. R. J. Phys. Chem. B 2006, 110, 17155.
7
495
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496

Journal of the American Chemical Society
Article
(
42) Clifford, J. N.; Palomares, E.; Nazeeruddin, M. K.; Grat
Nelson, J.; Li, X.; Long, N. J.; Durrant, J. R. J. Am. Chem. Soc. 2004,
26, 5225.
43) Mee, S. P. H.; Lee, V.; Baldwin, J. E. Chem.Eur. J. 2005, 11,
294.
44) Barolo, C.; Nazeeruddin, M. K.; Fantacci, S.; Di Censo, D.;
Comte, P.; Liska, P.; Viscardi, G.; Quagliotto, P.; De Angelis, F.; Ito,
S.; Gratzel, M. Inorg. Chem. 2006, 45, 4642.
45) Yu, Q.; Liu, S.; Zhang, M.; Cai, N.; Wang, Y.; Wang, P. J. Phys.
Chem. C 2009, 113, 14559.
46) Kim, J.-J.; Choi, H.; Paek, S.; Kim, C.; Lim, K.; Ju, M.-J.; Kang,
H. S.; Kang, M.-S.; Ko, J. Inorg. Chem. 2011, 50, 11340.
47) Planells, M.; Pelleja, L.; Clifford, J. N.; Pastore, M.; De Angelis,
F.; Lopez, N.; Marder, S. R.; Palomares, E. Energy Environ. Sci. 2011, 4,
820.
48) Gao, F.; Wang, Y.; Zhang, J.; Shi, D.; Wang, M.; Humphry-
Baker, R.; Wang, P.; Zakeeruddin, S. M.; Gratzel, M. Chem. Commun.
008, 2635.
49) Wang, Z.-S.; Yamaguchi, T.; Sugihara, H.; Arakawa, H. Langmuir
005, 21, 4272.
50) Chiba, Y.; Islam, A.; Watanabe, Y.; Komiya, R.; Koide, N.; Han,
L. Jpn. J. Appl. Phys. 2006, 45, L638.
51) Fabregat-Santiago, F.; Bisquert, J.; Garcia-Belmonte, G.;
Boschloo, G.; Hagfeldt, A. Sol. Energy Mater. Sol. Cells 2005, 87, 117.
52) Nishida, J.-i.; Masuko, T.; Cui, Y.; Hara, K.; Shibuya, H.; Ihara,
M.; Hosoyama, T.; Goto, R.; Mori, S.; Yamashita, Y. J. Phys. Chem. C
010, 114, 17920.
53) Marinado, T.; Nonomura, K.; Nissfolk, J.; Karlsson, M. K.;
Hagberg, D. P.; Sun, L.; Mori, S.; Hagfeldt, A. Langmuir 2010, 26,
592.
54) Koumura, N.; Wang, Z.-S.; Miyashita, M.; Uemura, Y.;
Sekiguchi, H.; Cui, Y.; Mori, A.; Mori, S.; Hara, K. J. Mater. Chem.
009, 19, 4829.
55) Matar, F.; Ghaddar, T. H.; Walley, K.; DosSantos, T.; Durrant, J.
R.; O’Regan, B. J. Mater. Chem. 2008, 18, 4246.
56) Montanari, I.; Nelson, J.; Durrant, J. R. J. Phys. Chem. B 2002,
06, 12203.
57) Reynal, A.; Forneli, A.; Palomares, E. Energy Environ. Sci. 2010,
, 805.
58) Privalov, T.; Boschloo, G.; Hagfeldt, A.; Svensson, P. H.; Kloo,
L. J. Phys. Chem. C 2009, 113, 783.
̈
zel, M.;
1
(
3
(
̈
(
(
(
1
(
̈
2
(
2
(
(
(
2
(
2
(
2
(
(
1
(
3
(
7
496
dx.doi.org/10.1021/ja300828f | J. Am. Chem. Soc. 2012, 134, 7488−7496