A simple triaryl amine-based dual functioned co-adsorbent for highly efficient dye-sensitized solar cells

Article abstract of DOI:10.1039/c2jm16021h

We have developed 4-(bis(9,9-dimethyl-9H-flouren-2-yl)amino)benzoic acid (HC-Acid) to use as an alternative co-adsorbent to deoxycholic acid (DCA) in organic dye-sensitized solar cells (DSSCs). It has dual functioned effects, that is, the prevention effec

Full text of DOI:10.1039/c2jm16021h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 3786
www.rsc.org/materials
PAPER
A simple triaryl amine-based dual functioned co-adsorbent for highly efficient
dye-sensitized solar cells†
Hae Min Song, Kang Deuk Seo, Min Soo Kang, In Tack Choi, Sang Kyun Kim, Yu Kyung Eom, Jung Ho Ryu,
*
Myung Jong Ju and Hwan Kyu Kim
Received 20th November 2011, Accepted 14th December 2011
DOI: 10.1039/c2jm16021h
We have developed 4-(bis(9,9-dimethyl-9H-flouren-2-yl)amino)benzoic acid (HC-Acid) to use as an
alternative co-adsorbent to deoxycholic acid (DCA) in organic dye-sensitized solar cells (DSSCs). It has
dual functioned effects, that is, the prevention effect of the p–p stacking of organic dye molecules like
DCA and the light harvesting effect at shorter-wavelength regions. When the HC-Acid was used as the
co-adsorbent onto the TiO2 surface with the organic dye NKX2677 sensitized solar cell, an extremely
high conversion efficiency of 9.09% was achieved under 100 mW cmꢀ2 AM 1.5G simulated light (Jsc
¼
18.01 mA cmꢀ2, Voc ¼ 0.663 V, and FF ¼ 76.16). As we herein demonstrate, an increase in short-circuit
photocurrent density (Jsc) of 20%, open-circuit photovoltage (Voc) of 11% and overall conversion
efficiency (h) of 38% occurred in comparison to the NKX2677 based solar cell. As a consequence, this
new class of a low molecular weight organic co-adsorbent (HC-Acid) is a promising candidate as an
alternative co-adsorbent with a new function not seen in DCA or CA derivatives for highly efficient
dye-sensitized solar cells.
dye/electrolyte interface, several kinds of additives have been
Introduction
introduced as co-adsorbents onto the TiO2 surface to improve
the photovoltaic performance of DSSCs. Deoxycholic acid
(DCA),14 hexadecylmalonic acid (HDMA),15 guanidinium thio-
cyanate,16 tert-butylpyridine,17 and citric acid18 have been added
to dye sensitized solar cells in order to improve the photovoltaic
performance of DSSCs. Moreover, many attempts for achieving
a high power conversion efficiency in DSSCs, for example, co-
sensitization by using multilayered dyes,19 the stepwise co-
sensitization based on organic dyes having complementary
spectral absorption in the visible region,20 molecular cocktails for
efficient panchromatic DSSCs using two dyes,21,22 and light
harvesting effect by co-sensitization with energy relay dyes,23,24
have been widely studied and developed. Furthermore, recently,
to reduce the dye aggregation, an unsymmetrical starburst tri-
phenylamine-based cyanine organic sensitizer25 having a bulky-
shaped structure, di-branched organic sensitizers26 with one
donor and two anchoring groups has been reported.
Dye-sensitized solar cells (DSSCs),1 mainly composed of dye-
sensitizer, wide band gap semiconductor oxide, redox electrolyte
and counter electrode, have been intensively studied and devel-
oped because of their relatively high solar-to-electricity conver-
sion efficiency and low cost.2–11 DSSCs employing Ru-sensitizers
have shown the highest h value of over 11%.12 However, in view
of the cost and environmental demand, organic dyes have
attracted much attention due to their many advantages over Ru-
dyes, such as diversity of molecular structures, high molar
extinction coefficient, and simple synthesis.13
Upon light excitation, initial charge separation occurs in the
adsorbed dye molecules and electrons are injected from their
excited states into the conduction band (CB) of the semi-
conductor. However, excited electrons are injected not only into
the CB of the semiconductor but also into the trap states and
recombine with oxidized dye molecules. In addition, it is possible
for free electrons on a semiconductor surface to recombine with
acceptors such as I3ꢀ ions. Such a charge recombination leads to
losses in both the short-circuit current (Jsc) and the open-circuit
photovoltage (Voc), resulting in a decrease in h. To reduce the
possible charge recombination pathways occurring at the TiO2/
Among the metal-free organic dyes studied in DSSCs,
coumarin-based dyes27 are somewhat promising sensitizers for
DSSCs based on TiO2 films because of their good photoresponse
in the visible region, good long-term stability under one sun
soaking,28 and appropriate lowest unoccupied molecular orbital
(LUMO) levels matching the conduction band of TiO2.
However, p–p stacking of organic dye molecules usually occurs
because of the strong intermolecular interaction. The strong p-
electron interaction in oligothiophene, in particular, allows easy
formation of p-stacks or lamellar structure.29 Although p–p
Center for Advanced Photovoltaic Materials (ITRC), Department of
Advanced Materials Chemistry, Korea University, 208 Seochang,
Jochiwon, Chungnam, 339-700, Korea. E-mail: hkk777@korea.ac.kr
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c2jm16021h
3786 | J. Mater. Chem., 2012, 22, 3786–3794
This journal is ª The Royal Society of Chemistry 2012

View Article Online
stacking is advantageous to light harvesting because of its broad
feature in the UV-vis absorption spectrum, p-stacked aggregates
usually lead to inefficient electron injection and thus results in
low power conversion efficiency.30,31 Prohibition of p–p stacking
with additive in the dye solution is a typical way to improve
efficiency of organic dye-sensitized solar cells suffering from the
p–p stacking problem. Coadsorption of dyes with additives32
and structural modification of dye molecules33 are proven to be
effective to dissociate p–p stacking or dye aggregation and thus
to improve solar cell efficiency.
Very recently, we also reported that a carbazole-based low
molecular weight co-adsorbent34 having multiple functioned
effects, that is, the prevention effect of the p–p stacking, the light
harvesting effect at shorter-wavelength regions and the hole
conducting function through a redox cascade process, was
studied for improving the photovoltaic performance of DSSCs.34
In order to reduce the possible charge recombination path-
ways occurring at the TiO2/dye/electrolyte interface, in this
study, we synthesized a triaryl amine-based 4-(bis(9,9-dimethyl-
9H-flouren-2-yl)amino)benzoic acid (hole conductor-acidified
co-adsorbent: HC-Acid) as an alternative co-adsorbent to DCA.
HC-Acid has a low molecular Y-shaped structure (Scheme 1) and
can also absorb the shorter-wavelength regions so that the light
harvesting effect at shorter-wavelength regions is anticipated to
increase the photocurrent of DSSCs. Also, we propose possible
energy or charge transfer processes of HC-Acid in the TiO2 film
(Scheme 2) used as the co-adsorbent onto the TiO2 surface with
the organic dye NKX2677 sensitized solar cell.
Scheme 2 Schematic diagram of the possible charge or energy transfer
pathways occurring at a dyed TiO2/HC-Acid/electrolyte interface. ET1
means a direct injection of electron from the HC-Acid to the TiO2
surface. ET2 and ET3 correspond to the energy transfer from the HC-
Acid to the dye, and quenching of the HC-Acid by triiodide, respectively.
were carried out with a VersaSTAT 3 (AMETEK) using a typical
three electrode electrochemical cell in a solution with 0.1 M
tetrabutylammonium hexafluorophosphate (TBAPF6) as the
supporting electrolyte in water-free THF at a scan rate of 50 mV
sꢀ1 at room temperature, which was degassed with argon gas for
10 min prior to the scan. Dye-adsorbed TiO2 on conductive glass
was used as the working electrode, a Pt wire as the counter
electrode, and an Ag/AgCl electrode as the reference electrode.
The potential of the reference electrode was calibrated by
ferrocene, and all potentials mentioned in this study are against
the normal hydrogen electrode (NHE). The dye adsorption
amount on the TiO2 surface was estimated spectroscopically
after the dye had been desorbed in a 0.1 M NaOH in 50 : 50
(vol%) EtOH : H2O.
Experimental
Characterizations
1H-NMR spectra were recorded at room temperature with
a Varian Oxford 300 spectrometer and chemical shifts were
reported in ppm units with tetramethylsilane as an internal
standard. 13C NMR spectra were recorded with a BRUKER
DRX 300 instrument. Mass spectra were obtained with a JEOL
JMS-700 spectrometer in EI mode. UV-vis absorption spectra of
dye-loaded films were recorded with a Shimadzu UV-2401PC
spectrophotometer. Cyclic voltammetry (CV) measurements
Materials
2-Iodo-9,9-dimethylfluorene was easily prepared according to
the methods in the literature.35,36 4-Aminobenzonitrile,
Scheme 1 Molecular structures of (a) HC-Acid, (b) DCA, and (c) NKX2677, and the right is a snapshot of HC-Acid derived from a semi-empirical
method (AM1). F: front view and S: side view.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3786–3794 | 3787

View Article Online
1,10-penanthroline and copper iodide were purchased from
Sigma-Aldrich inc. NKX2677 was synthesized according to the
literature.27 Other reagents and chemicals were purchased from
Samchun Chemical Co. Tetrahydrofuran (THF) was dried over
and distilled from sodium and benzophenone under an atmo-
sphere of dry nitrogen, and p-xylene was dried over and distilled
from CaH2 under an atmosphere of dry nitrogen. Other chem-
icals and reagents were used as-received without further
purification.
blade printing and then dried for 2 h at 25 ꢁC. The scattering
layer coated TiO2 electrodes were gradually heated under an air
flow at 325 ꢁC for 5 min, at 375 ꢁC for 5 min, at 450 ꢁC for 15 min
and at 500 ꢁC for 15 min. The resulting layer was composed of an
8 mm thick transparent layer and an 8 mm thick scattering layer,
ꢁ
which were tꢁreated again by TiCl4 at 70 C for 30 min and sin-
tered at 500 C for 30 min. The TiO2 electrodes were immersed
into a 0.3 mM NKX2677 dye solution (with 40 mM DCA or 10
mM HC-Acid added) in a mixture of EtOH and THF (v/v, 2 : 1)
and kept at room temperature for 24 h to complete the sensitizer
uptake. Counter electrodes were prepared by coating with a drop
of H2PtCl6 solution (2 mg of Pt in 1 ml of ethanol) on an FTO
plate. The dye-adsorbed TiO2 electrode and Pt-counter electrode
were assembled into a sealed sandwich-type cell. A drop of the
electrolyte was then introduced into the cell, which was
composed of 0.6 M 1,2-dimethyl-3-propyl imidazolium iodide,
0.05 M iodine, 0.1 M LiI, and 0.5 mM tert-butylpyridine in
CH3CN. It was introduced into the inter-electrode space from
the counter electrode side through predrilled holes. The drilled
holes were sealed with a microscope cover glass slide and Surlyn
to avoid leakage of the electrolyte solution.
Synthesis of 4-(bis(9,9-dimethyl-9H-fluoren-2-yl)amino)
benzonitrile (1)
A mixture of 2-iodo-9,9-dimethylfluorene (7.45 g, 23.28 mmol),
4-aminobenzonitrile (1.25 g, 10.58 mmol), 1,10-penanthroline
(0.19 g, 1.06 mmol), CuI (0.2 g, 1.06 mmol), and K2CO3 (4.39 g,
31.74 mmol) in dry p-xylene (13 ml) was refluxed for 48 h under
a N2 atmosphere. The reaction mixture was quenched with water
and extracted with ethyl acetate. The organic layer was collected,
and the aqueous layer was washed with ethyl acetate. The
combined organic layers were dried over MgSO4, filtered and
concentrated in vacuo. The residue was purified by flash column
chromatography (n-hexane/DCM ¼ 2/3) to afford the title
compound (1) (3.5 g, 66%): 1H NMR (300 MHz, CDCl3) d 7.69–
7.65 (m, 4H), 7.48–7.11 (m, 14H).
Photovoltaic measurements of the solar cells
Photoelectrochemical data were measured using a 1000 W xenon
light source (Oriel, 91193) that was focused to give 1000 W mꢀ2
,
the equivalent of one sun at Air Mass (AM) 1.5G, at the surface
of the cell. The light intensity was adjusted with a Si solar cell that
was double-checked with an NREL-calibrated Si solar cell (PV
Measurement Inc.). The applied potential and measured cell
current were measured using a Keithley model 2400 digital
source meter. The current–voltage characteristics of the cell with
the active area of 0.16 cm2 under these conditions were deter-
mined by biasing the cell externally and measuring the generated
photocurrent. This process was fully automated using Wave-
metrics software. Electrochemical impedance spectroscopy (EIS)
measurements were performed using an impedance analyzer
(VersaSTAT 3, AMETEK) connected to a potentiostat under
dark conditions at room temperature. The spectra were scanned
in a frequency range of 0.1–105 Hz and ac amplitude of 10 mV at
room temperature. The EIS results were analyzed using
ZSimpWin 3.21 software.
Synthesis of 4-(bis(9,9-dimethyl-9H-fluoren-2-yl)amino)benzoic
acid (2)
4-(Bis(9,9-dimethyl-9H-fluoren-2-yl)amino)benzonitrile (2 g,
3.98 mmol) and potassium hydroxide (2.3 g, 41 mmol) in 2-
ethoxyethanol/THF/H2O (2 : 1 : 2, 50 ml) were refluxed for 24 h.
The reaction mixture was allowed to cool at room temperature
and then acidified with 6 M HCl solution until white precipitate
appeared. The precipitates were filtered and dissolved again with
dichloromethane and washed several times with distilled water.
The combined organic layers were dried over MgSO4, filtered
and concentrated in vacuo. The residue was purified by gravity
column chromatography (DCM/MeOH ¼ 1/1) to afford the title
compound HC-Acid (1.8 g, 87%): 1H NMR (300 MHz, DMSO)
d 12.58 (b, 1H), 7.83–7.77 (m, 6H), 7.5 (d, J 6 Hz, 2H), 7.35–7.25
(m, 6H), 7.13 (d, J 6 Hz, 2H), 7.07–6.99 (m, 2H); 13C NMR (75
MHz, CDCl3) d ¼ 167.31, 154.97, 153.27, 151.46, 146.09, 145.68,
138.06, 135.08, 134.57, 130.86, 126.97, 124.37, 123.80, 122.89,
121.26, 119.83, 54.87, 46.49, 26.61 ppm. EI-mass (m/z): calcu-
lated: 521.24 (M + H+), found: 521 (M+).
Results and discussion
Dye absorption behaviors
Fig. 1 shows the absorption and emission spectra of the HC-Acid
in THF solution. The absorption peak (lmax) was observed at 359
nm and the emission peak (lmax) was observed at 450 nm. The
molar extinction coefficient 3 of HC-Acid was estimated to be
36 700 Mꢀ1 cmꢀ1 at 359 nm. Fig. 2(a) shows the UV-visible
absorption spectra of 0.3 mM NKX2677, 0.3 mM NKX2677 +
40 mM DCA, NKX2677 + 10 mM HC-Acid in 8 mm TiO2 films.
NKX2677 exhibited an absorption maximum at 511 nm in
ethanol solution,29 but the absorption peak was blue-shifted to
474 nm for a dye-loaded TiO2 film, obtained by exposing the thin
TiO2 film in dye solution (0.3 mM) until the saturation of dye
adsorption on the TiO2 surface was reached. Such a blue shift is
attributed to the solvent effect and deprotonation of carboxylic
Fabrication of the dye-sensitized solar cells
FTO glass plates (Pilkington) were cleaned in a detergent solu-
tion using an ultrasonic bath for 1 h, rinsed with water and
ethanol. The FTO glass platesꢁwere immersed in an aqueous
solution of 40 mM TiCl4 at 70 C for 30 min and washed with
water and ethanol. The first TiO2 layer of 8 mm thickness was
prepared by doctor blade printing TiO2 paste (Solaronix, Ti-
Nanoxide T/SP) and then dried for 2 h at 25 ꢁC. The TiO2
electrodes were gradually heated under an air flow at 325 ꢁC for 5
min, at 375 ꢁC for 5 min, at 450 ꢁC for 15 min and at 500 ꢁC for
15 min. A paste for the scattering layer containing 400 nm sized
anatase particles (CCIC, PST-400C) was deposited by doctor
3788 | J. Mater. Chem., 2012, 22, 3786–3794
This journal is ª The Royal Society of Chemistry 2012

View Article Online
those of NKX2677 and NKX2677/HC-Acid-based devices,
which may be effectively caused by the breakup of H-aggregation
of NKX2677.
The inset in Fig. 2(a) shows the absorption maxima peak
(lmax) at 385 nm for the TiO2 film solely loaded with HC-Acid.
Compared to the absorption peak of HC-Acid in THF solution,
TiO2 film loaded with HC-Acid shows the maximum absorption
peak to be red-shifted by ꢂ24 nm. This red-shift, favourable for
light harvesting in the longer wavelength of the UV region, can
be anticipated to increase the photocurrent of DSSCs. We also
measured the dye adsorption amount by the dye desorption into
a basic solution followed by spectroscopic determination. As
seen in Fig. 2(b), the dye amount on the TiO2 film decreased from
2.50 ꢃ 10ꢀ7 mol cmꢀ2 to 1.80 ꢃ 10ꢀ7 mol cmꢀ2 and 1.49 ꢃ 10ꢀ7
mol cmꢀ2, leading to a 28% and 41% drop in dye adsorption with
DCA and HC-Acid, respectively. As a result, the Y-shaped
structure of HC-Acid is allowed to be more efficient than DCA in
decreasing dye adsorption. Moreover, as shown in Fig. 2(b), it
indicates that, when comparing the dye amount adsorbed, HC-
Acid adsorbed about 7 times more. Judging by the difference of
structure between DCA and HC-Acid, a coadditive with a bulky
structure may be necessary for effective spacing of adjacent
organic dye molecules. As seen in Scheme 1, the molecular sizes
from the side view of NKX2677, DCA and HC-Acid were esti-
Fig. 1 Absorption and emission spectra of 3.5 ꢃ 10ꢀ5 M HC-Acid in
THF.
ꢀ
ꢀ
ꢀ
mated to be 8.98 A, 6.0 A and 13.37 A, respectively. In the case of
HC-Acid, the dihedral angles of the phenyl group and the fluo-
rene unit with a nitrogen atom as the central, using a semi-
empirical method (AM1), are estimated to be 54ꢁ.
Electrochemical properties
Fig. 3 shows the typical CV of NKX2677 on the TiO2 surface
using 0.1 M TBAPF6 in water-free THF. For comparison,
different dye-loaded films were obtained by dipping the TiO2
films in (a) 0.3 mM NKX2677, (b) 0.3 mM NKX2677 + 40 mM
DCA, and (c) 0.3 mM NKX2677 + 10 mM HC-Acid in THF
overnight, and the inset in Fig. 3 is 10 mM HC-Acid only.
NKX2677 displayed two reversible one-electron anodic
Fig. 2 (a) The absorption spectra of NKX2677 (dotted line), NKX2677/
DCA (dashed dotted line) and NKX2677/HC-Acid (straight line) on the
TiO2 film (thickness 8 mm). The inset is the absorption spectrum of the
TiO2 film loaded with HC-Acid alone. (b) The absorption spectra of
the dye desorbed solution in a 0.1 M NaOH in 50 : 50 (vol%) EtOH : H2O.
acid upon adsorption.32 When 40 mM DCA was included in the
dye solution, the dye-sensitized TiO2 film co-grafted with DCA
showed a similar absorption peak position. When 10 mM HC-
Acid was substituted for DCA on the TiO2 surface, the absorp-
tion peak was red-shifted from 474 to 511 nm compared with
Fig. 3 Cyclic voltammograms of the TiO2 films after immersion for 12 h
in different solutions: (a) 0.3 mM NKX2677, (b) 0.3 mM NKX2677 + 40
mM DCA and (c) 0.3 mM NKX2677 + 10 mM HC-Acid. The inset is CV
of the TiO2 film after immersion for 12 h in 10 mM HC-Acid.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3786–3794 | 3789

View Article Online
Table 1 Energy levels of the NKX2677, NKX2677/DCA, NKX2677/
HC-Acid and HC-Acid alone adsorbed TiO2 electrodes
Electrode
HOMO/Va
LUMO/V
Gap/eVb
NKX2677
0.92
0.92
0.91
1.11
ꢀ0.90
ꢀ0.91
ꢀ0.89
ꢀ1.06
1.82
1.85
1.85
2.17
NKX2677/DCA
NKX2677/HC-Acid
HC-Acid
a
The first oxidation potential vs. NHE in Fig. 3 was adopted as the
HOMO. The absorption threshold in Fig. 2(a) was taken as the gap
between the HOMO and the LUMO.
b
oxidation steps at half-wave potentials of 0.92 and 1.39 V.
Similar to NKX2677, a co-adsorbent DCA electrode exhibits
two reversible couples at 0.89 and 1.33 V. In the case of HC-Acid
coadsorption, the first oxidation potential is 1.12 V and the
second is 1.84 V, which is similar to HC-Acid alone. The lowest
unoccupied molecular orbital (LUMO) level is determined by
using the expression LUMO ¼ HOMO ꢀ gap, where the gap is
estimated from the absorption threshold in Fig. 2(a). The energy
levels of these three electrodes are summarized in Table 1. The
NKX2677 adsorbed to the TiO2 films can be electrochemically
oxidized and reduced in a reversible pattern. Generally, initial
oxidation of NKX2677 molecules followed by the electron
hopping across the TiO2 nanoparticle surface is one accepted
mechanism,37–40 for charge transfer in a working electrode. The
formation of p–p stacking of NKX2677, due to the increase of
the dye adsorbed amount, was also evidenced by the strong
current intensity signal as shown in Fig. 3. That is, after the
coadsorption of HC-Acid with the dye on the TiO2 surface, the
significant decrease in current intensity in the CV signal suggests
that HC-Acid more effectively decreases the dye adsorbed
amount than DCA. This different behavior in electrochemical
activity between the two electrodes, i.e., NKX2677/DCA and
NKX2677/HC-Acid, indicates that the coadsorption of HC-Acid
would be more favorable than DCA for suppressing the recom-
bination of electrons with electron acceptors in the electrolyte
I3ꢀ/I2 or preventing possible intermolecular electron losses
between two neighboring dye molecules in a DSSC device. In
other words, the coadsorption of HC-Acid would be more
effective than DCA for forming an insulting spacer among dye
molecules, which blocked the recombination of injected electrons
Fig. 4 IPCE spectra of DSSCs based on (a) NKX2677, (b) NKX2677/
DCA and (c) NKX2677/HC-Acid. The inset is the IPCE spectrum of the
TiO2 film loaded with HC-Acid alone.
Acid-based solar cell was ca. 45% at 360 nm. These results
indicate that it has the good light harvesting effect at shorter-
wavelength regions (see Table 2). As shown in Fig. 4, the
maximum IPCE is only ꢂ77% for NKX2677 sensitized solar
cells. It is much lower than 85%, taking into account the light loss
(ꢂ15%) by the reflection and absorption of the TCO glass. Upon
addition of DCA or HC-Acid in the dye solution, IPCE was
increased remarkably. The maximum IPCE increased from 73%
to 77% when 40 mM DCA was added in the dye solution, and
increased to 83% with 10 mM HC-Acid in the dye solution. Upon
HC-Acid coadsorption, the maximum IPCE improved remark-
ably, up to 67% at 360 nm, compared to those of the other two
devices, which could be contributed to the increased photocur-
rent of the DSSCs. Taking the ꢂ85% transmittance of the TCO
glass into account, the IPCE value for TiO2 film HC-Acid is
almost 100%, even without anti-reflecting coating film. Because
dye monomer can inject electrons into TiO2 more efficiently than
aggregate,31 the coadsorbed dye/DCA TiO2 film obtained from
the exposure of TiO2 film to the dye solution with 40 mM DCA
may contain a mixture of aggregate and dye monomer judged
from the IPCE spectrum. On the other hand, the coadsorbed dye/
HC-Acid TiO2 film obtained from exposure of TiO2 film to the
dye solution with 10 mM HC-Acid is mainly composed of dye
monomer, as judged from the high IPCE. Moreover, a plateau
from 500 to 600 nm was observed in the IPCE action spectrum
with IPCE of ꢂ85% on the dye/HC-Acid TiO2 film. It is evi-
denced that HC-Acid is effective to dissociate formation of p–p
stacking by the strong intermolecular interaction. This high
IPCE value means that all the three efficiencies, such as electron
injection, light harvesting, and electron-collection efficiency, are
unity.
ꢀ
with I3 ions in the electrolyte and/or the oxidized dye cations
generated after electron injection in a DSSC device.
Solar cell performances
Incident photon-to-electron conversion efficiency (IPCE) as
a function of wavelength was measured to evaluate the photo-
response of the photoelectrodes in the whole spectral region.
Fig. 4 shows the IPCE spectra for TiO2 films before and after
coadsorption with DCA or HC-Acid. The inset in Fig. 4 shows
the IPCE spectrum for TiO2 film loaded HC-Acid alone. The
photovoltaic performance of the HC-Acid-based solar cell was
obtained with a short circuit current (Jsc) of 2.57 mA cmꢀ2, an
open circuit voltage (Voc) of 655 mV, a fill factor (FF) of 71%,
and an overall energy conversion efficiency (h) of 1.21% under
standard AM 1.5 conditions. The maximum IPCE of the HC-
The lower charge collection efficiency or lower IPCE is
ascribed to charge recombination caused by dye aggregation or
close p–p stacking.40,41 Wang et al. reported that the coad-
sorption of DPA40,42 or HDMA43 with ruthenium polypyridine
dyes could improve both Jsc and Voc. However, DPA or HDMA
coadsorbs parallelly to the dye molecules and the linear alkyl
chain cannot separate adjacent molecules efficiently and thus
3790 | J. Mater. Chem., 2012, 22, 3786–3794
This journal is ª The Royal Society of Chemistry 2012

View Article Online
a
Table 2 Photovoltaic performance parameters of the DSSCs with different co-adsorbents under AM 1.5G (100 mW cmꢀ2
)
Dye adsorbed
amount/10ꢀ7 mol cmꢀ2
Device
Jsc/mA cmꢀ2
Voc/mV
FF (%)
h (%)
NKX2677
2.51
1.81
1.72
1.49
3.41
14.96
15.54
17.61
18.01
2.57
595
640
663
663
655
74.05
74.47
75.70
76.16
71.47
6.59
7.41
8.83
9.09
1.21
NKX2677/DCA
NKX2677/HC-Acidb
NKX2677/HC-Acidc
HC-Acidc
a
Performance of DSSCs was measured with a black metal mask (0.16 cm2). Electrolyte: 0.6 M DMPII, 0.05 M I2, and 0.1 M LiI in CH3CN. b HC-Acid
concentration was 5 mM. c HC-Acid concentration was 10 mM.
cannot prohibit the p–p stacking. Because of this, it was revealed
that they could not work on coumarin dye of NKX2700 in terms
of efficiency improvement contrary to the DCA.38 Moreover,
recently, di-branched organic sensitizers26 constituted by one
donor and two anchoring groups having a bulky-shaped struc-
ture and a novel unsymmetrical organic cyanine sensitizer25 that
contain the tailored starburst triphenylamine moiety acting as an
electron donor have been reported for suppressing aggregation,
disfavouring molecular stacking between dye molecules resulted
in improving the photovoltaic performance in DSSCs. As
mentioned above, for effective spacing of adjacent molecules and
efficiently prohibiting the p–p stacking, a coadditive having
a bulky-shaped structure may be necessary. As illustrated in
Scheme 1, the chemical structure of HC-Acid has a more bulky-
shaped structure than DCA. Thus, HC-Acid can act as a more
effective spacer than DCA among dye molecules and thus
suppresses the p–p interaction of the dye molecules, which
retards charge recombination and hence significantly improves
the performance of the DSSCs. Moreover, as shown in Table 1,
the LUMO of HC-Acid is slightly higher than that of NKX2677,
the electron from excited NKX2677 dye cannot directly move to
HC-Acid, in addition, it has a light harvesting effect at shorter-
wavelength regions as seen in IPCE data. In other words, it has
dual functioned effects, in which it can act as spacers between the
dye molecules and have the light harvesting effect in the shorter-
wavelength regions.
The current density–voltage (J–V) curves are shown in Fig. 5.
Their photovoltaic performance is summarized in Table 2. The
photocurrent density (Jsc) increased in the same order of inte-
grated photocurrent as the IPCE action spectra in Fig. 4. The
NKX2677-based DSSC produced h ¼ 6.59%, Jsc ¼ 14.96 mA
cmꢀ2, Voc ¼ 0.595 V, and FF ¼ 74.05. The NKX2677/DCA-
based DSSC produced h ¼ 7.41%, Jsc ¼ 15.54 mA cmꢀ2, Voc
¼
0.640 V, and FF ¼ 74.47. NKX2677/HC-Acid-based produced
h ¼ 9.09%, Jsc ¼ 18.01 mA cmꢀ2, Voc ¼ 0.663 V and FF ¼ 76.16.
HC-Acid only provides h ¼ 1.21%, Jsc ¼ 2.57 mA cmꢀ2, Voc
¼
0.656 V, and FF ¼ 71.47. Obviously, coadditive exchange from
DCA to HC-Acid results in high efficiency, which is attributed to
the significant enhancement of Jsc and Voc mainly originating
from its dual functioned effect. The above photovoltaic perfor-
mance results indicate that coadsorption with HC-Acid is more
effective in improving solar cell performance than DCA.
Effect of co-additives on the open-circuit voltage of DSSCs
Generally, in the DSSCs, Voc is defined as the voltage difference
between the chemical potential of the electrolyte redox (Ered) and
quasi-Fermi potential of electrons (EFn) in the semiconductor,
TiO2, which could be described as eqn (1) (ref. 44):
ꢀ
ꢁ
kBT
e
Ne
n
Voc ¼ Ered ꢀ ECB ꢀ g
ln
(1)
where g is the characteristic constant of TiO2 tailing states, kB is
the Boltzmann constant, T is the temperature, e is the elementary
charge, and Ne is the effective density of states at the TiO2
conduction band edge. Assuming that Ered is constant, Voc will
be determined by the position of the conduction band edge (ECB
)
and the concentration of electrons (n) in TiO2. These two
parameters are closely related to the surface charge and charge
recombination, respectively. The position of the conduction
band edge will be determined by the surface charge supplied by
interface species such as ions, additives, and dye molecules. Any
change in the surface charge will shift the conduction band edge
position and influence the Voc. The number of electrons (n) in
TiO2 is determined by the balance between electron injection and
electron recombination. Possible factors influencing Voc for
DSSCs based on NKX2677/DCA and NKX2677/HC-Acid
involve the TiO2 surface blocking and conduction band edge
shift, which will be discussed in detail in the following sections.
Electrochemical impedance spectroscopy (EIS) is a powerful
tool for identifying electronic and ionic transport processes in
DSSCs, which provides valuable information for the
Fig. 5 J–V curves of DSSCs based on (a) NKX2677, (b) NKX2677/
DCA and (c) NKX2677/HC-Acid. The inset is the J–V curve of HC-Acid
alone, and the TiO2 film thickness and active area of 16(8 + 8) mm and
0.16 cm2, respectively.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3786–3794 | 3791

View Article Online
NKX2677/DCA (34 U) < NKX2677/HC-Acid (39 U), which is
consistent with the Voc values measured in the devices. The
smaller Rct value means the electron recombination from the
TiO2 to electron acceptors in the electrolyte occurring more
easily, and thus results in lower Voc. Clearly, electron recombi-
nation based on NKX2677 is faster than that of NKX2677/DCA
and NKX2677/HC-Acid. The coadsorption leads to the forma-
tion of a more compact monolayer, comprised of the dye and
DCA or HC-Acid as the co-adsorbent, than the adsorption of
dye alone. This insulating molecular layer effectively shields the
ꢀ
back electron transfer from the TiO2 to I3 ions and gives
a higher Voc. In this study, a coumarin organic sensitizer
NKX2677 is coadsorbed with HC-Acid, a combination that
enhances the Voc by 68 mV and increases the overall conversion
efficiency by 38%.
To investigate the influences of the shift in the TiO2 conduc-
tion band edge and charge recombination on Voc, impedance
spectra were also measured by varying the applied potential at
equal intervals in the vicinity of Voc. By fitting the EIS curves,
another important parameter for DSSCs, electron lifetime (s),
could be extracted from the Cm and Rct using s ¼ CmRct.48,49 The
fitted Cm and Rct values together with s are plotted in Fig. 7. This
exponential rise with the increase of forward bias is a typical
behaviour of Cm that is described by eqn (2) (ref. 50):
ꢂ
ꢃ
e2
a
Cm ¼
exp
ðEred þ eVa ꢀ ECB
Þ
(2)
kBT
kBT
where e is the elementary charge, kB is the Boltzmann constant, T
is the temperature (298 K), a is a constant related to the distri-
bution of electronic states below the conduction band, and Ered is
the chemical redox potential of the redox couples in the elec-
trolyte. Assuming that Ered is constant, Cm is governed by the
forward bias (Va) and ECB. As shown in Fig. 7(a), at a given value
of Va the Cm for both NKX2677/HC-Acid and NKX2677/DCA
is higher than that of NKX2677, indicating a negative shift of the
conduction band edge. Generally, the negative shift of ECB leads
to improvement of Voc of DSSCs. Cm between the NKX2677/
HC-Acid and the NKX2677/DCA has almost similar value, but
Rct of the NKX2677/HC-Acid is slightly higher than that of the
NKX2677/DCA. This means that it is charge recombination,
rather than the position of conduction band edge, that governs
Voc. At the forward bias of ꢀ0.65 V, the fitted s increases in the
order of NKX2677/HC-Acid (21 ms) > NKX2677/DCA (34 ms)
> NKX2677 (39 ms), giving an explanation of the different Voc
yielded by DSSCs based on the three devices. One electron loss
Fig. 6 (a) Nyquist plots and (b) Bode phase plots of electrochemical
impedance spectra for the DSSCs based on NKX2677 (open squares),
NKX2677/DCA (open circles) and NKX2677/HC-Acid (open triangles).
The inset is the equivalent circuit used to fit the impedance spectra.
understanding of photovoltaic parameters.45–47 Fig. 6 shows the
typical EIS Nyquist and Bode phase plots measured under the
dark conditions at a forward bias of ꢀ0.65 V. The equivalent
circuit (inset) in Fig. 6(b) was used to fit the experimental data of
all of the samples. Rs is the series resistance accounting for the
transport resistance of the TCO and the electrolyte. Cm and Rct
are the chemical capacitance and the charge recombination
resistance at the TiO2/electrolyte interface, respectively. CPt and
RPt are the interface capacitance and charge transport resistance
at the Pt/electrolyte interface, respectively. The larger semicircle
at the middle frequency range represents the interfacial charge
transfer resistance (Rct) at the dyed TiO2/electrolyte interface.
The fitted Rct increased in the order of NKX2677 (21 U) <
Fig. 7 (a) Chemical capacitance (Cm), (b) interfacial charge transfer resistance (Rct), and (c) electron lifetime (s) fitted from impedance spectra under
a series of different forward bias: NKX2677 (filled squares), NKX2677/DCA (open circles) and NKX2677/HC-Acid (filled triangles).
3792 | J. Mater. Chem., 2012, 22, 3786–3794
This journal is ª The Royal Society of Chemistry 2012

View Article Online
channel is the electron capture by electron acceptors, i.e., I3ꢀ/I2
ions in the electrolyte. The insertion of HC-Acid, forming an
insulating layer, blocks the electron recombination to the I3ꢀ ions
and thus enhances the electron lifetime. The longer electron
lifetime contributed to a higher Voc. However, the electron life-
time in TiO2 was not significantly enhanced, compared to that of
another HC-Acid in previous article reported by our group,34
which has multiple functioned effects, such as the prevention
effect of the p–p stacking, the light harvesting effect at shorter-
wavelength regions and the hole conducting function through
a redox cascade process. It indicates that the present HC-acid has
dual-functioned effects, such as the prevention effect of the p–p
stacking and the light harvesting effect at shorter-wavelength
regions.
Fig. S1†), and the separation distance between the donor of HC-
Acid and the acceptor of NKX2677 dye may be sufficiently
nearby in the solution state.
In this study, under the similar solution concentrations of each
solution estimated from real dye-sensitized solar cells fabricated
here, as depicted in Fig. 8, the increase of the emission spectrum
of NKX2677 over 600 nm was not observed, may be due to the
fact that the molar excitation coefficient of NKX2677 has an
extremely low value of 1800 Mꢀ1 cmꢀ1 at 354 nm, and its PL
quantum efficiency is also very low at 354 nm (see Fig. S2†). The
excitation spectrum profile of a mixture of HC-Acid and
NKX2677 (lem ¼ 670 nm) is similar to those of HC-Acid and
NKX2677 (see Fig. S3†). From this result, it is presumed that the
energy transfer from HC-Acid to NKX2677 happens to be low as
follows: the molar excitation coefficient of NKX2677 is very low
in the range of 1800–7400 Mꢀ1 cmꢀ1 at the wavelength region of
350–390 nm (see Fig. S2†), so that the energy transfer efficiency
from HC-Acid to NKX2677 should be expected to be low
(pathway ET2 in Scheme 2). Triiodide is a highly mobile ion that
is known as a ‘perfect quencher’, meaning that collisions with the
HC-Acid have a near unity probability of quenching the excited
Photophysical properties and possible charge transfer mechanism
For further understanding on the cell performance enhancement
by HC-acid, we have investigated the photophysical properties of
HC-Acid and NKX2677 in the solution state (see the ESI† for
details). Fig. 8 shows the PL emission spectra with different
conditions at the excitation wavelength (lex) of 354 nm in
acetonitrile. Each solution was made to similar concentrations
estimated from real dye-sensitized solar cells fabricated here (the
mole ratio of HC-Acid/NKX2677/I2 ¼ 1.5/1/2). Upon the
photoexcitation at 354 nm, the PL maximum wavelength of HC-
Acid, NKX2677 and I2 appeared at 419, 430 and 428 nm,
respectively. The PL emission intensities of both NKX2677 and
I2 are extremely lower than that of HC-Acid. As shown in Fig. 8,
the PL emission maxima of both the mixture of HC-Acid/
NKX2677 and the mixture of HC-Acid/NKX2677/I2 appeared
around 490 nm, and the PL emission intensity of HC-Acid was
significantly diminished about 51% and 56% in the system of HC-
Acid/NKX2677 and HC-Acid/NKX2677/I2, respectively.
ꢀ
state.23,52 This quenching effect by I3 ions appeared to be rela-
tively lower about 5% in our system (pathway ET3 in Scheme 2).
For a real dye-sensitized solar cell based on HC-Acid, the energy
transfer efficiencies of ET2 and ET3 seem to be very low, due to
the molecular size difference between NKX2677 and HC-Acid.
Thus, HC-Acid cannot extremely influence the photovoltaic
performance enhancement through the energy transfer pathways
ꢀ
of HC-Acid into NKX2677 and Iꢀ/I3 in the real dye-sensitized
solar cell systems. In other words, the electron injection pathway
of ET1 is dominant over the other energy transfer pathways of
ET2 and ET3 in Scheme 2.
Conclusions
Once excited, the HC-Acid can transfer its excited energy to
a certain acceptor,21,22 and the energy transfer rate is affected as a
function of the separation distance (r) between HC-Acid and
a nearby acceptor molecule (SD).23,51 HC-Acid is assumed to be
The role of 4-(bis(9,9-dimethyl-9H-flouren-2-yl)amino)benzoic
acid as the co-adsorbent on the TiO2 surface with NKX2677
sensitizer was investigated. To confirm the role of HC-Acid, we
compared it with a deoxycholic acid (DCA). Co-grafted DCA or
HC-Acid on the TiO2 surface reduced the adsorption amount of
NKX2677 and diminished the H-aggregation formation between
the dye molecules onto TiO2 as seen from the IPCE and
absorption spectral data on the TiO2 film. HC-Acid reduced the
recombination process at the TiO2 surface acting like a blocking
layer to improve Voc by slowing recombination, which man-
ifested itself in a longer electron lifetime. Moreover, the breakup
of p-stacked aggregates might improve the electron injection
yield, thus increasing Jsc. In other words, it has dual functioned
effects, which are the prevention effect of the p–p stacking of
organic dye molecules such as the DCA and the light harvesting
effect at shorter-wavelength regions. With the HC-Acid used as
the co-adsorbent for the organic dye NKX2677 sensitized solar
cell, an extremely high power conversion efficiency of 9.09% was
€
within the Forster radius (Ro) of the NKX2677 dye (see
achieved under 100 mW cmꢀ2 AM 1.5G simulated light (Jsc
¼
18.01 mA cmꢀ2, Voc ¼ 0.663 V, and FF ¼ 76.16). Thus, the dual
functioned coadsorbent is a promising candidate as a new co-
additive for highly efficient dye-sensitized solar cells to improve
UV light harvesting through a charge transfer process.
Fig. 8 PL emission spectra for (a) 9.9 ꢃ 10ꢀ6 M HC-Acid, (b) 6.6 ꢃ 10ꢀ6
M NKX2677, (c) 1.25 ꢃ 10ꢀ5 M I2, (d) 9.9 ꢃ 10ꢀ6 M HC-Acid + 6.6 ꢃ
10ꢀ6 M NKX2677, and (e) 9.9 ꢃ 10ꢀ6 M HC-Acid + 6.6 ꢃ 10ꢀ6
NKX2677 + 1.25 ꢃ 10ꢀ5 M I2 at lex of 354 nm in acetonitrile.
M
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 3786–3794 | 3793

View Article Online
4680; (b) A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran,
Md. K. Nazeerudin, E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeerudin
Acknowledgements
€
and M. Gratzel, Science, 2011, 334, 629.
23 B. E. Hardin, E. T. Hoke1, P. B. Armstrong, J.-H. Yum, P. Comte,
This research was supported by MKE (The Ministry of Knowledge
Economy), Korea, under the ITRC support program supervised by
the IITA (Institute for Information Technology Advancement
(IITA-2008-C1090-0804-0013)), WCU (The Ministry of Education
and Science) program (R31-2008-000-10035-0) and by the
Converging Research Centre Program through the National
Research Foundation of Korea (2010K000973).
ꢁ
T. Torres, J. M. J. Frechet, M. K. Nazeeruddin, M. Gratzel and
M. D. McGehee, Nat. Photonics, 2009, 3, 406.
€
24 B. E. Hardin, A. Sellinger, T. Moehl, R. Humphry-Baker,
€
J.-E. Moser, P. Wang, S. M. Zakeeruddin, M. Gratzel and
M. D. McGehee, J. Am. Chem. Soc., 2011, 133, 10662.
25 J. Tang, W. Wu, J. Hua, J. Li, X. Li and H. Tian, Energy Environ. Sci.,
2009, 2, 982.
26 A. Abbotto, N. Manfredi, C. Marinzi, F. D. Angelis, E. Mosconi,
€
J.-H. Yum, Z. Xianxi, M. K. Nazeeruddin and M. Gratzel, Energy
Environ. Sci., 2009, 2, 1094.
Notes and references
27 Z.-S. Wang, Y. Cui, K. Hara, Y. Dan-oh, C. Kasada and A. Shinpo,
Adv. Mater., 2007, 19, 1138.
28 D. E. Janzen, M. W. Burand, P. C. Ewbank, T. M. Pappenfus,
€
1 B. O’Regan and M. Gratzel, Nature, 1991, 353, 737.
2 M. Gratzel, Nature, 2001, 414, 338.
€
3 P. Wang, C. Klein, R. Humphry-Baker, S. M. Zakeeruddin and
€
ꢁ
H. Higuchi, D. A. da Silva Filho, V. G. Young, J.-L. Bredas and
M. Gratzel, J. Am. Chem. Soc., 2005, 127, 808.
4 S. R. Song, C. Lee, H. Choi, J. Ko, J. Lee, R. Vittal and K. J. Kim,
Chem. Mater., 2006, 18, 5604.
K. R. Mann, J. Am. Chem. Soc., 2004, 126, 15295.
29 A. C. Khazraji, S. Hotchandani, S. Das and P. V. Kamat, J. Phys.
Chem. B, 1999, 103, 4693.
5 (a) N. Robertson, Angew. Chem., 2006, 118, 2398; (b) N. Robertson,
Angew. Chem., Int. Ed., 2006, 45, 2338.
€
30 A. Kay and M. Gratzel, J. Phys. Chem., 1993, 97, 6272.
31 K. Hara, Y. Dan-oh, C. Kasada, Y. Ohga, A. Shinpo, S. Suga,
K. Sayama and H. Arakawa, Langmuir, 2004, 20, 4205.
32 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.
33 K. Hara, M. Kurashige, Y. Dan-oh, C. Kasada, A. Shinpo, S. Suga,
K. Sayama and H. Arakawa, New J. Chem., 2003, 27, 783.
34 B. J. Song, H. M. Song, I. T. Choi, S. K. Kim, K. D. Seo, M. S. Kang,
M. J. Lee, D. W. Cho, M. J. Ju and H. K. Kim, Chem.–Eur. J., 2011,
17, 11115.
35 C.-H. Huang, S.-H. Yang, K.-B. Chen and C.-S. Hsu, J. Polym. Sci.,
Part A: Polym. Chem., 2006, 44, 519.
36 S. H. Lee, T. Nakamura and T. Tsutsui, Org. Lett., 2001, 3,
2005.
37 D. P. Hagberg, T. Edvinsson, T. Marinado, G. Boschloo, A. Hagfeldt
and L. Sun, Chem. Commun., 2006, 2245.
38 P. Bonhote, E. Cogniat, S. Tingry, C. Barbe, N. Vlachopoulos,
6 A. Islam, F. A. Chowdhury, Y. Chiba, R. Komiya, N. Fuke,
N. Ikeda, K. Nozaki and L. Han, Chem. Mater., 2006, 18, 5178.
7 M. S. Kang, J. B. Oh, K. D. Seo, H. K. Kim, J. Park, K. Kim and
N.-G. Park, J. Porphyrins Phthalocyanines, 2009, 13, 798.
8 M. J. Lee, K. D. Seo, H. M. Song, M. S. Kang, Y. K. Eom,
H. S. Kang and H. K. Kim, Tetrahedron Lett., 2011, 52,
3879.
9 K. D. Seo, H. M. Song, M. J. Lee, M. Pastore, C. Anselmi,
€
F. D. Angelis, M. K. Nazeeruddin, M. Gratzel and H. K. Kim,
Dyes Pigm., 2011, 90, 304.
10 D. H. Lee, M. J. Lee, H. M. Song, B. J. Song, K. D. Seo, M. Pastore,
C. Anselmi, S. Fantacci, F. D. Angelis, M. K. Nazeeruddin and
H. K. Kim, Dyes Pigm., 2011, 91, 192.
11 N. S. Baek, J.-H. Yum, X. Zhang, H. K. Kim, M. K. Nazeeruddin
€
and M. Gratzel, Energy Environ. Sci., 2009, 2, 1082.
12 Z. Wang, M. Yanagida, K. Sayama and H. Sugihara, Chem. Mater.,
2006, 18, 2912.
13 M. K. Nazeerudin, A. Kay, I. Rodicio, R. Humphry-Baker,
€
F. Lenzmann, P. Comte and M. Gratzel, J. Phys. Chem. B, 1998,
102, 1498.
39 T. A. Heimer, S. T. D’Arcangelis, F. Farzad, J. M. Stipkala and
G. J. Meyer, Inorg. Chem., 1996, 35, 5319.
40 Z.-S. Wang, Y. Cui, Y. Dan-ho, C. Kasada, A. Shinpo and K. Hara,
J. Phys. Chem. C, 2007, 111, 7224.
41 X. Ren, Q. Feng, G. Zhou, C.-H. Huang and Z.-S. Wang, J. Phys.
Chem. C, 2010, 114, 7190.
42 P. Wang, S. M. Zakeeruddin, R. Humphry-Baker, J. E. Moser and
€
€
E. Muller, P. Liska, N. Viachopoulos and M. Gratzel, J. Am.
Chem. Soc., 1993, 115, 6382.
ꢁ
14 (a) A. Moraderia, I. Lopez-Duarte, B. O’Regan, M. V. Martınez-
Dıaz, A. Forneli, E. Palomares, T. Torres and J. R. Durrant, J.
ꢁ
ꢁ
Mater. Chem., 2009, 19, 5016; (b) X. Ren, Q. Feng, G. Zhou,
C. H. Huang and Z. S. Wang, J. Phys. Chem. C, 2010, 114,
7190.
15 M. K. Nazeeruddin, F. D. Angelis, S. Fantacci, A. Selloni,
€
G. Viscardi, P. Liska, S. Ito, B. Takeru and M. Gratzel, J. Am.
Chem. Soc., 2005, 127, 16835.
16 F. Gao, Y. Wang, D. Shi, J. Zhang, M. Wang, X. Jing, R. Humphry-
€
M. Gratzel, Adv. Mater., 2003, 15, 2101.
43 P. Wang, S. M. Zakeeruddin, P. Comte, R. Charvet, R. Humphry-
€
Baker and M. Gratzel, J. Phys. Chem. B, 2003, 107, 14336.
44 A. Uasmi, S. Seki, Y. Mita, H. Kobayashi, H. Miyashiro and
N. Terada, Sol. Energy Mater. Sol. Cells, 2009, 93, 840.
45 R. Kern, R. Sastrawan, J. Ferber, R. Stangl and J. Luther,
Electrochim. Acta, 2002, 47, 4213.
46 (a) J. Bisquert, J. Phys. Chem. B, 2002, 106, 325; (b) J. Bisquert, Phys.
Chem. Chem. Phys., 2003, 5, 5360.
47 F. Fabregat-Santiago, J. Bisquert, L. Cevey, P. Chen, M. Wang,
€
Baker, P. Wang, S. M. Zakeeruddin and M. Gratzel, J. Am. Chem.
Soc., 2008, 130, 10720.
17 H. Tian, X. Yang, J. Cong, R. Chen, J. Liu, Y. Hao, A. Hagfeldt and
L. Sun, Chem. Commun., 2009, 6288.
18 A. Reynal and E. Palomares, Energy Environ. Sci., 2009, 2, 1078.
19 J. N. Clifford, E. Palomares, M. K. Nazeeruddin, R. Thampi,
€
M. Gratzel and J. R. Durrant, J. Am. Chem. Soc., 2004, 126,
5670.
20 H. Choi, S. Kim, S. O. Kang, J. Ko, M.-S. Kang, J. N. Clifford,
€
S. M. Zakeeruddin and M. Gratzel, J. Am. Chem. Soc., 2009, 131,
558.
€
ꢁ
48 Q. Wang, S. Ito, M. Gratzel, F. Fabregat-Santiago, I. Mora-Sero,
J. Bisquert, T. Bessho and H. Imai, J. Phys. Chem. B, 2006, 110,
25210.
€
A. Forneli, E. Palomares, M. K. Nazeeruddin and M. Gratzel,
Angew. Chem., Int. Ed., 2008, 47, 8259.
21 (a) J. J. Cid, J.-H. Yum, S.-R. Jang, M. K. Nazeeruddin,
ꢁ
49 J. Bisquert, F. Fabregat-Santiago, I. Mora-Sero, G. Garcia-Belmonte
ꢁ
€
E. M. Ferrero, E. Palomares, J. Ko, M. Gratzel and T. Torres,
and S. Gimenez, J. Phys. Chem. B, 2009, 113, 17278.
50 F. Fabregat-Santiago, J. Bisquert, G. Garcia-Belmonte, G. Boschloo
and A. Hagfeldt, Sol. Energy Mater. Sol. Cells, 2005, 87, 117.
Angew. Chem., 2007, 119, 8510; (b) Y. Chen, Z. Zeng, C. Li,
W. Wang, X. Wang and B. Zhang, New J. Chem., 2005, 29, 773; (c)
X.-F. Wang, Y. Koyamaa, O. Kitaob, Y. Wadad, S.-I. Sasakie,
H. Tamiakie and H. Zhou, Biosens. Bioelectron., 2010, 25, 1970.
€
51 T. Forster, Discuss. Faraday Soc., 1959, 27, 7.
52 J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Kluwer
€
22 (a) J.-H. Yum, S.-R. Jang, P. Walter, T. Geiger, F. Nuesh, S. Kim,
J. Ko, M. Gratzel and M. K. Nazeeruddin, Chem. Commun., 2007,
Academic/Plenum Publishers, New York, 1999.
€
3794 | J. Mater. Chem., 2012, 22, 3786–3794
This journal is ª The Royal Society of Chemistry 2012