Preparation and Characterization of Indium and Gallium doped Transparent ZnO films for Solar cell Applications

In this work, the effect of indium (In) and gallium (Ga) dopants on the structural, optical and electrical properties of ZnO thin films was studied. ZnO thin films were deposited on glass substrates at 400°C using the spray pyrolysis deposition technique. X-ray diffraction (XRD) results indicated that both undoped and doped ZnO films had (002) preferred orientation. The undoped ZnO films were found to exhibit high transmittance above 80%, while indium-doped (In:ZnO) and galliumdoped (Ga:ZnO) films had transmittance above 60% and 70% respectively. From the Hall Effect measurements, doping improved the conductivity of the ZnO thin films however, In:ZnO films showed higher electrical conductivity compared to Ga:ZnO films. Electron probe microanalysis (EPMA) results were used to confirm the presence of the respective dopants in the thin film samples.


INTRODUCTION
Solar cells consist of various materials, each performing a specific function which contributes to the overall efficiency of the solar cell. The Transparent Conducting Oxide/Film (TCO) is an important component of the solar cell design. This component is responsible for the collection and transporting electrons generated after the photoexcitation process. It is required that, the TCO be optically transparent and electrically conductive.
TCOs are also used in other electro-optic devices such as light emitting diodes (LED) touch and flat panel displays 1,2 . Various materials have been used as TCO in solar cells including tin doped indium oxide (ITO), fluorine-doped tin oxide (FTO) and indium-doped cadmium oxide (In : CdO). The abovementioned materials have disadvantages such as high cost, toxicity and unavailability. An alternative TCO material is Zinc oxide (ZnO). Zinc oxide is a wide-band-gap semiconductor with a direct band gap of 3.3 eV at room temperature 3 . The wide band gap makes zinc oxide transparent to light with high transmittance. The lattice parameters 'a' and 'c' of ZnO which has a wurtzite crystalline structure are 3.24 °A and 5.19 °A respectively 4,5 . The undoped ZnO has a very low electrical conductivity and hence requires doping to improve the conductivity. Factors that influence the choice of dopants include valency, electronegativity and ionic radius. Some of the common dopants used to improve conductivity of ZnO are aluminum and fluorine. Even though much work has been done to study the effects of these dopants on ZnO, there still remains some gaps in understanding the behaviour of these dopants [6][7][8][9] .
There are several thin film deposition techniques employed in the fabrication of TCOs 10,13 . These deposition systems can be classified as physical vapor deposition (PVD), chemical vapor deposition (CVD) and solution growth. Most PVD and CVD techniques require complex equipment setup and operate under low pressure or high vacuum conditions. This makes these techniques very expensive. An alternative technique which has a relatively simple setup, operates in atmospheric conditions and can be scaled up easily is spray pyrolysis. Spray pyrolysis has been used to grow ZnO films with very good structural, optical and electrical properties 14 .
In this work, indium doped ZnO and gallium doped ZnO thin films are prepared by spray pyrolysis and characterized to determine their optical, electrical and structural properties.

MATERIALS AND METHOD
The undoped ZnO, In-doped ZnO and Ga-doped ZnO thin films were deposited on soda lime glass substrates using spray pyrolysis. The glass substrates were washed in an ultrasonic cleaner for half an h using separate solutions of ethanol (95%), acetone and distilled water. After cleaning, the glass substrates were preheated at a temperature of 400°C to remove any traces of impurities that may be present even after cleaning. 0.2M aqueous solutions of zinc acetate (99.9% -Sigma Aldrich) were used as the precursor solution for the deposition process. The sources of indium and gallium dopants were indium (III) chloride and gallium oxide respectively and the doped ZnO films were prepared by mixing known concentration of aqueous solutions of indium (III) chloride and gallium oxide to the zinc acetate solution. The films were deposited at a temperature of 400° in a non-vacuum environment. The transmittance of the films was measured using a Genesys 10S UV-Vis spectrophotometer. The parameters used were a minimum wavelength of 350nm, with scan interval 15nm and a maximum wavelength of 800nm.
The surface morphology of the films were studied using a field emission gun scanning electron microscope (FEG-SEM), JEOL 5800 and operating in the secondary mode.
The XRD analysis was performed using a scan axis of 2θ from 20.00 to 80.00°. A step size of 0.2 was also programmed. The anode material for the X-ray diffractometer was Cu with Kα radiation of wavelength 0.154nm.
Hall Effect Measurements was carried out on doped zinc oxide thin films. The doped samples were subjected to a magnetic field with electric current passing through the magnetic field in a perpendicular direction. Molten lead was deposited at the four corners of the sample and allowed to solidify to form contacts. Thin wires were embedded in the lead contacts and these served as ports for the probes. The experiment was carried out at a temperature of 300K (27°C). For the purposes of this experiment, a magnetic field of 300mT was created by supplying current to a pair of electromagnets. The magnetic strength was measured with a Phywe Teslameter. For each sample (In:ZnO 1wt% -5wt% / Ga:ZnO 1wt% -5wt%), a minimum current of -10mA (reverse current) was applied. The current was varied by a step size of 1mA up to a maximum of 10mA. At each current, the corresponding voltage created was recorded using a digital multimeter. Voltage correction was done based on the voltages recorded at 0mA for each sample to attain the actual Hall voltages (VH). A plot of the input current against the Hall voltage was generated and the slope of the graph used to determine the Hall coefficient.
Electron Probe Microanalysis (EPMA) using JEOL 5800 was carried out on In:ZnO and Ga:ZnO samples to ascertain the presence of the intended dopants and other elements in the ZnO structure. The analysis also gave further information about the relative concentration of these elements.

RESULTS AND DISCUSSIONS
Optical transmittance transmittance may be due to the homogeneity of the precursor solution. As the dopant concentration was increased, incomplete dissolution of the dopant compounds in the precursor solution could have affected the film formation process during the deposition 16,17 . The 1wt% Ga:ZnO sample recorded the highest average transparency (87.56%). The aim of carrying out this characterization was to find out how much light the TCO transmits and absorbs within the ultraviolet and visible range of the electromagnetic spectrum (EMS). For the In:ZnO samples, the optical transmittance increased steadily until a concentration of 5wt% sample where there was a drop (Fig.1a). The increase in optical transmittance for the In:ZnO may be attributed to good crystal properties. As the crystallinity of the ZnO structure improves, less scattering of light occurs and the transmittance values increase. The average optical transmittance of the In:ZnO was above 60% with the 4wt% In:ZnO sample recording the highest average transmittance value of 72.98%. For the Ga:ZnO thin films, there was a steady decrease in the transparency of the films as the dopant concentration was increased as shown in Fig. 1b. This will cause high light scattering that will affect the transmittance values. The drop in The band gap for the doped samples was obtained by extrapolating the linear portion of the Tauc plot of (αhν) 2 against hν as shown in Fig. 2. The band gap in eV is the intercept of the extrapolated line on the x-axis. The undoped ZnO samples had an average band gap of 3.25eV. In Fig. 2a, the band gap of 1wt% In:ZnO and 2wt% were 3.125 and 3.129eV respectively. The Ga:ZnO films recorded an average of 3.25eV as is in Fig. 2b. Comparing the band gap values for the undoped ZnO, In:ZnO and Ga:ZnO, it could be suggested that, the reduction in resistivity of the In:ZnO can be attributed to the narrowing of the band gap 18,19 . Also, the band gap of Ga:ZnO did not change significantly from the undoped ZnO hence the effect of the gallium dopant in reducing the resistivity was not as significant as In:ZnO. The X-ray diffraction analysis was used to study the crystal structure of the doped zinc oxide samples and determine if a peak shift occurred because of the doping of ZnO. Fig. 3 shows the X-ray diffraction pattern for the doped ZnO thin films. The XRD diffraction peaks occurring at 2θ=31.09, 34.54, 36.34, 56.63 and 62.94 correspond to the (100), (002), (101), (110), (103), planes of the hexagonal structure of the doped ZnO analyzed. The most dominant peak for both 1wt% -indium doped zinc oxide (In:ZnO) and gallium doped zinc oxide (Ga:ZnO) was the (002) peak. However, the (002) peak of the Ga:ZnO sample had higher intensity as compared to the In:ZnO sample. The peak for the preferred (002) orientation occurred at 34.414° and 34.422° for In:ZnO and Ga:ZnO respectively as expected 14 . From this data, it can be said that, there was no peak shift after doping ZnO with indium and Gallium. Other less dominant peaks that showed on the XRD pattern for the In:ZnO sample were (100), (101), (110) and (103). This can be attributed to the difference in atomic radii of indium and zinc [20]. Indium has a larger atomic radius (1.56Å) compared to zinc (1.42Å). The difference in atomic radius causes stress in the ZnO structure when indium replaces zinc. This is seen in the various peaks except that of the preferred (002) orientation in the In:ZnO sample. Gallium however has a smaller atomic radius (1.36Å) compared to zinc and indium. From  Fig. 5, it is observed that, the (002) orientation from Ga:ZnO is most dominant and the other peaks have less intensities when compared. The smaller atomic radius of gallium allows it fit with less strain on the zinc oxide structure as compared to indium. This therefore explains the difference in peak intensities for the (002) peaks and the appearance of other peaks and corresponding low intensities for both samples. The surface morphology of In:ZnO and Ga:ZnO doped films made up of 5wt % were studied using the scanning electron microscope (see Fig. 4). In Fig. 4A and 4B, it is evident that, there is a distinct variation in the surface morphology of In:ZnO and Ga:ZnO doped films. Ga:ZnO (Fig. 4A), had cylindrical rod-shaped crystallites while that of In:ZnO (Fig. 4B), had characteristic hexagonal slices. These plate-like structures spread across the surface fused together and overlapping to give a rough surface. The Ga:ZnO crystallites however were clustered in pockets with varying diameters. The difference in crystallite structure and distribution serves as an explanation for the different optical transparency values recorded.

CONCLUSION
UV-Vis spectroscopy, Hall Effect, X-ray diffraction, and EPMA analysis were carried out on the undoped and doped ZnO thin films prepared by spray pyrolysis technique. U V-Vis spectrophotometry analysis showed that the In:ZnO thin films had transparency above 60% and the transparency increased with increasing dopant concentration. For the Ga:ZnO thin films, a maximum transparency of 88% was recorded for the 1wt% sample. From the Hall Effect measurement and sheet resistance analysis, the conductivity, electron mobility and carrier concentration decreased with increasing dopant concentration. The In:ZnO samples however exhibited higher conductivities, mobility and charge concentration compared to the Ga:ZnO.
The X-ray diffraction results revealed that both In:ZnO and Ga:ZnO thin films were highly oriented with dominant (002) peaks.
Electron probe microanalysis results showed that, both indium and gallium were present in their respective doped films. The composition of indium in mass percent for both In:ZnO samples was higher than that of gallium in the respective Ga:ZnO samples. This indicated that, more indium dopant had been successfully incorporated in the ZnO during the spray pyrolysis process. Based on the results obtained from the doped samples both doped thin films can be used TCOs in opto-electronic devices such as solar cells.