The SPT setup utilized is illustrated in Figure 1 [19] . A precursor solution which contains constituent reactant compounds is atomized in a nozzle to tiny droplets which are then sprayed onto a preheated substrate. The surface of the substrate must be sufficiently hot to initiate chemical reaction between the precursors in the droplet solution. Specifically, the droplet must still contain enough reactants in solution after reaching the substrate [20] . A film of stable compounds subsequently forms that adheres to the substrate due to chemical reaction and thermal decomposition of the solution.

The primary focus of the study was the processing of mixed oxide nanocomposite films (ZnO + In₂O₃ and SnO₂ + In₂O₃). However, the single oxides (ZnCl₂ and SnCl₄) were first investigated as benchmarks for the mixed oxides.

Zinc oxide was deposited using aqueous solution of zinc chloride (ZnCl₂) precursor while tin oxide utilized tin chloride (SnCl₄) precursor solution in water. The chemical reactions can be formulated thus [21] :

It should be noted that the reaction of tin chloride solution is a heterogeneous reaction that occurs in the vapor phase [12] .

In order to deposit mixed oxides, tin chloride and indium nitrate were dissolved together in water to synthesize SnO₂ + In₂O₃. Zinc chloride and indium nitrate were used for deposition of ZnO + In₂O₃ for different composition ratios. Each precursor was fed into a 2 mm-diameter duct at a flow rate of 1 ml/min [22] . The solution was injected through a 1 mm-diameter round spray nozzle and then atomized at 1 bar air pressure. The solution in the nozzle was then sprayed onto a heated aluminum oxide (Al₂O₃) coated substrate. The distance between the spray nozzle and the hot substrate was kept constant at 10 cm for

the experiments. This distance was chosen based on the results of previous studies [19] [23] . The thin film was finally annealed for 30 minutes after deposition to the desired structures at 450˚C [8] [24] [25] . The annealing process promotes adhesion of the film to the substrate. These conditions were kept constant for all other sets of experiments performed.

The substrate Temperature (T_s) and solution concentration (C) were used as the control parameters. In order to assess the results of mixed oxide deposition, the optimum case from single oxide deposition was also considered.

Two sets of experiments were designed for deposition of ZnO. In the first set of experiments (Cases 1 and 2), the concentration was maintained constant (C = 0.1 mol/lit) while the temperature was varied from 400˚C to 450˚C. In the second set (Cases 3 and 4), the concentration was increased from C = 0.2 mol/lit for Case 3 to 0.3 mol/lit for Case 4 while the temperature was constant at 400˚C [26] .

The same approach was used for deposition of SnO₂. In the first set (Cases 5 to 8), the substrate temperature (T_s) was varied from 350˚C to 500˚C at 50 degrees increments for different sets of experiments while the concentration was constant at 0.25 mol/lit. In the second set (Cases 9 and 10), the temperature was kept constant at 450˚C while the concentration was varied from 0.1 to 0.5 mol/lit [12] [27] .

In order to deposit mixed oxide films, the precursor concentration ratio was changed for new sets of experiments. Indium nitrate was mixed with zinc chloride to produce 25 wt% ZnO + 75 wt% (Case 11) and 80 wt% ZnO + 20 wt% In₂O₃ (Case 12). For these cases, temperature and solution concentration were kept constant at 400˚C and 0.1 mol/lit, respectively. For deposition of SnO₂ + In₂O₃, indium nitrate was added to produce 80 wt% SnO₂ + 20 wt% (Case 13) and 95 wt% SnO₂ + 5 wt% In₂O₃ (Case 14).

The deposited thin films were characterized by X-ray photoelectron spectroscopy (XPS) and Scanning electron microscopy (SEM). Surface morphology was studied using a Zeiss ULTRA-55 FEG SEM system which used Schottky field emission source with a resolution of 1 nm @ 15 KV and 1.7 nm @ 1 KV and STEM detector. Surface spectroscopy was performed by Physical Electronics 5400 ESCA system to detect elemental composition.

Figure 6 and Figure 7 show the SEM micrographs of ZnO + In₂O₃ and SnO₂ + In₂O₃ thin films on Al₂O₃ substrate. Figure 6(a) and Figure 6(b) are the results for ZnO + In₂O₃ at T_s = 400˚C and concentration ratios of 25% of ZnO and 80% of ZnO respectively. It has been found that an un-doped ZnO film will form a non-uniform structure [39] . By doping with indium, the uniformity is enhanced

and grain size is increased [39] . At low concentration ratio of ZnO (Figure 6(a)), ZnO nano wires are formed. By increasing the amount of ZnO (from 25% in Figure 6(a) to 80% in Figure 6(b)), crystallization occurs and a well-structured thin film is produced. These results exhibit the same trend that was observed in a previous Figure 2 for ZnO films. Thus by increasing the amount of zinc in solution, there is a better chance to produce zinc oxide crystals or rods. In Figure 6(b), the growth of ZnO rods occurs fully due to the growth of side walls, resulting in the formation of nano tubes [35] [40] [41] . By increasing the indium dopant, the grain size decreases. This trend is attributed to the stresses applied by the mixture which limits the growth of the grain size [42] .

Figure 7 shows the SEM micrographs of SnO₂ + In₂O₃ thin films deposited on Al₂O₃ substrate at T_s = 450˚C and at a concentration ratio of 80% of SnO₂ and 95% of SnO₂ respectively. The results show that small spherical particles are formed at the low concentration of tin (Figure 7(a)). When the concentration of tin is high (Figure 7(b)), the synthesized film is similar to the single oxide thin film of Case 5, resulting in cracking and formation of a thick film.

Image processing techniques were applied to the SEM micrographs in order to calculate the particle size distribution, specifically the mean diameter of each particle. The average particle size was determined by considering all particle dimensions in the film. The mean area of particles () was calculated by considering the total number of particles and the space intervals thus [43] :

where is actual mean area, A is the assumed mean area, i is the class interval between particles, d is the deviation of midpoint from assumed mean, and N is total number of particles considered. The mean diameter was then obtained from assuming spherical particles.

Table 2 shows the average particle size and standard deviation from average particle size obtained for some of the films produced. The standard deviation values show the dispersion of particle sizes from the measured average particle size for each case. Smaller values of standard deviations indicate narrower range of particle size distribution from average value. It implies simultaneous chemical reactions inside droplets and subsequently a homogenous particle growth on

the substrate due to the balance between the substrate temperature and the concentrations of precursors in solution. Figure 8 and Figure 9 show the particle size distributions obtained for ZnO single oxide and ZnO + In₂O₃ mixed oxide films respectively.

Figure 8 shows that by increasing the concentration of ZnO, the particles grow larger. At low concentration, the particles are smaller and the size distribution is less homogenous (Table 2, Case 1) as was also observed in a previous study [36] . At C = 0.2 mol/lit and T_s = 400˚C (Table 2, Case 3), a narrow particle size distribution is observed with high probability of a more homogenous film. Comparing the standard deviation values of cases 1 to 3 shows that more particles within 1σ is observed when solution concentration is high and subsequently the homogeneity of particle size is higher. Figure 9 for ZnO + In₂O₃ mixed oxide shows that the addition of indium oxide component to ZnO results in the growth of particles over a wider range of sizes. Higher standard variation values represents that there is more deviation from average size when a mixed precursor is deposited on the surface. It means that the particles in mixed oxides cases are produced with less homogeneity compared to the single oxides (ZnO) of Figure 8. In the mixed oxide film cases of Figure 9, the results show that a more homogenous particle distribution is obtained when ZnO is the dominant precursor. The experimental results show that by increasing the amount of ZnO, crystallization occurs and a well-structured thin film is produced. There is therefore a better chance to produce zinc oxide crystals or rods at the higher ZnO concentration [35] . Thus the overall particle size is reduced when the concentration of indium is increased (Table 1, Case 11). This trend may be attributed to the fact that indium ions limit the growth of ZnO particles on the surface [39] ) which has been also observed in a previous study of the authors [44] or by the increased stresses applied by the mixture which limits the growth of grain size [42] .

Figure 10 shows the particle size distribution for SnO₂ single oxide and Zno + In₂O₃ mixed oxide films. The single oxide is deposited in the shape of spherical

grains with sharp boundaries as indicated in a previous Figure 3(d). The mixed metal oxide is formed with rough structure with larger spherical grains as shown in the previous Figure 7(a). The results show that in the latter case, the particles agglomerate and form larger-sized particles as depicted in Figure 10 (Table 2, Cases 10 and 13).