Figure shows the XRD θ
spectra of the GZO thin films on glass substrates. The diffraction peak of GZO (0002) appears at 2θ
= 33.64°, which means that the c
-axis oriented GZO films were grown on glass substrates. GZO thin films are well known to have a hexagonal wurtzite structure with a preferred growth orientation along the c
-axis due to their lowest surface free energy. The GZO films on glass showed a very low resistivity of 5.95 × 10-4
Ω cm (sheet resistance of 6.53 Ω/sq), which is comparable to that of the commercially used FTO glasses. The carrier concentration and Hall mobility of the GZO films were 6.82 × 1020
and 15.37 cm2
/Vs, respectively. The high carrier concentration of the GZO films can be attributed to the substitution of Ga3+
caused by the supply of sufficient thermal energy [8
]. Furthermore, the deposition at the optimal substrate temperature leads to high mobility with the improvement in crystallinity. From these results, it is concluded that the deposited GZO films can be used for transparent electrodes. However, as mentioned above, a blocking layer should be needed in DSSC applications because of susceptibility to acid and oxidation at high temperature.
XRD θ-2θ pattern of the GZO thin films deposited on glass substrates.
Figure shows the optical transmittance spectra for the GZO glasses with the TiO2 blocking layer. The average transmittance is approximately 80% in the visible region, which implies that the fabricated GZO glasses are highly transparent enough to be applied to the DSSCs even after the deposition of the TiO2 blocking layer.
Optical transmittance spectra for the GZO glasses with the TiO2 blocking layer.
Figures and show the SEM images of the TiO2 nanoparticles coated on the GZO glasses with the TiO2 blocking layer and normal GZO glasses, respectively. It can be seen that the porous TiO2 nanoparticles adhere uniformly on both glasses. There is no large difference between them, but the morphology of the TiO2 nanoparticles on the TiO2 blocking layer is more ordered and spherical.
SEM images of the TiO2 nanoparticles. The TiO2 nanoparticles were coated on the (a) GZO glasses with the TiO2 blocking layer and (b) normal GZO glasses.
Figure shows the photocurrent density-voltage [J-V
] characteristics of the fabricated DSSCs using the GZO/TiO2
, GZO, and FTO glasses measured under one sun condition. The estimated photovoltaic parameters are summarized in Table . It should be noted that the performance of the DSSCs with the normal GZO glasses is relatively poor (3.36%) compared to that with the FTO though the GZO and FTO have similar electrical resistivity. However, when the TiO2
blocking layer is employed, the fabricated DSSCs show an improvement in the short-circuit current [Isc
] and fill factor [FF], and as a result, a conversion efficiency of 4.02% is obtained, which is 19.6% higher than that of the DSSCs with the normal GZO glasses. These results prove that the TiO2
blocking layer plays a role in protecting the GZO films as was expected. Furthermore, the DSSCs with the GZO/TiO2
glasses show slightly better characteristics than those with the commercially used FTO glasses, which is ascribed to the fact that the TiO2
blocking layer can prohibit the recombination of injected electrons in the GZO with the electrolyte effectively. In particular, the improvement in FF is clearly found, which is due to the improved electrical contact between the GZO and TiO2
J-V characteristics of the fabricated DSSCs using the GZO/TiO2, GZO, and FTO glasses.
Photovoltaic parameters of the fabricated DSSCs
Figure presents the Nyquist plots of the electrochemical impedance spectra of the fabricated DSSCs with and without the TiO2
blocking layer. It is known that the semicircles in the frequency regions of 103
, 1 to 103
, and 0.1 to 1 Hz are associated with the charge transport at the TiO2
/TCO or Pt/electrolyte interface, TiO2
/dye/electrolyte interface, and Nernstian diffusion in the electrolyte, respectively [10
]. The first circle of the DSSCs with the GZO/TiO2
glasses is smaller than that with the GZO glasses, which indicates that the charge transport at the TiO2
nanoparticles/GZO is easier with the TiO2
blocking layer. The larger shunt resistance of the DSSCs with the GZO/TiO2
glasses corresponding to the second circle is also seen. Therefore, the improvement in the efficiency with the TiO2
blocking layer can be explained by the small series resistance and large shunt resistance [11
Electrochemical impedance spectra of the fabricated DSSCs with and without the TiO2 blocking layer.