Journal of

Electroanalytical
Chemistry
Journal of Electroanalytical Chemistry 570 (2004) 257–263
www.elsevier.com/locate/jelechem

Effect of the thickness of the Pt film coated on a counter electrode

on the performance of a dye-sensitized solar cell
Xiaoming Fang 1, Tingli Ma *, Guoqing Guan, Morito Akiyama,
Tetsya Kida, Eiichi Abe
Institute for Structural and Engineering Materials (ISEM, AIST Kyushu), National Institute of Advanced Industrial Science and Technology (AIST),
807-1, Shuku, Tosu, Saga 841-0052, Japan
Received 21 January 2004; received in revised form 23 March 2004; accepted 10 April 2004
Available online 19 June 2004

Abstract

The effects of the thickness and morphology of a Pt film coated on a counter electrode on the performance of a dye-sensitized
solar cell (DSC) were investigated. Deposition of a Pt film ranging in thickness from 2 to 415 nm gradually decreases the sheet
resistance of the counter electrode. No significant difference in the charge-transfer resistance at the electrolytejcounter electrode
interface was observed for a Pt film thickness ranging from 25 to 415 nm. A high energy conversion efficiency of approximately 5%
can be obtained for DSCs based on a counter electrode with a very thin Pt film of 2 nm, as well as with a 415-nm thick Pt film. These
results are important for reducing production costs by reducing the required amount of expensive platinum.
Ó 2004 Elsevier B.V. All rights reserved.

Keywords: Pt film thickness; Counter electrode; Dye-sensitized solar cell; Performance; Sheet and charge-transfer resistance

1. Introduction which is usually an organic solvent containing the io-

Since a high solar energy conversion efficient (10%)
dye-sensitized solar cell (DSC) was reported by Gr€atzel
and co-workers, the DSC has become the most prom-
ising alternative to conventional silicon solar cells [1–6].
The photoelectrode of the DSC consists of a thin film
of porous nanocrystalline TiO2 nanoparticles anchored
with a ruthenium complex as the dye-sensitizer, which
harvests the light energy. Photoexcitation of the dye
leads to a fast injection of electrons into the conduction
band of TiO2 , and then the photo generated electrons
pass through the interconnecting network of TiO2
particles and are collected by the conducting glass
substrate. The oxidized dye is subsequently regenerated
by electron donation from the electrolyte solution
*
Corresponding author. Tel.: +81-942-81-3640; fax: +81-942-81-
3690.
E-mail address: tingli-ma@aist.go.jp (T. Ma).
1
Also at College of Chemical Engineering, South China University
of Technology, Guangzhou 510640, PR China.
dide (I
)/triiodide (I


3 ) couple. The I3 ions formed in
this regeneration process diffuse to the counter elec-
trode, where they are rapidly reduced to I
by capture
of the electrons migrating from the external circuit.
Overall, the DSC generates electric power from light
without undergoing any permanent chemical transfor-
mation [7–12].
Counter electrodes, as one important component in
DSCs, are usually constructed of conducting glass sub-
strates coated with Pt films. The roles of the counter
electrode are to transfer electrons arriving from the ex-
ternal circuit back to the redox electrolyte and to cata-
lyze the reduction of the redox couple in order to keep
the overvoltage low at photocurrent densities up to 20
mA/cm2 , where the Pt serves as an electrocatalyst
[4,13,14].
In recent years, regenerative DSCs have been under
extensive investigation, which has focused mainly on
the following aspects: the identification and synthesis
of dye molecules that match the requirement of
absorbing all light below a threshold wavelength of

0022-0728/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jelechem.2004.04.004
258 X. Fang et al. / Journal of Electroanalytical Chemistry 570 (2004) 257–263

approximately 920 nm [15–18]; theoretical analysis on 2.2. Fabrication and characterizations of platinized
charge-transport processes and models in mesoporous counter electrodes
nanocrystalline TiO2 films [19–21]; substitution of sol-
id-state heterojunctions for the liquid electrolyte [22– Platinized counter electrodes were fabricated by de-
27], etc., whereas, few studies have been performed on positing Pt particles onto conducting glass sheets using
the counter electrode in the DSC. To reduce the fab- an r.f. magnetron sputtering instrument (TOKUDA,
rication cost of the counter electrodes, Suzuki et al. MODEL CFS-4ES-231USC; power, 200 W; deposition
[28] applied carbon nanotubes to the counter electrodes

rate, 2.8 A/s). The Pt film thickness was controlled by
in place of expensive Pt, and Saito et al. [29] used variation of the deposition time. Measurement of the
chemically polymerized poly (3,4-ethylenedioxythioph- exact thickness of the Pt film was impossible for thin
ene) on a conductive glass as a counter electrode. films on the very rough surface of the FTO glass sub-
However, the conversion efficiencies of the DSCs based strate. The thickness of the Pt film was estimated for the
on these new types of counter electrodes were lower convenience of discussion. The relationship between the
than that based on the platinized counter electrode. Pt film thickness and the deposition time is shown in
Until now, platinized FTO glass has been the preferred Table 1.
material for the counter electrode of a DSC due to the The surface microstructures of the FTO glass and the
excellent catalytic activity of Pt for triiodide reduction; platinized FTO glass counter electrodes prepared were
however, less attention has been paid to the effect of Pt observed using a scanning electron microscope (SEM,
film thickness on the properties of platinized counter Hitachi S–5200).
electrodes. The sheet resistances of the platinized counter elec-
We have performed systematic studies on counter trodes were measured by a four-dot method (HL5500
electrodes, including the effect of Pt film thickness on the PC Hall effect measurement system).
performance of the DSCs and the development of novel Electrochemical impedance spectroscopy (EIS) was
substrate materials for the counter electrode. employed to determine the charge-transfer kinetic per-
In this paper, we report the properties of platinized formance at the electrolytejplatinized counter electrode
counter electrodes with Pt films of different thicknesses, interface. For EIS measurements, a sandwich cell con-
including surface microstructures, sheet resistances, sisting of two identical platinized FTO electrodes, be-
light-reflecting properties, charge-transfer resistances of tween which a Surlyn film (Dupont, 50 lm) was inserted
the platinumjelectrolyte interface, with a view to eluci- as a spacer, was used. The active area of the electrodes
dating the effects of the Pt film thickness of counter was about 0.8 cm2 . The cell was filled with an electro-
electrodes on the performance of DSCs. lyte, which was composed of 0.1 M LiI + 0.1 M I2 + 0.6
M 1,2-dimethyl-3-propylimidazolium iodine + 0.5 M
tert-butylpyridine in methoxypropionitrile. EIS spectra
2. Experimental were obtained at zero bias potential and 10 mV ampli-
tude over the frequency range of 0.05–105 Hz.
2.1. Preparation of TiO2 working electrodes The light-reflecting property of the platinum mirror
surface of the counter electrode was determined by
Porous TiO2 films were prepared by a screen-printing measuring the UV–Vis reflection spectra of the dye-
technique (Screen: Mitani Electronics, MEC3000) onto sensitized TiO2 film, which faced the platinized counter
an FTO glass sheet (Asahi Glass Co. Ltd., fluorine- electrode and was irradiated from the front and back
doped SnO2 overlayer, FTO, sheet resistance: 10 X/sq) glass side, using a UV–Vis spectrophotometer equipped
with a viscous paste of TiO2 powder (P25, Nippon
Aerosil). The paste was prepared using 100% polyeth-
ylene 600 as a dispersion medium, developed by Ma Table 1
et al. [30]. Twenty-five pieces of homogeneous TiO2 film The performances of platinized counter electrodes with Pt films of
(area: 0.2 cm2 ) were formed on an FTO glass sheet at different thicknesses
one time. The thickness of the TiO2 films obtained was Sputtering Pt film Rs (X/cm2 ) Rct (X/cm2 )
approximately 14 lm. After drying at 50–60 °C for 15 time (s) thickness (nm)
min on a hot plate, the TiO2 films were sintered at 450 0 0 10 1.5  107
°C for 30 min, impregnated with a 0.05 M titanium 7 2 8.85 0.8
36 10 6.50 1.2
tetrachloride solution in a hermetic glass bottle at 70 °C
90 25 3.19 2.1
for 30 min, followed by a firing at 500 °C for 30 min. 180 50 2.28 1.6
The TiO2 films obtained were immersed in an ethanol 360 100 1.18 2.1
solution of cis-bis(isothiocyanato)bis(2,20 -bipyridyl-4,40 - 720 200 0.60 1.7
dicarboxylato)- ruthenium(II)bis-tetrabutylammonium 1080 300 0.39 1.7
1500 415 0.32 2.1
(N719 dye, 5  10
4 M) for 12 h at 25 °C.
 X. Fang et al. / Journal of Electroanalytical Chemistry 570 (2004) 257–263
with an integrating sphere setup (JASCO V-570). The
UV–Vis absorption spectra for the configuration of
clamping the TiO2 film with the counter electrodes
containing Pt films of various thicknesses were also
measured.
2.3. Photovoltaic characterizations
Current–voltage curves of the DSCs based on the
counter electrodes with Pt films of different thickness
were obtained by scanning a bias voltage while mea-
suring photocurrents under white light irradiation (100
mW/cm2 ) using a sandwich solar cell constructed of the
dye-sensitized TiO2 electrode and the platinized counter
electrode. A 300 W Xe short ARC lamp (Ushio inc.,
Japan) served as a light source. The photon flux im-
pinging on the cell was determined by a power meter
(Nova, Ophir Optronics Ltd.) and a calibration cell (BS-
520, s/n 019, Bunkoh-Keiki Co., Ltd.). The composition
of the electrolyte solution used was as the same as that
described in Section 2.2.
3. Results and discussion
3.1. Surface morphologies of Pt films sputtered on FTO
glass
For the surfaces of counter electrodes with Pt films of
2- and 10-nm thickness, no obvious mirror appearance
was observed, whereas the counter electrodes with Pt
Fig. 1. Surface morphologies of FTO glass and platinized FTO glass counter electrodes: (a) FTO glass substrate; (b–d) platinized counter electrode
with Pt films of 100-nm, 200-nm, and 415-nm thicknesses.
259
films from 25-nm to 415-nm thickness possessed good
mirror surfaces. Fig. 1 shows the SEM surface mor-
phology of bare FTO glass and FTO glass coated with
Pt films of 100-, 200- and 415-nm thickness. As is shown
in Fig. 1(a), the bare FTO glass has a rough scale-like
surface structure due to the fluorine-doped SnO2 layer
of the 740-nm thick coating on the glass sheet, because
such a structure is considered to be helpful for the ad-
hesion of the Pt film on the surface. It was also observed
that the 100-nm thick Pt film sputtered on an FTO glass
substrate is constructed of Pt nanoparticles with a grain
size of ca.14 nm, as is shown in the inset of Fig. 1(b), and
for the Pt films of 2-, 10-, 25- and 50-nm thickness, the
sizes of the Pt nanoparticles are less than 10 nm (not
shown here). When the Pt film thickness increased to
200 nm, the grain size of Pt particles grew to 40–60 nm
(inset of Fig. 1(c)). Furthermore, as the Pt film thickness
increased to 415 nm, the granular surface microstructure
disappeared, and an uneven Pt film was formed on the
F-doped SnO2 layer, as is shown in the inset of Fig. 1(d).
It can be concluded that the grain size of the Pt particles
gradually increased and their porosity declined with the
increase in deposition time, due to the aggregation of the
Pt nanoparticles during sputtering.
3.2. Effect of Pt film thickness on sheet resistances of
counter electrodes
Fig. 2 illustrates the dependence of the sheet resis-
tance Rsh of the platinized counter electrode on the Pt
film thickness, that is, on the sputtering time. The sheet
260 X. Fang et al. / Journal of Electroanalytical Chemistry 570 (2004) 257–263
Fig. 2. The dependence of sheet resistance of platinized counter elec-
trodes on Pt film thickness.
resistance of the counter electrode decreased gradually
with an increase in the sputtering time from 7 to 1500 s,
and correspondingly, in the Pt film thickness from 2 to
415 nm. The slope of the curve can be divided into three
parts: the Rsh decreased remarkably for a sputtering time
between 7 and 90 s, with a Pt film thickness between 2
and 25 nm; the Rsh decreased gradually for thicknesses
between 25 and 100 nm with deposition times ranging
from 90 to 360 s, and it decreased slowly for sputtering
times between 360 and 1500 s, when the Pt film thickness
was between 100 and 415 nm. The variation of Rsh can
be interpreted as follows: according to the SEM mi-
crographs, it can be conjectured that the surface of the
FTO glass was not fully covered with Pt particles within
the deposition time of 90 s, while the FTO surface was
gradually fully covered during deposition times from 90
to 360 s. Moreover, when the deposition time exceeded
360 s, the grain size of the Pt particles increased and the
Pt film became dense due to the aggregation of the
sputtered Pt particles, so that the Rsh values were not
decreased markedly. These results indicate that the
conductivity of the counter electrode can be improved
by depositing a Pt film on a FTO glass; however, when
the Pt film thickness exceeds 100 nm, further deposition
gives no significant improvement in the conductivity.
3.3. Effect of the Pt film thickness on the charge-transfer
resistance
During the mediator regenerative cycle in DSC, the
oxidized species, I

3 , must be converted back (reduced)
to iodide on the counter electrode, according to the
following reaction:
I


3 þ 2e ! 3I


ð1Þ
The previous reaction is confined to the surface of the
counter electrode. In order to compare the reaction rate
on Pt films with different thicknesses, the charge-transfer
resistances at the Ptjelectrolyte interface were deter-
mined by EIS measurements, as reported in the litera-
ture [13,14,31–34]. Fig. 3 shows a typical impedance
spectrum, which was obtained from an experimental cell
consisting of two identical platinized FTO electrodes
with a Pt film of 25-nm thickness. The equivalent circuit
for this type of cell is illustrated in Fig. 4, as was de-
scribed in the literature [14,33]. For high frequencies
around 100 kHz, the impedance is dominated by the
ohmic serial resistance (Rs ) , which represents the resis-
tance of the platinized counter electrode and the elec-
trolyte [31]. On the other hand, the impedance is
dominated by the RC network of the Ptjelectrolyte in-
terface, consisting of the charge-transfer resistance and
the capacitance (C) in the frequency range between 100
Hz and 10 kHz, and by the Nernst diffusion impedance
(ZN ) in the range of 1–10 Hz [33].
The Rct values for cells with different Pt film thick-
nesses were obtained by fitting EIS spectra using the
above equivalent circuit, and the values are summarized
in Table 1. The Rct decreased drastically from 1.5  107
(FTO) to 0.8 X/cm2 by coating a Pt layer as thin as 2
nm, as is shown in Table 1. On the other hand, the Rct
values are between 0.8 and 2.1 X/cm2 for Pt film thick-
nesses ranging from 2 to 415 nm, that is, no marked
difference was observed. Because Pt serves as a catalyst
for triiodide reduction, the drastic decrease in Rct implies
an acceleration of the I
3 reduction rate. As was dis-
cussed in Section 3.1, the grain size of the Pt particles
increased and their porosity declined, when the Pt film
thickness increased from 2 to 415 nm. This may be
Fig. 3. A typical electrical impedance spectrum for a symmetric Pt thin-
layer cell taken at zero bias potential and 10 mV amplitude.
Fig. 4. Equivalent circuit for the impedance spectrum in Fig. 3. N:
Nernst diffusion impedance; Rct charge-transfer resistance of one
electrode; C: double-layer capacity of one electrode; Rs : serial resis-
tance.
 X. Fang et al. / Journal of Electroanalytical Chemistry 570 (2004) 257–263
caused by the decline in catalytic activity, because the
catalytic activity generally depends on the surface area
of the catalysts. On the other hand, as was discussed in
Section 3.2, the sheet resistance (Rsh ) of the platinized
counter electrodes decreased with the increase in Pt film
thickness, that is, the interconnection of Pt particles was
improved, which contributes to the electron transfer at
the counter electrodes. Therefore, when the thickness of
the Pt film was greater than 415 nm, the combined ef-
fects of the catalytic activity and conductivity of the
platinized counter electrodes resulted in no marked
difference in the charge-transfer resistance at the inter-
face of the electrolyte and counter electrodes. According
to the comments of one referee, the small change in Rct
may be explained by the high catalytic activity of small
Pt particles, and only medium activity for larger parti-
cles. The medium activity is counterbalanced by the fact
that there is more Pt present (greater area) in thicker
films.
3.4. Effect of Pt film thickness on performances of the
DSCs
It is known that the counter electrode should have the
following characteristics for good operation of dye-
sensitized nanocrystalline solar cells: (1) good conduc-
tivity for transferring electrons, (2) excellent catalytic
activity for triiodide reduction, and (3) light-reflecting
ability for improving light-harvesting efficiency [6–10].
We have already discussed the properties of conductivity
and catalytic activity. The next subject concerns the
light-reflecting properties of counter electrodes with Pt
mirror surfaces. Fig. 5 shows the UV–Vis reflection
spectra of the TiO2 film without and with platinized
counter electrodes. The UV–Vis absorption spectra are
shown in Fig. 6. The reflection spectra of the TiO2 film
irradiated from the TiO2 film side are different from that
irradiated from the glass side due to the reflective effect
Fig. 5. The UV–Vis reflection spectra of the TiO2 photoelectrode il-
luminated from the TiO2 side and from the glass side: (a) with plati-
nized counter electrode and electrolyte; (b) with platinized counter
electrode; (c) illuminated from TiO2 side; (d) illuminated from the glass
side.
261
Fig. 6. The UV–Vis absorption spectra of a TiO2 film illuminated from
the back glass side of the TiO2 film that was clamped to platinized
counter electrodes with Pt films of different thicknesses.
of the FTO glass substrate. On the other hand, when the
TiO2 film and the platinized counter electrode were
fabricated together, the absorption intensity was slightly
stronger than that without the platinized counter elec-
trode. These results indicated that the reflective property
of the Pt film could slightly improve the light-harvesting
efficiency of the dye adsorbed onto the TiO2 film. Fur-
thermore, a broad absorption was observed below 500
nm, due to the electrolyte, when it was injected between
the two electrodes. When the Pt film thickness was in the
range from 25 to 415 nm, we observed no difference
between the absorbance of the TiO2 film clamped to the
platinized counter electrode. However, it should be
mentioned that there was increased reflection for these
spectra in the 700–800 nm region. This result showed
that the incident light-reflecting ability of the Pt mirror
has an effect on the light-harvesting efficiency of the dye,
when the incident light passes through the 14-lm thick
TiO2 layer; however, it was not marked. Therefore, it
can be concluded that the incident light-reflecting ability
of the Pt film of the counter electrode is not a key factor
in the previous three properties.
To investigate exactly the effect of the Pt film thick-
ness of the counter electrode on the performance of the
DSC, all the dye-sensitized TiO2 films used were pre-
pared at one time by the screen-printing method, so that
they could be considered to have the same microstruc-
ture. Table 2 shows the performance characteristics of
the DSCs composed of counter electrodes with Pt films
of different thicknesses, and the photocurrent density–
voltage (j–V) curves of the DSCs are shown in Fig. 7.
When a FTO glass without Pt film was used as the
counter electrode, the open-circuit photovoltage (Voc ) of
the DSC was only 120 mV, the short-circuit photocur-
rent density (jsc ) was about 1.9 mA/cm2 , and the overall
energy-conversion efficiency achieved was only 0.38%,
as is shown in Table 2. Such a low conversion efficiency
262 X. Fang et al. / Journal of Electroanalytical Chemistry 570 (2004) 257–263

Table 2 tween 2 and 415 nm. Therefore, a Pt film as thin as 2 nm

Performance characteristics of dye-sensitized solar cells composed of is sufficient for a DSC with methoxypropionitrile as the
counter electrodes with Pt films of different thicknesses
solvent of the electrolyte. This conclusion is in agree-
Pt film Voc jsc Fill g ment with the results obtained by Gr€atzel and co-
thickness (nm) (mV) (mA/cm2 ) factor (%) (%)
workers [14] and Hauch and Georg [33].
0 227 4.05 13 0.12 There is a possibility that should be considered,
2 680 12.50 60 4.90 namely, that the catalytic activity of the Pt film may be
10 705 11.35 61 4.92
25 702 11.60 59 4.99
reduced and saturated due to the agglomeration of the
50 690 12.10 62 5.17 Pt particles under the high r.f sputter power. For this
100 701 11.25 62 4.89 reason, we compared the performance of DSCs with Pt
200 698 11.40 64 5.08 counter electrodes sputtered under high (200 W) and
300 685 11.60 63 5.03 low (50 W) r.f power, and no definite difference was
415 694 12.50 60 5.18
observed. Therefore, it was clear that no improvement
of performance was achieved for DSCs coated with a
thick Pt film on an FTO glass.
Because of the durability and stability of the DSC
and the concern that a small amount of platinum might
dissolve in the electrolyte by oxidation and complex
formation with iodide/triiodide [35], accordingly, a too
thin Pt film should be avoided in the fabrication of the
DSCs. We tested the DSC cells having platinized
counter electrodes with different Pt film thicknesses, in
contact with the electrolyte solution for three months.
No deterioration in the performances of the cells was
observed, even for the counter electrode with a Pt film as
thin as 25 nm. These findings are important in practical
applications of DSCs, especially, in view of reducing
costs by being able to use a small amount of expensive
Fig. 7. Photocurrent–voltage curves of dye-sensitized solar cells com-
posed of counter electrodes with Pt films of different thicknesses.
platinum.

is due to the lack of catalytic activity of the bare FTO 4. Conclusions

glass for triiodide reduction, as is evident from the
previously mentioned fact that the Rct at the interface of
an FTO glass and the electrolyte is very large. On the
contrary, when a 2-nm thick Pt film was sputtered on an
FTO glass, the Voc and jsc of the DSC were strikingly
increased to 680 mV and 12.5 mA/cm2 , and the overall
energy conversion efficiency achieved was up to 4.90%.
This result means that the electron transfer rate at the
electrolytejcounter electrode interface is increased, due
to the catalytic activity of the Pt film coated on the FTO
glass. It was clarified that the catalytic activity of the Pt
film is a key factor for the performance of the DSC. On
the other hand, we observed no apparent difference in
Voc , jsc and the conversion efficiency of the DSCs based
on Pt film thicknesses between 2 and 415 nm. As dis-
cussed in Section 3.3, when the Pt films thickness is
greater than 415 nm, the charge-transfer resistance at
the interface of the electrolytejcounter electrodes ex-
hibited no marked difference, so that the DSC composed
of a counter electrode with a thin Pt film can give as high
a conversion efficiency as that with a thick Pt film. The
results obtained indicate that the thickness of a Pt film
coated on an FTO glass has no significant influence on
the performance of the DSC for Pt film thicknesses be-
We investigated the effect of the thickness of Pt films
sputter-coated on an FTO glass on the performance of
DSCs. In summary, the following conclusions were
reached: (1) The conductivity of the counter electrode
was gradually improved by depositing the Pt film at a
thickness from 2 to 415 nm. However, when the Pt film
thickness exceeded 100 nm, further Pt deposition gave
no significant improvement in the conductivity; (2) The
charge-transfer resistance at the electrolytejcounter
electrode interface showed no obvious change with in-
creased thickness of the Pt film; (3) The light-reflecting
ability of the Pt mirror was not a key property in the
performance of the DSC; (4) A high conversion effi-
ciency of ca. 5% could be obtained for the DSC based on
a thin Pt film of 2 nm. It was thus shown that the Pt film
thickness had no significant influence on the perfor-
mance of the DSC.
Acknowledgements
The authors acknowledge discussions with Dr. Sato-
shi Uchida, Inst. Multidisciplinary Res. Adv. Mat. Res.
 X. Fang et al. / Journal of Electroanalytical Chemistry 570 (2004) 257–263
Bldg. Chem. React. Sci., Tohoku Univ. This research
was supported by the Japan Science and Technology
Corporation.
References
[1] B. O’Regan, M. Gr€atzel, Nature 353 (1991) 737.
[2] M.K. Nazeeruddin, A. Kay, R. Humphry-Baker, E. M€ uller, P.
Liska, N. Vlachopoulos, M. Gr€atzel, J. Am. Chem. Soc. 115
(1993) 6382.
[3] C.J. Barbe, F. Arendse, P. Comte, M. Jirousek, F. Lenzmann, V.
Shklover, M. Gr€ atzel, J. Am. Ceram. Soc. 80 (1995) 3157.
[4] A. Kay, M. Gr€ atzel, Sol. Ener. Mater. Sol. Cell 44 (1996) 99.
[5] M.K. Nazeeruddin, P. Pechy, T. Renouard, S.M. Zakeeruddin, R.
Humphry-Baker, P. Comte, P. Liska, L. Cevey, E. Costa, V.
Shklover, L. Spiccia, G.B. Deacon, C.A. Bignozzi, M. Gr€atzel, J.
Am. Chem. Soc. 123 (2001) 1613.
[6] M.K. Nazeeruddin, R. Humphry-Baker, P. Liska, M. Gr€atzel, J.
Phys. Chem. B 107 (2003) 8981.
[7] A. Hagfeldt, M. Gr€atzel, Chem. Rev. 95 (1995) 49.
[8] N. Papageorgiou, P. Liska, A. Kay, M. Gr€atzel, J. Electrochem.
Soc. 146 (1999) 898.
[9] M. Gr€ atzel, Curr. Opin. Colloid Interface Sci. 4 (1999) 314.
[10] M. Gr€ atzel, Nature 414 (2001) 338.
[11] Y. Liu, A. Hagfeldt, X.-R. Xiao, S.-E. Lindquist, Sol. Ener.
Mater. Sol. Cell 55 (1998) 267.
[12] K. Sayama, H. Sugihara, H. Arakawa, Chem. Mater. 10 (1998)
3825.
[13] N. Papageorgiou, Y. Athanassov, M. Armand, P. Bonh^ ote, H.
Pettersson, A. Azam, M. Gr€atzel, J. Electrochem. Soc. 143 (1996)
3099.
[14] N. Papageorgiou, W.F. Maier, M. Gr€atzel, J. Electrochem. Soc.
144 (1997) 876.
[15] T. Renouard, R.-A. Fallahpour, Md.K. Nazeeruddin, R. Hump-
hry- Baker, S.I. Gorelsky, A.B.P. Lever, M. Gr€atzel, Inorg. Chem.
41 (2002) 367.
263
[16] J.-J. Lagref, Md.K. Nazeeruddin, M. Gr€atzel, Synthetic Metals
138 (2003) 333.
[17] T. Ma, K. Inoue, H. Noma, K. Yao, E. Abe, J. Photochem.
Photobiol. A: Chem. 152 (2002) 207.
[18] T. Ma, K. Inoue, K. Yao, H. Noma, T. Shuji, E. Abe, J. Yu, X.
Wang, B. Zhang, J. Electroanal. Chem. 537 (2002) 31.
[19] S.Y. Huang, G. Schlichth€ orl, A.J. Frank, J. Phys. Chem. B 101
(1997) 2576.
[20] J. van de Lagemaat, A.J. Frank, J. Phys. Chem. B 105 (2001)
11,194.
[21] L.M. Peter, N.W. Duffy, R.L. Wang, K.G.U. Wijayantha, J.
Electroanal. Chem. 524–525 (2002) 127.
[22] U. Bach, D. Lupo, M. Gr€atzel, Nature 395 (1998) 583.
[23] T. Kitamura, M. Maitani, S. Ynagida, Chem. Lett. (2001)
1054.
[24] M.A. De Paoli, A.F. Nogueira, D.A. Machado, C. Longo,
Electrochim. Acta 46 (2001) 4243.
[25] C. Longo, A.F. Nogueira, M.A. De Paoli, J. Phys. Chem. B 106
(2002) 5925.
[26] B. O’Regan, F. Lenzmann, R. Muis, J. Wienke, Chem. Mater. 14
(2002) 5023.
[27] S. Murai, S. Mikoshiba, H. Sumino, S. Hayase, J. Photochem.
Photobiol. A 148 (2002) 33.
[28] K. Suzuki, M. Yamaguchi, M. Kumagai, S. Yanagida, Chem.
Lett. 32 (2003) 28.
[29] Y. Saito, T. Kitamura, Y. Wada, S. Yanagida, Chem. Lett. 31
(2002) 1060.
[30] T. Ma, T. Kida, M. Akiyama, K. Inoue, S. Tsunematse, K. Yao,
H. Noma, E. Abe, Electrochem. Commun. 5 (2003) 369.
[31] J. van de Lagemaat, N.-G. Park, A.J. Frank, J. Phys. Chem. B 104
(2000) 2044.
[32] R. Kern, R. Sastrawan, J. Ferber, R. Stangl, J. Luther, Electro-
chim. Acta 47 (2002) 4213.
[33] A. Hauch, A. Georg, Electrochim. Acta 46 (2001) 3457.
[34] N. Koide, H. Imai, T. Mitate, L.Y. Han, XXIst Intern. Conf. on
Photochem., Book of Abstracts 2003, 2P031.
[35] E. Olsen, G. Hargen, S.E. Linquist, Sol. Ener. Mater. Sol. Cell 63
(2000) 267.