A high voltage dye-sensitized solar cell using a nanoporous NiO photocathode

Article abstract of DOI:10.1246/cl.2005.500

An np tandem dye-sensitized solar cell (np-DSC) comprising a titanium dioxide (TiO₂) anode and a nickel oxide (NiO) cathode prepared as porous nanostructured films by a sol-gel method using a triblock copolymer template (P123, HO(OCH₂-CH₂)₂₀(OCH(CH ₃)CH₂)₇₀(OCH₂CH₂) ₂₀OH))CH₂ exhibited a high open circuit voltage (V _oc) of 0.92 V that surpassed the long time champion data, ≈0.8 V for conventional DSC. Copyright

Full text of DOI:10.1246/cl.2005.500

5
00
Chemistry Letters Vol.34, No.4 (2005)
A High Voltage Dye-sensitized Solar Cell using a Nanoporous NiO Photocathode
y
ꢀ
Akihiko Nakasa, Hisanao Usami, Seiichi Sumikura, Satoshi Hasegawa, Toshiki Koyama, and Eiji Suzuki
Department of Fine Material Engineering, Shinshu University, 3-15-1, Tokida, Ueda 386-8567
y
Department of Functional Polymer Science, Faculty of Textile Science and Technology,
Shinshu University, 3-15-1, Tokida, Ueda 386-8567
(Received November 29, 2004; CL-041437)
An np tandem dye-sensitized solar cell (np-DSC) compris-
DSC. When irradiated from the anode side, the solar light is part-
ly absorbed by dye-1, yielding Voc1. The light passing through
the dyed anode and the electrolyte reaches the cathode and is ab-
sorbed by dye-2, yielding Voc2. Thus an np-DSC is able to utilize
wide spectrum of solar light when the dye-2 is designed to ab-
sorb the longer wavelength light in comparison with the dye-1.
Lindquist and co-workers fabricated a np-DSC comprising a
ing a titanium dioxide (TiO2) anode and a nickel oxide (NiO)
cathode prepared as porous nanostructured films by a sol–gel
method using a triblock copolymer template (P123, HO(OCH2-
CH2)20(OCH(CH3)CH2)70(OCH2CH2)20OH))CH2 exhibited a
high open circuit voltage (Voc) of 0.92 V that surpassed the long
time champion data, ꢁ0:8 V for conventional DSC.
ꢂ
ꢂ
conventional TiO2 dye-sensitized anode, an I /I3 electrolyte
3
and a dye-sensitized NiO cathode. Voc of 0.73 V reported by
Since Gr a¨ tzel and co-workers reported a highly efficient
DSC,1,2 it has been extensively investigated as one of the most
interesting alternatives to current solar cell technology for the
conversion of sunlight into electrical energy. Typical DSC com-
prises a dye-adsorbing porous, nanocrystalline titanium dioxide
them was, however, still lower than the best Voc of n-DSC. In
our work, the anode and the cathode are dye-sensitized TiO2
and NiO films, respectively, as used by Lindquist and co-work-
ers, but we prepared them differently, intending to increase dye-
adsorbing surface area of NiO film. The both photoelectrodes
were prepared by a sol–gel method using P123 triblock copoly-
mer as a nanostructure template: P123 has been applied for prep-
ꢂ
ꢂ
film interpenetrated by a liquid electrolyte containing an I /I3
red/ox couple. Because of n-type behavior of the dye-sensitized
titanium oxide electrode, the conventional DSC is also called
n-DSC. Although many researchers have tried to improve the
performance, that is, Voc, short circuit current density (Jsc),
and the fill factor, reported more than ten years ago by Gr a¨ tzel
group,1 no one has significantly surpassed their data.
4
,5
aration of crack-free nanoporous films.
The preparation of photoanode TiO2 film is described briefly
in the following. 1.25 g of P123 was added to 3 g of titanium iso-
propoxide dissolved in a mixed solvent of acetyl acetone and 2-
propanol, and stirred for 3 h. Then the solution was spin-coated
(400 rpm for 3 s, then 2000 rpm for 10 s) on SnO2: F conducting
glass (FTO) (Asahi glass co. ltd., 10 ꢀ/ ), and dried at room
temperature. This process was repeated three times. The FTO
substrate was immersed in an alkaline bath for 2 h before the spin
coating. Finally, the spin-coated precursor film was calcined at
,2
In this paper, we will report on an np-DSC exhibiting a Voc,
ꢁ
0:92 V, significantly higher than the best Voc, ꢁ0:8 V, of con-
1,2
ventional DSC reported by Gr a¨ tzel and co-workers. Replace-
ment of a photo-inactive cathode of a Gr a¨ tzel type cell by a
dye-sensitized p-type semiconductor cathode results in an np-
DSC. Figure 1 illustrates principle of photocurrent generation
by np-DSC. Voc of np-DSC is equal to Voc1 þ Voc2, where
Voc1 is equal or close to Voc of single n-DSC. Hence Voc of
np-DSC is intrinsically higher than that of conventional n-
ꢃ
450 C for 30 min in air. Nanostructured NiO film for photocath-
odes was similarly prepared by a sol–gel method using NiCl2 as
a source material and P123 as a template. The TiO2 and NiO
films were immersed into dehydrated ethanol solution contain-
ꢂ4
0
0
ing 10 M of cis-di(thiocyanato)bis(4,4 -dicarboxy-2-2 -bipyri-
dine) ruthenium(II) (Kojima Chemicals Co. Ltd., Ru(dcbpy)2-
photoanode
CB
photocathode
CB
(
nyldene)ethylidene]-2-thioxo-4-thiazolidine (MC, Hayashibara
NCS)2) or 3-carboxymethyl-5-[2-(3-octadecyl-2-benzothiazoli-
6
e
Biochemical Laboratories, Inc.) for 12 h, respectively.
Ru(dcbpy)2(NCS)2 and MC were used without purification.
Prior to fabrication of DSC, the dye-coated TiO2 and NiO films
were rinsed with dehydrated ethanol and dried using dry N2 gas.
Counter electrodes for n- and p-DSC were prepared by deposit-
ing Pt/Pd on ITO by sputtering with Pt/Pd (7:3) target. DSC
were fabricated by coupling the TiO2 anode with the counter-
electrode for n-DSC, the counter electrode with the NiO cathode
for p-DSC, and the TiO2 anode with the NiO cathode for np-
DSC, and filling the space between the both electrodes with an
Voc1 e
h
h
υ
e
V
oc
2
dye1
υ 2
VB
e
VB
dye2
ꢂ
ꢂ
I /I3 electrolyte PN50 (Solaronix SA). The active cell area
ꢂ2
2
electrolyte
was 0.12 cm . The cells were irradiated at 100 mWcm with
a light source simulating AM 1.5 global solar radiation. The
photocurrent–bias voltage characteristics were obtained using
ꢂ1
a standard three-electrode setup. Scanning rate was 10 mVs
7
.
FE-SEM images, reported elsewhere, of the TiO2 (anatase
Figure 1. Principle of photocurrent generation by np-DSC.
Copyright ꢀ 2005 The Chemical Society of Japan

Chemistry Letters Vol.34, No.4 (2005)
501
8
Table 1. The characterization of various type DSCs
b
c
Jsc
mAcm
Voc
/V
FF
/–
ꢀ
/%
6
4
2
0
2
4
Photocurrent
(a)
type
ꢂ2
/
n
p
5.83
1.00
3.62
2.73
0.762
0.093
0.918
0.887
ꢂ2
0.53
0.30
0.19
0.32
2.360
0.027
0.660
0.780
a
a
np (n-side)
np (p-side)
0
0.2
0.4
0.6
0.8
1
All date collected with 100 mWcm solar illumination.
paragraph is illumination side.
FF is fill factor.
-
-
Darkcurrent
a
b
c
ꢀ
is photoconversion efficiency.
Bias voltage / V
The np-DSC exhibited Voc of 0.92 V which is higher than the
1
.2
best Voc, ꢁ0:8 V, of conventional DSC (Figure 2c). This result
demonstrated that the np tandem would be a suitable structure
for a high voltage type DSC. Jsc and Voc of our np-DSC were
higher by approximately 70 and 27%, respectively, than the re-
ported np-DSC data. Jsc of our np-DSC, 3.7 mAcm , was
smaller than that of our n-DSC, which was presumably due to
the insufficient number of carriers generated by the photoca-
thode. We expect that increase of NiO film thickness would en-
hance dye adsorption, resulting in a higher current density.
Choice of dyes is also important to increase the current density.
The combination of dyes, Ru(dcbpy)2(NCS)2 as dye-1 and MC
as dye-2, applied in this work is not optimal, because the absorp-
tion edge of dye-2, 589 nm, was shorter than that of dye-1,
1
(b)
Photocurrent
Lindquist et al.
0
0
0
0
.8
.6
.4
.2
0
3
ꢂ2
-
-
-
0.2 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12
Darkcurrent
0.4
0.6
Bias Voltage / V
6
7
50 nm.
Difference of the irradiation direction, from p- or n-side, did
5
4
3
2
1
0
1
2
(c)
n-side
not make significant difference of the np-DSC performance. This
suggested that the cathode absorbed minor amount of the light,
giving insignificant difference of irradiation intensity at the
anode whether it was irradiated directly or through the cathode.
The performance of n-, p-, and np-DSC is summarized in
Table 1.
In conclusion, np-tandem DSC can generate a Voc higher
than n-DSC. When carrier generation at the photocathode is aug-
mented to a level at the photoanode of conventional n-DSC, Jsc
and the solar-to-electricity conversion efficiency (ꢀ) of np-DSC
would approach and surpass, respectively, those of conventional
n-DSC. Further development including enhancement of dye ad-
sorption at the cathode and choice of optimum combination of
dyes is required to increase cathode photocurrent.
Photocurrent
p-side
0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Darkcurrent
1
-
-
Bias voltage / V
Figure 2. The photocurrent vs bais voltage of various type
DSCs. (a) n-DSC, (b) p-DSC, (c) np-DSC, and n-side and p-side
shows illumination direction.
phase) and NiO films indicated the formation of nanoparticles of
10–30 and 30–40 nm diameters, respectively.
The photocurrent density–bias voltage curves of n-, p-, and
np-DSC are shown in Figure 2. The n-DSC exhibited a Voc of
.76 V (Figure 2a), which is comparable to that of well-prepared
This work was supported by the Cooperative Link for
Unique Science and Technology for Economy Revitalization
(
CLUSTER) of Japan’s Ministry of Education, Culture, Sports,
0
Science and Technology. And Dr. A. Nakasa was supported
by the Ministry’s 21st Century COE program.
conventional n-DSC. The result indicated that the sol–gel meth-
od using P123 as a template could be an alternative method for
preparing nanostructured TiO2 anodes of n-DSC. Although the
Jsc of our n-DSC was low, we expect that further increase in
the film thickness to 10 mm from ꢄ1 mm of this work would raise
Jsc to the level of conventional n-DSC. The photocurrent densi-
ty–bias voltage curve of our p-DSC is shown together with that
of the pioneering work by Lindquist and co-workers (Figure 2b);
Jsc and Voc are fivefold and by ꢁ15%, respectively, larger than
References
1
2
B. O’Regan and M. Gr a¨ tzel, Nature, 353, 737 (1991).
M. K. Nazeeruddin, A. Kay, I. Rodicio, R. Baker Humphry, E. Muller,
P. Liska, N. Vlachopoluos, and M. Gr a¨ tzel, J. Am. Chem. Soc., 115,
6
382 (1993).
3
4
5
6
7
J. He, H. Lindstr o¨ m, A. Hagfeldt, and S.-E. Lindquist, Sol. Energy
Mater. Sol. Cells, 62, 272 (2000).
H. Yun, K. Miyazawa, H. Zhou, I. Honma, and M. Kuwabara, Adv.
Mater., 13, 1377 (2001).
N. Bao, K. Yanagisawa, X. Lu, and X. Feng, Chem. Lett., 33, 346
3
their results. We presumed that the improvement of NiO nano-
structure owing to the application of the sol–gel method using
P123 increased dye-adsorption surface area, which enhanced
light absorption, resulting in increased number of photoactivated
electrons.
(
2004).
K. Sayama, K. Hara, N. Mori, M. Satsuki, S. Suga, S. Tsukagoshi,
Y. Abe, H. Sugihara, and H. Arakawa, Chem. Commun., 2000, 1173.
A. Nakasa, E. Suzuki, H. Usami, and H. Fujimatsu, submitted.
Published on the web (Advance View) March 5, 2005; DOI 10.1246/cl.2005.500