*Corresponding author; E-mail: crcho@pusan.ac.kr
I. INTRODUCTION
Photovoltaic devices generate clean energy that can
reduce the world’s dependence on conventional energy sources. Dye-sensitized
solar cell (DSSCs) might be the most efficient and stable excitonic photocells
[1, 2]. DSSCs with a metal oxide nanostructure have been studied as a low-cost
alternative to silicon because of the simple manufacturing process, large-scale
production capability, and potential applications as semiconductor-based p-n
junction solar cells [3, 4]. Recently, ZnO was used as a candidate material for
replacing TiO2 because the energy-band structure of ZnO
with a wide band gap (3.37eV) and its physical properties are similar to the
energy-band structure and physical properties of TiO2. Furthermore,
ZnO has higher electronic mobility, which would be remarkable for electron
transport (a high exciton binding energy of 60 meV), with possible reduced
recombination loss [5].
In particular, the synthesis of one-dimensional
nanostructures of ZnO, replacing the existing nanoparticle structure, is
attracting a lot of interest owing to their unique properties and integration
into DSSCs [6]. On the other hand, the increase in efficiency achieved using
ZnO-based DSSCs has been limited because of the generation of the Zn2+/dye
complex at the interface between ZnO and dye. ZnO is immersed in a solution of
sensitizer dyes, and a small amount of Zn2+ ions from the ZnO surface are
dissolved in the solution. During the process, a Zn2+/dye complex is
formed on the ZnO surface. Among several dyes, N719 dye with two carboxyl
groups having very low acidity is used to reduce the formation of a Zn2+/N719
dye complex on the surface of ZnO [7]. To increase dye adsorption, recent
research has focused on the production of –O and –OH radicals on the surface of
the ZnO layer by using physical and chemical treatment [8-10]. Among the
various techniques for surface modification, plasma treatment is a simple,
low-cost, and mass-production method, however, the effect of an atmospheric-pressure (AP)
plasma on solar cell efficiency has not
been reported.
In this article, we report the growth of ZnO nanorods
(NRs) achieved using a chemical bath deposition (CBD) method. For surface
modification, the grown ZnO NRs were annealed and treated with an AP plasma containing
the reactive gases Ar/O2, Ar/N2, and Ar/H2.
The crystal structures, morphologies and surface chemical bonding states of the
ZnO NRs were investigated. In this paper, the results for the dye adsorption and the efficiency
of the ZnO NRs-based DSSCs are discussed for the first time.
II. EXPERIMENTS AND DISCUSSION
ZnO thin films were
deposited on a F-doped SnO2 (FTO)/glass substrates by RF magnetron sputtering. For
growing ZnO NRs, 400 ml of an aqueous solution containing 0.03-M zinc acetate
dehydrate (ZnAc) [Zn(O2CCH3)2(H2O)2,
98% purity] and 0.03-M hexamethylenetetramine (HMT) [(CH2)6N4,
99% purity] as a precursor was prepared at room temperature. The substrate was
immersed at the bottom of the beaker containing the prepared aqueous solution,
and the beaker was maintained at 92°C for 2.5 h for the growth of ZnO on the
substrate. The substrate was reversed during the growth process. After the
completion of the growth process, the NRs were rinsed with deionized water and
dried in air to remove any residual salt and organic materials (U-ZnO). After the NRs had been grown for 10 h, they
were annealed at 400°C for 2 h (A-ZnO) and were treated with an AP plasma (FemtoScience,
PlasmafluxTM) containing O2, N2 and H2
gases at a RF power of 100 W for 30 min (O-ZnO, N-ZnO and H-ZnO). The flow rate
of Ar was 40 sccm, and that of each of the reactive gases, i.e., O2,
N2, and H2 was 5 sccm. To examine their photovoltaic
behavior, we immersed the ZnO NRs in a solution of
cis-bis(isothiocyanato)bis(2,2’-bipyridyl-4,4’dicarboxylato)-ruthenium(II)bis
tetrabuthylammonium dye (N719, Solaronix) for 24 h at room temperature. A
dye-adsorbed ZnO electrode was assembled with a Pt counter electrode to form a
DSSC. One drop of an electrolyte solution (Iodolyte AG-50, Solaronix) was
infiltrated between the two electrodes. The physicochemical properties of the
ZnO NRs were investigated using field emission scanning electron microscopy
(FESEM) (S-4700, Hitachi), X-ray diffractometry (XRD) (X’pert pro, Phillips),
and X-ray photoelectron spectroscopy (XPS) (ESCALAB250, VG Scientific). The dye
adsorption of the NRs was measured using a UV-VIS-NIR spectrophotometer (Cary 5000,
Varian). The current density-voltage (J-V) characteristics of cells with active
areas of 0.25 cm2 were measured using an AM 1.5 solar simulator
fitted with a Xe lamp.
Figure 1 shows SEM images of the ZnO
NRs grown on ZnO/FTO/glass. The ZnO NRs had a hexagonal shape and were arranged
in a vertical array in the [001] direction on the ZnO seed layer. After a few
minutes, the NRs nucleated on the seeds in the ZnO thin layer. The diameters of
these NRs were significantly larger than those of the ZnO nanoparticle seeds in
a thin film. In the initial reaction step, after 2.5 h, the growth of ZnO NRs
stopped, and their average length and
diameter were approximately 1.75 μm and 76 nm, respectively (Fig. 1(a)). The
length and the diameter of the NRs increased with increasing duration of the
repeated reaction (5, 7.5, and 10 h) at the same precursor concentration, as
shown in Figs. 1(b-d). The average length and diameter
of the NRs after 10 h were approximately 3.8 μm and 150 nm, respectively. Moreover,
the NR diameter increased steadily during the growth period, but at a much slower rate than
the NR length. The growth rate of the ZnO NRs in the vertical direction was
very fast during the initial reaction; however, the rate reduced with increasing reaction time because of an increase
in NR growth in the radial direction. Figure 1(e) shows the morphology of A-ZnO NRs that were annealed to remove any residual
organics material and to increase the crystallinity. The shapes of the AP-plasma-treated O-ZnO, N-ZnO and H-ZnO NR samples did not show
any significant changes (Figs. 1(f-h)).
Structural
analyses of the NR samples grown on ZnO/FTO/glass for various reaction times
and annealed and treated with Ar/O2, Ar/N2 and Ar/H2
AP plasma (shown in Fig. 2) were carried out using XRD. For the ZnO NRs with a
hexagonal wurtzite structure, the (002) peak was dominant, and the peak
intensity increased with increasing duration of the repeated reaction because
of NR growth in the [002] direction, as shown in Figs. 2(b-d). The (002)/(101)
peak height ratio increased with time because only those NRs that had a c-axis
almost normal to the substrate, NRs that contributed to the (002) diffraction
peak, continued to grow. The peak intensity corresponding to A-ZnO NR was much
higher than that corresponding to U-ZnO NRs (Fig. 2(e)). This phenomenon was
attributed to the removal of residual organic molecules and an increase in
crystallinity, resulting from the arrangement of interior lattices. On the
other hand, there were no significant changes in the peak intensities
corresponding to the O-ZnO, N-ZnO and H-ZnO NRs samples.
XPS was
used to obtain more information about the surfaces of the ZnO NRs. Figure 3 shows
the surface chemical bonding states in the Zn 2p and O 1s XPS spectra. Figure 3(a)
shows the O 1s XPS spectra deconvoluted using a Gaussian-Lorentz fit. Herein, the
peak at the binding energy of 530.1 eV corresponds to the Zn-O bond in
the wurtzite structure. The peak at 531.5 eV was assigned to the -OH bond
in zinc oxyhydroxide species with an oxygen-deficient region in the ZnO matrix.
The binding energy of 532.3 eV was attributed to weakly bound oxygen, such as
the oxygen in adsorbed H2O and O2 [11]. The Zn 2p spectra
show a doublet peak at binding energies of 1021.2 eV and 1044.3 eV
corresponding to the Zn 2p3/2 and Zn 2p1/2 core levels,
respectively (Fig. 3(b)) [12]. The increase in the Zn 2p peak area of the
treated ZnO NRs was greater than that for the U-ZnO NRs but there was no change
in the peak area ratios of Zn 2p3/2 to Zn 2p1/2.
Figure 4 shows the absorption spectra and the J-V curves for the N719-sensitized
ZnO-NR-based DSSCs. A broad absorption band peaking at 520 nm can be observed,
which was assigned to the metal-to-ligand charge-transfer transition (MLCT) of
N719 [13-15]. The dyes was adsorbed on the ZnO NRs after the formation of H2O
in chemical reactions involving -O, -OH radicals on the surface of the ZnO NRs and -COOH radical in the dye molecules [16, 17]. The absorption intensity for
A-ZnO NRs was similar to that for U-ZnO NRs because of a decrease in the
concentration of the -OH radical and an
increase in the concentration of the -O radical. Furthermore,
dye adsorption of O-ZnO and N-ZnO NRs increased when -OH radicals were added during the Ar/O2, and the
Ar/N2 AP-plasma treatments.
Therefore, the intensity of this band, which is a characteristic of dye
adsorption, increased with increasing concentration of A-ZnO, O-ZnO and N-ZnO NRs.
In particular, O-ZnO NRs showed the highest dye adsorption because the increase
in the concentration of the -OH radical was more than
that for N-ZnO NRs. On the other hand, H-ZnO NRs showed less dye adsorption than
A-ZnO NRs because of a significantly decreased concentration of -O radicals on the NR surface.
The open-circuit voltage (Voc) of the U-ZnO was 0.63 V, but
those for the A-ZnO, O-ZnO, N-ZnO, and H-ZnO NRs were 0.71, 0.81, 0.80, and
0.80 V, respectively. The value of Voc for U-ZnO NRs was less than
those for A-ZnO NRs and plasma-treated ZnO NRs because of the presence of a
relatively large amount of residual organic molecules on the surfaces of the
ZnO NRs. The short-circuit current density (Jsc) of the U-ZnO NRs
(1.76 mA/cm2) was smaller than those for the A-ZnO NRs (2.41 mA/cm2),
O-ZnO NRs (3.37 mA/cm2), N-ZnO NRs (2.72 mA/cm2) and
H-ZnO NRs (2.18 mA/cm2). In particular, the value of Jsc was
highest for O-ZnO NRs and was relatively low for H-ZnO NRs as compared to N-ZnO
and A-ZnO NRs. Consequently, the PV efficiency of O-ZnO
NRs (1.21%) was higher than those of U-ZnO NRs (0.51%), A-ZnO NRs (0.83%),
N-ZnO NRs (0.99%), and H-ZnO NRs (0.79%). The significant change in the cell efficiencies of O-ZnO NRs and N-ZnO NRs based DSSCs might be
due to the analysis and measurement conditions. The XPS data were generally
measured for a few top NR layers instead of the sidewalls of the NRs. On the
other hand, the plasma affects the entire sample. The
effect of annealing and O2-AP-plasma treatment in series increased
the PV efficiency to 62% and 137% more than that of the untreated samples,
respectively.
III. CONCLUSION
The grown ZnO NRs were oriented along the [001] direction and had a
hexagonal wurtzite structure. The shape of the NRs was determined and the
length of the NRs were observed to increase with increasing reaction time. There were no
significant changes in the morphologies and structures of A-ZnO, O-ZnO, N-ZnO
and H-ZnO NRs after the annealing and plasma-treatment. The concentrations of -O and -OH radicals in
A-ZnO, O-ZnO, and N-ZnO NRs increased because of the supply of these polar groups (-O and -OH) produced in AP plasmas containing
reactive gases. O-ZnO NRs showed the highest level of
dye absorption because of the largest polar
components. Consequently,
O-ZnO-NR-based DSSCs showed higher efficiency (1.21%) than the other DSSC
samples because of the increased crystallinity and dye adsorption. Our results support the fundamental insights for enhancing the
efficiency of nanostructure based DSSCs by treating them with AP plasmas
containing reactive gases.
ACKNOWLEDGMENT
This study was supported by a Korea Research
Foundation Grant (KRF-2008-313-D00607).
REFERENCES
[1]
M. K. Nazeeruddin, P. Pechy and T. J. Renouard,
J. Am. Chem. Soc. 123, 1613
(2001).
[2]
P. Wang, M. S. Zakeeruddin,
J. E. Moser, K. M. Nazeeruddin, S. Takashi and M. Graetzel, Nature Mater. 2, 402 (2003).
[3]
M. K. Nazeeruddin, F. D. Angelis,
S. Fantacci, A. Selloni, G. Viscardi, P. Liska, S. Ito, B. Takeru and M.
Grätzel, J. Am. Chem. Soc. 127, 16835 (2005).
[4]
G. Oskam, Z. S. Hu, R. L. Penn, N.
Pesika and P. C. Searson, Phys. Rev. E 66, 011403 (2002).
[5]
Q. Zhang, C. S.
Dandeneau, X. Zhou, and G. Cao, Adv. Mater. 21, 4087 (2009).
[6]
H. Horiuchi, R. Katoh, K. Hara,
M. Yanagida, S. Murata, H. Arakawa and M. Tachiya, J. Phys. Chem. B 107, 2570
(2003).
[7]
F.
Yan, L. Huang, J.
Zheng, J. Huang, Z. Lin, F. Huang and M. Wei, Langmuir 26, 7153 (2010).
[8] J. Chung, J. Lee and S. Lim, Physica B 405, 2593 (2010).
[9] S. K. Park, C. Kim, J. H. Kim, J. Y. Bae and Y. S. Han, Curr.
App. Phys. 11, 131
(2011).
[10] C. Kim, J. T. Kim, H. Kim, S. H. Park, K. C. Son and Y. S. Han, Curr. App. Phys. 10, 176
(2010).
[11] X. G. Han, H. Z. He, Q.
Kuang, X. Zhou, X. H. Zhang, T. Xu, Z. X. Xie and L. S.
Zheng, J. Phys. Chem. C 113, 584 (2009).
[12] N. S. Ramgir,
D. J. Late, A. B. Bhise, M. A. More, I. S. Mulla, D. S. Joag and K.
Vijayamohanan, J. Phys. Chem. B 110, 18236 (2006).
[13] N. Chantarat, Y. W. Chen, S. Y. Chen and C. C. Lin,
Nanotechnology 20, 395201
(2009).
[14] G. Benkö, J. Kallioinen,
J. E. I. Korppi-Tommola, A. P. Yartsev and V. Sundström, J.
Am. Chem. Soc. 124, 489 (2002).
[15] Y. Zhang, Z. B. Xie and J.
Wang, Nanotechnology 20, 505602 (2009).
[16] K. Ahn, Y. S. Jeong,
H. U. Lee, S. Y. Jeong, H. S. Ahn, H. S. Kim, S. G. Yoon and C.
R. Cho, Thin Solid Films 518, 4066 (2010).
[17] F. M. Fowkes, J. Phys. Chem. 382 (1962).
