Enhanced performance of a dye-sensitized solar cell with an electrodeposited-platinum counter electrode

Article abstract of DOI:10.1016/j.electacta.2007.10.074

A counter electrode was prepared for a dye-sensitized solar cell (DSSC) through electrochemical deposition of mesoporous platinum on fluorine-doped tin oxide glass in the presence of a structure-directing nonionic surfactant, octaethylene glycol monohexadecyl ether (C₁₆EO₈). The DSSC fabricated with the electrochemically deposited Pt (ED-Pt) counter electrode rendered a higher solar-to-electricity conversion efficiency of 7.6%, compared with approximately 6.4% of the cells fabricated with the sputter-deposited or most commonly-employed thermal deposited Pt counter electrodes. This enhanced efficiency is attributed to the higher short-circuit photocurrent arising from the increases in the active surface area and light reflection as well as the decrease in the sheet resistance of the ED-Pt film, relative to those of the Pt films prepared by the other two deposition methods. The sputter-deposited Pt film yielded almost the same photovoltaic characteristics as the thermal deposited Pt film. The Pt films were characterized by FE-SEM, AFM, cyclic voltammetry, chronoamperometry, electrochemical impedance spectroscopy, sheet resistance measurements, adhesion tests, and light reflection tests.

Full text of DOI:10.1016/j.electacta.2007.10.074

Available online at www.sciencedirect.com
Electrochimica Acta 53 (2008) 2890–2896
Enhanced performance of a dye-sensitized solar cell with an
electrodeposited-platinum counter electrode
Chang Ho Yoona, R. Vittala, Jiwon Leeb, Won-Seok Chaec, Kang-Jin Kima,∗
a Department of Chemistry, Korea University, Seoul 136-713, South Korea
b Samsung SDI Co. Ltd., Gyeonggi-do 449-577, South Korea
c Department of Chemistry, Daejin University, Pocheon, Gyounggi-do 487-711, South Korea
Received 4 September 2007; received in revised form 24 October 2007; accepted 29 October 2007
Available online 21 December 2007
Abstract
A counter electrode was prepared for a dye-sensitized solar cell (DSSC) through electrochemical deposition of mesoporous platinum on fluorine-
doped tin oxide glass in the presence of a structure-directing nonionic surfactant, octaethylene glycol monohexadecyl ether (C16EO8). The DSSC
fabricated with the electrochemically deposited Pt (ED-Pt) counter electrode rendered a higher solar-to-electricity conversion efficiency of 7.6%,
compared with approximately 6.4% of the cells fabricated with the sputter-deposited or most commonly-employed thermal deposited Pt counter
electrodes. This enhanced efficiency is attributed to the higher short-circuit photocurrent arising from the increases in the active surface area
and light reflection as well as the decrease in the sheet resistance of the ED-Pt film, relative to those of the Pt films prepared by the other two
deposition methods. The sputter-deposited Pt film yielded almost the same photovoltaic characteristics as the thermal deposited Pt film. The Pt
films were characterized by FE-SEM, AFM, cyclic voltammetry, chronoamperometry, electrochemical impedance spectroscopy, sheet resistance
measurements, adhesion tests, and light reflection tests.
© 2007 Elsevier Ltd. All rights reserved.
Keywords: Electrodeposited-Pt counter electrode; Thermal and sputter depositions; Active surface area; Charge transfer resistance; Dye-sensitized solar cell
1. Introduction
tion kinetics on the Pt/FTO electrode leads to minimum energy
loss. The light-reflecting character of the Pt is also desirable for
increasing the light harvesting efficiency of the sensitizing dye.
Until now, platinized FTO glass has been the preferred com-
bination for the counter electrode of a DSSC, due to the excellent
catalytic activity of Pt for the reduction of I3− ions and due to its
light reflecting nature. The counter electrode is usually prepared
by spreading a drop of hydrogen hexachloroplatinate (H2PtCl6)
2-propanol solution on a conducting glass substrate, followed
by its annealing at 450 ◦C. The platinized counter electrode of a
DSSC is also prepared by depositing Pt particles onto conduct-
ing glass by sputtering [2,3]. These two references also reported
that a 2 nm-thick sputtered-Pt is sufficient to obtain a reasonable
charge transfer resistance, one of the important parameters of a
DSSC.
High electrochemical activity of the counter electrode is
an important requirement for enhanced performance of a dye-
sensitized solar cell (DSSC). A thin layer of Pt is well established
as the catalyst on the counter electrode substrate, usually a
fluorine-doped tin oxide (FTO) conducting glass. When the
counter electrode is only FTO glass without a Pt layer, the con-
version efficiency of the pertinent cell is very low (0.1%) [1].
This indicates clearly that the function of the counter electrode
depends essentially on the deposited Pt on the substrate. The role
of Pt in the counter electrode of a DSSC is to catalyze the reduc-
tion of triiodide, I3−, ions that are produced by the oxidation of
iodide ions in the electrolyte by the oxidized dye. This cataly-
sis is required to reduce the voltage loss due to charge-transfer
overpotential at the counter electrode. The fast triiodide reduc-
The counter electrodes prepared with carbon-based mate-
rials such as activated carbon [4], carbon nanotubes [5], hard
carbon spherule [6], and acetylene-black spheres [7] were also
used in DSSCs; however, such cells generally showed lesser
efficiencies than that obtainable in the case of a DSSC with
∗
Corresponding author. Tel.: +822 3290 3127; fax: +822 3290 3121.
E-mail address: kjkim@korea.ac.kr (K.-J. Kim).
0013-4686/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2007.10.074

C.H. Yoon et al. / Electrochimica Acta 53 (2008) 2890–2896
2891
Pt counter electrode [8]. Recently, Murakami et al. employed
2. Experimental
carbon black as the counter electrode catalyst in a DSSC and
achieved a remarkable light-to-energy conversion efficiency of
9.1% under 100 mW cm−2 [9]. Saito et al. [10] used chemically
polymerized poly(3,4-ethylenedioxythiophene) on an ITO glass
to prepare the counter electrode and obtained a conversion effi-
ciency, comparable to that of a DSSC using a sputter-deposited
Pt counter electrode. Furthermore, Biancardo et al. reported
that a counter electrode prepared by spin-coating of poly(3,4-
ethylenedioxythiophene):polystyrenesulphonate (PEDOT:PSS)
had shown a similar I3−/I− catalytic activity to that of a Pt
counter electrode in a quasi-solid state dye-sensitized solar cell
[11].
The usual FTO substrate was replaced by stainless steel, Ni
and other plastic materials by Ma et al. [12], with Pt layer being
used as the catalyst in all the cases. They found that stainless
steel, nickel substrates and the plastic substrate, polyethylene
naphthalate film coated with tin-doped indium oxide (ITO-PEN
110) were not only stable in the electrolyte, but also rendered
efficiencies for their DSSCs that are comparable to that of a cell
with a platinized FTO counter electrode. Wang et al. prepared
a counter electrode for a DSSC by thermal decomposition of
H2PtCl6 on NiP-plated glass and the conversion efficiency of
the pertinent cell increased by 33%, compared with that of a cell
with a Pt/FTO counter electrode [13].
2.1. Preparation of the counter electrodes
The electrochemical preparation of a Pt counter electrode
(0.8 cm × 1.0 cm) for a DSSC was accomplished, using a poten-
tiostat/galvanostat (PAR EG&G Model 273A), in a home-made
three-electrode cell, assembled with a Pt coil counter electrode,
an Ag/AgCl reference electrode, and an FTO conducting glass
(Pilkington Co., TEC 7, approximately 80% visible transmit-
tance) as the working electrode. The deposition solution con-
sisted of 0.42 g of C16EO8 (octaethylene glycol monohexadecyl
ether), 0.3 ml of 8% H2PtCl6·H2O, and 0.5 ml of distilled water.
Prior to coating, the FTO conducting glass plate of dimensions
1.5 cm × 1.5 cmwassubjectedtoultrasoniccleaningusingwater
and then ethanol. The potentiostatic deposition of Pt film on the
FTO substrate was carried out at a constant potential of −0.06 V
vs. Ag/AgCl for 900 s. The Pt-electrodeposited conducting glass
was placed in distilled water for 2 h to remove any remaining
C16EO8. This process was repeated three times, each time rins-
ing the substrate with water, to ensure the complete removal of
the residual surfactant. The Pt-deposited FTO glass was dried at
100 ◦C for 10 min and annealed at 450 ◦C for 30 min in the air.
Theheatingandannealingatfarhighertemperaturesforatotalof
40 min ensures the complete removal of the surfactant, because
the melting point of the nonionic surfactant, octaethylene glycol
monohexadecyl ether (C16EO8), is 43–44 ◦C.
Considering the vast literature published on various compo-
nents of DSSCs, it may be said that very few publications are
made on their counter electrodes, and that there hardly exist
reports on their electrochemical preparation. Though Papageor-
giou et al. have deposited Pt electrochemically on conducting
tin oxide glass, their focus in that report was only to prepare
an iodine/triiodide reduction catalyst for aqueous and organic
media and no photovoltaic properties of DSSCs were studied
[14].
For thermal deposition, a drop of 5 mM H2PtCl6·H2O in 2-
propanol was placed on the FTO glass substrate, followed by its
drying and then annealing at 380 ◦C for 15 min. Sputter deposi-
tion of Pt was carried out at the rate of 7 nm/min for 30 s using
Ar as the sputtering gas, employing a direct-current method with
a Hitachi E-1030 ion sputter. All the Pt counter electrodes of the
DSSCs used in this study had an area of 0.8 cm × 1.0 cm.
We present in this paper preparation of a platinum counter
electrode for a DSSC by electrodepositing mesoporous Pt on
a conducting substrate, using a nonionic surfactant, octaethy-
lene glycol monohexadecyl ether (C16EO8), and the greatly
enhanced performance of the pertinent DSSC, with reference
to those of cells fabricated using thermal and sputter-deposited
Pt counter electrodes. C16EO8 was selected for achieving a
regular structure for the electrochemically deposited Pt; this
is based on the fact that this nonionic surfactant forms liquid
crystalline phase which can be used as a template for the produc-
tion of mesoporous Pt. [15]. This enhanced performance of the
DSSC with the electrodeposited-Pt counter electrode assumes
its importance in view of the considerations that thermal depo-
sition is not well-suited for flexible plastic electrodes due to
high-temperature annealing, and that sputter deposition is not
cost efficient due to its vacuum process. That is to say that
metallic platinum can be obtained directly on flexible plastic
substrates in the case of electrodeposition and heating process
can thus be avoided, which is not possible for such substrates
through thermal deposition, because in the thermal deposition
heating at 100 ◦C and then annealing at 450 ◦C is required for
the formation of metallic platinum from H2PtCl6·H2O on the
substrate.
2.2. Preparation of the photoanode and fabrication of the
DSSC
TiO2 photoanode was prepared as follows. A thin layer of
non-porous TiO2 was deposited on a cleaned FTO glass from
5% Ti(IV) butoxide in ethanol by spin coating at 3000 rpm, fol-
lowed by its annealing at 450 ◦C. A porous TiO2 film of 10 ␮m
thickness was formed on this coated substrate by a doctor blade
technique as follows. Ti-Nanoxide D paste was placed in a con-
tainer and stirred for 2 h. From this stirred paste a TiO2 film
was made on the above pretreated FTO glass, using the doctor
blade technique. The film was dried at 70 ◦C for 10 min before
removing the tape used for fixing the thickness of the film, and
then annealed at 450 ◦C for 30 min. The resulting TiO2 film was
sensitized with N719 dye by immersing it for 24 h in 0.3 mM
ethanolic solution of the N719 dye (Ru(II)L2(NCS)2:2TBA,
where L = 2,2ꢀ-bipyridyl-4,4ꢀ-dicarboxylic acid). A 2-electrode
sandwiched DSSC was fabricated according to the procedure
given elsewhere [16]. The TiO2 photoanode had an active area
of 0.4 cm × 0.4 cm. The electrolyte consisted of 0.05 M I2, 0.1 M
LiI, 0.6 M 1,2-dimethyl-3-hexylimidazolium iodide, and 0.5 M
4-tert-butylpyridine in 3-methoxypropionitrile.

2892
C.H. Yoon et al. / Electrochimica Acta 53 (2008) 2890–2896
2.3. Characterization of the Pt films and the DSSCs
All chemicals were of analytical grade and used without
further purification. All solutions were prepared using Milli Q
(18.2 Mꢀ cm) H2O. Octaethylene glycol monohexadecyl ether
(C16EO8, 98% pure), and 3-methoxypropionitrile (C4H7NO,
99% pure) were purchased from Fluka. Ti-Nanoxide D paste
(TiO2), N719 dye were purchased from Solaronix SA and their
assay was not available. Ti(IV) butoxide (97% pure), hydro-
gen hexachloroplatinate (IV) hydrate (H2PtCl6·H2O), iodine,
lithium iodide (LiI), 4-tert-butylpyridine, and 2-propanol are
more than 99% pure and purchased from Aldrich. Lithium
perchlorate (LiClO4, 95%) was also purchased from Aldrich.
Anhydrous ethanol was 99.9% pure and obtained from Carlo
Erba and acetonitrile was from Junsei and 99.5% pure. 1,2-
dimethyl-3-hexylimidazoliumiodide(C11H21N2I)wasobtained
from C-TRI and was 98% pure.
Photocurrent density–voltage (J–V) curves of the DSSCs fab-
ricated with different Pt counter electrodes were obtained using
a Keithley M236 source measure unit. A 300 W Xe arc lamp
(Oriel) with an AM 1.5 solar simulating filter for spectral cor-
rection served as the light source, and its light intensity was
adjusted to 100 mW cm−2 by using a Si solar cell. To comple-
ment the photocurrent behavior of the DSSCs, electrochemical
impedance spectra were recorded over a frequency range of
0.05–105 Hz with an AC amplitude of 5 mV using an EG&G
PARC 273A potentiostat with an M1025 frequency–response
detector.
A Hitachi S-4300 field emission-scanning electron micro-
scope (SEM) was used to obtain the surface microstructure of
mesoporous Pt, deposited by different techniques, as well as to
obtain the thickness of the TiO2 film. Atomic force microscopy
(AFM, model: XW-100, PSI Co.) was used to obtain two-
dimensional micrographs of the Pt-coated films, for an area of
5 ␮m × 5 ␮m in each case. The quantitative values of the root-
mean-square roughness (Rrms) of the films were obtained by
using the built-in software of the AFM apparatus. Sheet resis-
tances of the counter electrodes were measured by the four-point
methodusingamodelCMT-SR1000NofChangMinCo., Korea.
For measurement of light reflection of the Pt film, a light from
the light source, model 95 ion laser of Lexel Co was made inci-
dent to the film at an angle of 45◦. The adhesion of the Pt film
on the FTO substrate was tested according to the standard tape
test (ASTM D 3359, cross-cut tape test).
Cyclic voltammograms were obtained in the electrolyte solu-
tion, containing 1 mM I2, 10 mM LiI, and 0.1 M LiClO4 in
acetonitrile, at a scan rate of 100 mV s−1, using an EG&G
PARC 263A potentiostat. The electrochemical cell consisted of
Pt working electrode (0.8 cm × 1.0 cm), prepared by thermal,
sputter or electrochemical depositions, a Pt-wire auxiliary elec-
trodeand an Ag/AgCl referenceelectrode. Chronoamperometric
curves with the Pt films, prepared by thermal, sputter and elec-
trochemical depositions, were obtained with an EG&G PARC
263A potentiostat, using an electrochemical cell consisting of
the Pt-film working electrode (0.8 cm × 1.0 cm), a Pt-wire aux-
iliary electrode and an Ag/AgCl reference electrode. Hundred
millimolars of LiI was used as the electrolyte for the DSSC,
whereas 10 mM of LiI was used as the electrolyte for elec-
trochemical measurements. Sufficiently higher content of I2 is
required for a DSSC to sustain the reduction and oxidation reac-
tions of triiodide and iodide species at the counter electrode
and photoanode, respectively, for higher performance of the
cell. For electrochemical studies 10 mM of LiI was sufficient.
The presence of LiClO4 ensures suppression of the electrostatic
migration of electroactive species in the electrochemical cell,
which is absent in a DSSC.
3. Results and discussion
3.1. Characterization of the Pt films
The Pt films prepared by the three deposition methods
were characterized by FE-SEM, AFM, cyclic voltammetry, and
chronoamperometry. Fig. 1a, b, and c show plane view FE-SEM
images of the Pt films deposited on the FTO glasses by thermal,
sputter, and electrochemical deposition methods, respectively.
Both the thermal deposited (TD) and sputter-deposited (SD) Pt
films show scale-like structures with granules. The TD plat-
inum film has granules of imprecise boundaries in a nebulous
environment, whereas the SD platinum film has well-defined
Fig. 1. FE-SEM images of the Pt films deposited by (a) thermal, (b) sputtering, and (c) electrochemical techniques on FTO substrates.

C.H. Yoon et al. / Electrochimica Acta 53 (2008) 2890–2896
2893
Fig. 2. Two-dimensional AFM surface images of the 5 ␮m × 5 ␮m Pt films deposited by (a) thermal, (b) sputtering, and (c) electrochemical methods on FTO
substrates.
polyhedrons. Interestingly, the ED platinum shows spherical or
elliptical granules and their clusters. Furthermore, the morphol-
ogy of the ED film shows a robust mesostructure with accessible
pores. The pattern of pores in the ED film is also periodically
repeating. We believe that this regular structure in the ED film
is effectively a cast of the structure of the liquid crystal phase
formed in the electrochemical deposition process of Pt, because
C16EO8 was employed as a liquid crystal templating agent for
metal depositions for ordered mesoporosity and for increasing
the surface area of the deposited films [15,17]. A comparison
among the three Pt films clearly shows that the ED-Pt film is
more porous than the TD-Pt and SD-Pt films.
AFM topographic images of the TD-Pt, SD-Pt, and ED-Pt
films are shown in Fig. 2, which gives additional evidence for
the morphology of the films. The increased porosity of the ED-
Pt film compared with those of the TD-Pt and SD-Pt films can
be seen in the figure. The agglomerates in the ED film are of
about 0.3–0.5 ␮m, while their size distribution in the cases of
the TD-Pt and SD-Pt films can not be made with certain approxi-
mation, because of uncertain boundaries and irregular shapes of
granules associated with them. The Rrms values of the Pt films
were measured in order to identify the degree of the surface
roughness of the films and are shown in Table 1. The ED-Pt
film shows the largest Rrms value, approximately 43 nm over
a 5 ␮m × 5 ␮m area of the film, implying that the ED-Pt film
possesses the largest surface area among the three Pt films, in
agreement with the results of the AFM topographic images. The
Rrms values in Table 1 include the contribution from the FTO
substrates. Nevertheless, the values indicate qualitatively the rel-
ative roughness of the very thin films of less than 10 nm in all
the cases. As the amount of the associated platinum material
(∼5–10 ␮g/cm2) was insufficient for BET analysis, we could
not determine the specific surface areas of the platinum films.
The relative active surface areas of the Pt films were veri-
fied by cyclic voltammetry and chronoamperometry from the
rates of the oxidation of iodide and the reduction of iodine.
Fig. 3 shows the cyclic voltammograms of the Pt films in the
electrolyte solution, containing 10 mM LiI, 1 mM I2, and 0.1 M
LiClO4 in acetonitrile. The CVs were obtained at the scan rate
of 100 mV s−1, using a cell assembled with the Pt working
electrodes prepared by the three deposition methods, a Pt-wire
auxiliary electrode and an Ag/AgCl reference electrode. The
anodic peaks are denoted by I and II and the primes indicate
the respective cathodic peaks. In an anodic sweep, iodide is oxi-
dized sequentially to triiodide (peak I) and then to iodine (peak
II) according to the reactions (1) and (2), respectively [18].
3I− → I3− + 2e−
2I3− → 3I2 + 2e−
(1)
(2)
1000
800
I'
(a)
(b)
(c)
TD
SD
ED
600
400
II'
200
0
-200
-400
-600
-800
-1000
Table 1
Root-mean-square roughness (Rrms), reflectivity and sheet resistance of the
Pt films prepared by thermal (TD), sputter (SD), and electrochemical (ED)
depositions
I
II
-1200
-500
Deposition
method
Rrms
(nm)
Reflectivity at
Sheet resistance
(ꢀ/sq)
0
500
1000
1500
488 nm
633 nm
830 nm
Potential (mV vs Ag/AgCl)
TD
SD
ED
27
28
43
0.076
0.18
0.25
0.046
0.20
0.27
0.097
0.35
0.47
7.69
7.46
4.80
Fig. 3. Cyclic voltammograms with the TD, SD, and ED Pt films, obtained in
the electrolyte containing 1 mM I2, 10 mM LiI, and 0.1 M LiClO4 in acetonitrile,
at a scan rate of 100 mV s−1
.

2894
C.H. Yoon et al. / Electrochimica Acta 53 (2008) 2890–2896
When the potential scan is reversed, iodine is reduced first
17
16
15
14
13
12
11
10
9
17
16
15
14
13
12
11
10
9
to triiodide (peak IIꢀ), then to iodide (peak Iꢀ). Fig. 3 clearly
reveals larger anodic and cathodic peaks and associated charges
for the ED-Pt film than those for the TD-Pt and SD-Pt films. The
increased peak heights and areas (or equivalent charges) for the
ED-Pt film can be interpreted in terms of increased active surface
−
area of this film [19] and in terms of enhanced reduction of I3
and oxidation of I− at the ED-Pt working electrode, relative
to those at the film electrodes obtained by the other deposition
methods. Apparently, both of the reactions were facilitated by
the highly accessible pore structure of the ED-Pt film through
whichthediffusionsofI3− andI− ionscanbemorerapidandless
hindered than those in the TD-Pt and SD-Pt films. The increased
activesurfaceareaoftheED-Ptfilmisalsointunewithitsporous
structure as has been observed in its AFM image. This increased
active surface area of ED-Pt film−is desirable for DSSCs, because
of the fact that reduction of I3 at the counter electrode in a
DSSC entails larger active surface area.
8
8
7
7
6
6
5
5
4
4
TD
SD
ED
3
3
2
2
1
1
0
-0.8
0
0.0
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
Applied Potential (V)
Fig. 5. J–V curves of the DSSCs with the counter electrodes with TD-Pt, SD-Pt,
The increased active surface area of the ED-Pt film deduced
from the cyclic voltammograms is further verified by the
chronoamperometric plots shown in Fig. 4. Each plot was
obtained in an acetonitrile solution, containing 1 mM I2, 10 mM
LiI, and 0.1 M LiClO4 by using a Pt working electrode pre-
pared by one of the deposition methods, a Pt-wire auxiliary
electrode, andanAg/AgClreferenceelectrode. TheED-Ptwork-
in−g electrode indeed showed the largest reduction current for
I3 reduction and the largest oxidation current for I− oxida-
tion among the three Pt working electrodes, though the shapes
of the I3 reduction plots are somewhat different from each
other, probably due to the differences in the Pt film surface
conditions.
The high active surface area arising from a tailored
nanoporous nature of the ED-Pt film is advantageous for the
performance of a DSSC, because the catalytic activity depends
on the surface area of the catalyst, here the Pt film [14,19–21].
This is favorable for the reduction of I3− at the counter electrode,
and ED-Pt layers.
which in turn enhances the Jsc of a DSSC. We have measured
average sheet resistances of the three Pt films along with that of
a bare FTO glass substrate (Table 1). Surprisingly, the average
sheet resistance of the ED-Pt film is found to be far less than
those of the bare FTO glass (7.94 ꢀ/sq, not shown in the table)
and the films of TD-Pt and SD-Pt. This lowest sheet resistance
of the ED-Pt film may be attributed to its larger surface area, as
a result of the more porous structure, and probably to its strong
adhesion to the FTO substrate. To confirm the adhesion factor,
the Pt films were subjected to the standard adhesion test (ASTM
D 3359). The test revealed that the ED-Pt film adheres strongly
to its FTO substrate. We are not certain to make a comparative
statement with regard to the adhesion of the Pt films deposited
by the three methods, because the films were too thin on the
rough surface of the FTO glass plate to be able to differentiate
their adhesion strengths.
−
3.2. J–V curves of the DSSCs
Fig. 5 shows the J–V curves of the DSSCs with the TD-
Pt, SD-Pt, and ED-Pt counter electrodes. The photovoltaic
parameters are listed in Table 2. The short circuit photocurrent
density (Jsc) and the open-circuit voltage (Voc) of the DSSC
with the ED-Pt counter electrode with a deposition time of
900 s are 15.38 mA cm−2 and 0.73 V, compared with the Jsc
of 13.87 mA cm−2 and Voc of 0.70 V of the cell with a TD-
Pt counter electrode. Promoted by this increase in the Jsc and
Voc and an increase in the fill factor, the DSSC with the ED-
Table 2
Photovoltaic parameters of the DSSCs with the counter electrodes with TD-Pt,
SD-Pt, and ED-Pt layers
Deposition
Jsc (mA cm−2
)
Voc (V)
FF
η (%)
Fig. 4. Chronoamperometric curves with the Pt films in the electrolyte contain-
ing 10 mM LiI, 1 mM I2, and 0.1 M LiClO4 in acetonitrile. The curves were
obtained using the Pt working electrodes prepared by (a) thermal, (b) sputter,
and (c) electrochemical depositions.
TD
SD
ED
13.87
13.95
15.38
0.70
0.73
0.73
0.63
0.65
0.68
6.3
6.4
7.6

C.H. Yoon et al. / Electrochimica Acta 53 (2008) 2890–2896
2895
trodes, the difference in the fill factor in favor of the cell with the
ED-Pt electrode should be due to its lower charge transfer and
sheet resistances, relative to those of TD-Pt and SD-Pt counter
electrodes.
It is generally accepted that the counter electrode should have
good light reflecting characteristics to improve the light harvest-
ing by the sensitizing dye [6,8], which in turn would additionally
improve the Jsc of the pertinent DSSC. We measured light-
reflecting properties of the three electrodes with TD, SD, and
ED platinum layers. Table 1 shows the light reflectance values
of the electrodes at the wavelengths of 488, 633, and 830 nm.
It can be seen that the light reflectance of the ED-Pt film is far
higherthanthoseofTD-PtandSD-Ptfilmsinthevisibleandnear
IR region; this observation is qualitatively in consistency with
the Rrms values of the Pt films. The increased light reflectance in
the case of the ED-Pt counter electrode, relative to those of TD-
Pt and SD-Pt counter electrodes, is advantageous for reducing
the loss of the incident light due to transmittance, as evidenced
by the higher Jsc in the case of the cell with the ED-Pt counter
electrode.
Fig. 6. Nyquist plots of the DSSCs fabricated with the Pt counter electrodes pre-
pared by (a) electrochemical, (b) sputter, and (c) thermal depositions. The plots
were recorded over a frequency range of 0.05–105 Hz, with an AC amplitude of
5 mV.
Pt counter electrode (ED cell) has shown a solar-to-electricity
conversion efficiency of 7.6%, compared with 6.3% of the cell
prepared with the TD-Pt counter electrode. The DSSC with the
SD-Pt counter electrode has shown almost the same Jsc and
conversion efficiency as those of a cell with the TD-Pt counter
electrode. We observed that the Jsc and efficiency of the ED cell
increased gradually with the electrodeposition time up to 900 s,
and then remained at the same level up to 3600 s.
In this study, we focused on the preparation of Pt counter
electrodes by different deposition techniques and made a com-
parative investigation on the merits of such electrodes in terms of
photovoltaicparametersofDSSCs. Tofacilitatethiscomparative
investigation, we have not tried to improve the performance of
the DSSCs, e.g., by incorporating a TiO2 scattering layer, which
would certainly enhance the efficiency of the DSSCs [23]. In
eachcaseofdepositionofplatinumfilm, viz. thermal, sputterand
electrochemical depositions, we adopted their respective stan-
dard procedures, with the objective of obtaining the best possible
efficiency for the DSSCs under our experimental conditions. We
do not confirm the thickness of the platinum films to be exactly
the same, though care was taken to obtain films of equal thick-
ness. According to previous reports [2,3,24], thickness of the
platinum film has insignificant influence on the performance of
DSSCs and a 2 nm-thick Pt is sufficient to obtain a reasonable
charge transfer resistance.
The dependence of Jscs on the Pt counter electrodes was
further evidenced through electrochemical impedance spec-
troscopy. Fig. 6 compares resultant Nyquist plots of the DSSCs
with the three counter electrodes. The semicircles represent the
impedances related to charge transfer processes in the DSSCs.
As the variable is only the Pt counter electrode, contributions
to the impedance in virtue of the deposition methods are only
taken into consideration. The plots suggest qualitatively that
charge transfer resistance, Rct, of a cell increases in the order
of Rct of ED-Pt film < Rct of SD-Pt film < Rct of TD-Pt film. The
Rct change in this study essentially refers to the component of
resistance against electron transfer from the counter electrode
to I3−, because other elements of resistances are common to
all the DSSCs. A small Rct indicates a facile electron transfer,
whereas a large Rct implies a sluggish electron transfer [22]. The
smallest Rct value of the ED-Pt cell correlates well with the high-
est Jsc value of the corresponding DSSC, which is attributable
to the porous structure and increased surface area of the ED-Pt
film. This less charge transfer resistance at the ED-Pt/electrolyte
interface impli−es a more facile electron transfer from the counter
electrode to I3 ions and thus an enhanced Jsc for the pertinent
DSSC, compared with the Jscs of the cells with the TD-Pt and
SD-Pt counter electrodes. The increased fill factor of the DSSC
with the ED-Pt electrode, compared with those of cells with
TD-Pt and SD-Pt counter electrodes, also supports this obser-
vation. It is known that fill factor depends on the resistances
of the photoanode, electrolyte, and counter electrode. Compo-
nents of the resistances due to the photoanode and electrolyte
being the same for the cells with the different Pt counter elec-
4. Conclusions
Among the three counter electrodes prepared by depositing
their Pt films by electrochemical, sputter and most commonly-
employed thermal methods, the electrodeposited-Pt (ED-Pt)
counter electrode has rendered a solar-to-electricity conversion
efficiency of 7.6% for its DSSC, compared with approximately
6.4% of the cells prepared by using the other two types of counter
electrodes. Thesputter-depositedPtfilmyieldedalmostthesame
photovoltaic characteristics as the thermal deposited Pt film. The
improvedperformanceoftheDSSCwiththeED-Ptcounterelec-
trode is attributed to the increased active surface area, decreased
sheet resistance, and enhanced light reflection over the spectral
rangeextendingupto830 nmoftheED-Ptfilm, withreferenceto
the corresponding properties of the other Pt films. The improved
catalytic activity of the ED-Pt film for the reduction reaction
I3− + 2e− → 3I− and the decreased charge transfer resistance at
the counter electrode/electrolyte interface are correlated respec-
tively with the increased active surface area and the decreased

2896
C.H. Yoon et al. / Electrochimica Acta 53 (2008) 2890–2896
sheet resistance of the ED-Pt film. The enhanced Jsc is also
rationalized by the enhanced light reflection of the ED-Pt film,
which facilitates improved light harvesting by the sensitizing
dye.
[9] T.N. Murakami, S. Ito, Q. Wang, M.K. Nazeeruddin, T. Bessho, I. Cesar, P.
Liska, R. Humphry-Baker, P. Comte, P. Pe´chy, M. Gra¨tzel, J. Electrochem.
Soc. 153 (2006) A2255.
[10] Y. Saito, T. Kitamura, Y. Wada, S. Yanagida, Chem. Lett. 31 (2002) 1060.
[11] M. Biancardo, K. West, F.C. Krebs, J. Photochem. Photobiol. A: Chem.
187 (2007) 395.
Acknowledgments
[12] T. Ma, X. Fang, M. Akiyama, K. Inoue, H. Noma, E. Abe, J. Electroanal.
Chem. 574 (2004) 77.
[13] G. Wang, R. Lin, Y. Lin, X. Li, X. Zhou, X. Xiao, Electrochim. Acta 50
(2005) 5546.
[14] N. Papageorgiou, W.F. Maier, M. Gra¨tzel, J. Electrochem. Soc. 144 (1997)
876.
This work was supported by the Sol–Gel Innovation Project
and the MOCIE new and renewable energy R&D project under
contract 2006-N-PV12-P-05.
[15] G.S. Attard, C.G. Go¨ltner, J.M. Corker, S. Henke, R.H. Templer, Angew.
Chem. Int. Ed. Engl. 36 (1997) 1315.
[16] M.G. Kang, N.-G. Park, S.H. Chang, S.H. Choi, K.-J. Kim, Bull. Korean
Chem. Soc. 23 (2002) 140.
[17] P.A. Nelson, J.M. Elliott, G.S. Attard, J.R. Owen, Chem. Mater. 14 (2002)
524.
[18] K.J. Hansen, C.W. Tobias, J. Electrochem. Soc. 134 (1987) 2204.
[19] K.-S. Choi, E.W. Mc Farland, G.D. Stucky, Adv. Mater. 15 (2003) 2018.
[20] N. Papageorgiou, P. Liska, A. Kay, M. Gra¨tzel, J. Electrochem. Soc. 146
(1999) 898.
[21] M. Matsumiya, W. Shin, N. Izu, N. Murayama, Sens. Actuators B 93 (2003)
309.
[22] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamentals and
Applications, John Wiley & Sons, New York, 1980, p. 350.
[23] S. Hore, C. Vetter, R. Kern, H. Smit, A. Hinsch, Sol. Energy Mater. Sol.
Cells 90 (2006) 1176.
References
[1] K. Suzuki, M. Yamaguchi, M. Kumagai, S. Yanagida, Chem. Lett. 32(2003)
28.
[2] A. Hauch, A. Georg, Electrochim. Acta 46 (2001) 3457.
[3] X. Fang, T. Ma, G. Guan, M. Akiyama, T. Kida, E. Abe, J. Electroanal.
Chem. 570 (2004) 257.
[4] K. Imoto, K. Takahashi, T. Yamaguchi, T. Komura, J.-I. Nakamura, K.
Murata, Sol. Energy Mater. Sol. Cells 79 (2003) 459.
[5] T. Hino, Y. Ogawa, N. Kuramoto, Full Nanotubes Carb. Nanostruct. 14
(2006) 607.
[6] Z. Huang, X. Liu, K. Li, D. Li, Y. Luo, H. Li, W. Song, L. Chen, Q. Meng,
Electrochem. Commun. 9 (2007) 596.
[7] F. Cai, J. Chen, R. Xu, Chem. Lett. 35 (2006) 1266.
[8] M.K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Mu¨ller,
P. Liska, N. Vlachopoulos, M. Gra¨tzel, J. Am. Chem. Soc. 115 (1993)
6382.
[24] X. Fang, T. Ma, G. Guan, M. Akiyama, E. Abe, J. Photochem. Photobiol.
A: Chem. 164 (2004) 179.