Chromium (III) acetylacetonate, acetone and methanol were reagents obtained from Sigma-Aldrich while Zinc acetate dihydrate (Zn(CH₃COO)₂・2H₂O) was obtained from Fisher Scientific. All reagents were used without further purification.

The single solid source precursor was prepared from the mixture of Zinc acetate dihydrate and chromium acetylacetonate. Zinc acetate (0.05 M) was dissolved and stirred vigorously with a magnetic stirrer in a mix of water, acetone, and methanol at 25˚C. A 0.05 M prepared solution of chromium acetylacetonate in acetone was used as doping source, introduced based on stoichiometry, such that the ratio of atomic concentration of Cr with respect to Zn were 1, 2, 3 and 5 at%. [11] . The resulting solution was oven dried to constant weight at 60˚C and pulverized before pyrolysis.

The thin films were deposited at 420˚C on a soda lime glass substrate with nitrogen gas as a carrier gas with a flow of 2.5 dm³/min (Figure 1). The soda-lime substrates were mounted on stainless steel blocks within the heating chamber to ensure effective thermal contact. The time of deposition was two hours.

The FTIR spectrum of the precursor, undoped and doped is shown in Figure 2(a) and Figure 2(b). The spectrum show that the basic absorption bands are between 4000 and 450 cm⁻¹. The major peaks are: OH stretching or M-OH-M band in range of 3400 - 3500 cm⁻¹, C-H vibration in range of 2922 - 2852 cm⁻¹, carbonyl stretching (C=O) at 1542 cm⁻¹, C-C stretching at 1200, 1100 1002 cm⁻¹ and M-O band at below 1000 cm⁻¹. The bands for C-C stretching and bending shifted to the lower frequencies in the case of Cr doping. This shows that the Zn-O bond has been altered due to Cr doping. The peaks in the range of 428 - 688 cm⁻¹ correspond to the Zn-O bond stretching. The bands observed at 790 cm⁻¹ could be attributed to Cr-O bond stretching.

The SEM micrographs of the ZnO thin films deposited at 420˚C as a function of the Cr content were shown in Figures 3(a)-(e). The surface morphological studies shows the formation of clusters like structure of undoped ZnO while with addition of Cr, the film showed the crystalline grains on the surface. It also reveals that the Cr concentration has great influence on the crystallinity and grain size. As the doping concentration increases, the films surface is smooth and the grains are uniform with agglomerated secondary grains, which are rounded in shape and compactly stacked up. Further increase in dopant concentration results in a compact leave-like structure. The primary grain size decreases as the Cr concentration increases. It was observed that the grain size is in the order of 125, 100, and 80 nm, respectively for films deposited with 1%, 3% and 5% Cr dopant. This shows that as dopants concentrations increases, the film surface compactness also increases.

The relative smoothness of the surface and variation in the grain size which is comparable to the previous work reported by Aguilar-Leyva et al. (2007) [19]

may be due to the effect of the acetone used as solvent in order to improve the crystallization [20] [21] . This may also affect the sensing properties.

Figure 4 shows the X-ray diffraction patterns of as deposited ZnO thin films as a function of Cr dopant concentration in solution. This pattern agreed well with the JCPDS standard (No. 36 - 1451) data of hexagonal wurtzite structured ZnO bulk powder.

The films were polycrystalline in nature with indexed diffractions peaks corresponding to (100), (002), (101) and (110) lattice planes. Diffraction peak of (002) plane were observed for all the deposited thin film irrespective of their doping concentration having preferential crystalline orientation along the c-axis; (002) plane. The common intensity peak (002) in all the spectra shows that the quality of the films are good [19] [22] . The peaks belonging to Cr metal, oxides or any other impurity were absent, indicating a complete substitution of Cr metal in host compound (ZnO). Jayaraman et al. (2017) [23] have reported similar phenomenon in doped ZnO. Full width at half maximum (FWHM) varies with dopants concentration, resulting in variation in the average crystallite size (thickness). The average crystallite size was determined using Debye Scherrer’s formula and found to be 85, 125, 110 100 and 80 nm for ZnO film with 0%, 1%, 2%, 3% and 5% Cr as dopant respectively. This implies that there is an initial increase in grain size with addition of Cr and subsequently decrease in the grain size with increase in Cr content.

The optical band gap energy (Eg) was determined from transmission measurements by analyzing the optical data with the expression for the optical absorption coefficient.

The effect of increasing Cr dopant concentration on the optical transmittance of the ZnO thin films are shown in Figure 5(a). Generally the transmission decreases with increase in Cr concentration. Similar observation had been reported by Amjid Iqbal et al. (2013) [18] . This effect may be due to the thickness of the films and doping effect. The observed decrease in the optical band gap in transition metal doped II-IV semiconductor compounds may be explain in terms of sp-d spin exchange interaction between the localized d electrons of substituting the cations (transition metal ions) and band electrons [24] [25] .

The absorption coefficient (α) is calculated from the relationship;

α = 1 d ln 1 T (1)

where d is the thickness of the film and T is the transmittance. For a direct band gap material, the absorption coefficient is related to the energy of the photon by the expression as follows;

( α h ν ) 2 = K ( h v − E g ) (2)

where K is a constant of proportionality, hv is the energy of the photon and E_g is the band energy.

The absorption coefficient a was determined using Equation (1) while band gap of the films were obtained from the extrapolation of the linear portion of the graph of (αhν)² versus E_g to α = 0. The optical band gap of the undoped ZnO was found to be 3.32 eV while that of doped ZnO ranges from 3.32 eV to 3.10 eV (Figure 5(b)) [24] [26] .

The SIMNTRA software was used to fit the RBS data with elemental profiles. The depth profiles of the blank substrate, undoped and doped thin films, and as well as RBS spectra of the film were shown in Figures 6(a)-(d), and Figure 7(a) and Figure 7(b) respectively. The depth profile of the films though semi-quantitative, shows a constant bulk concentration of Zn, O and Cr throughout the thin films along with an estimated ratio of the thickness of the thin films to that of the substrate. While depth profile of blank substrate reveals that silicon (si) and oxygen (O) are the major constituent with oxygen having a concentration 83.66 atm %. (Figure 6(a)). The depth profile of the films (0%, 3% and 5% Cr doped) shown in Figures 6(b)-(d) reveal almost constant elemental concentration of oxygen in all the deposited films and a constant reduction of O concentration to 70.05 at% in the substrate layer. The reduction of O concentration from 84.66 to 70.05 at% shows that there is a diffusion of O from the substrate layer into the deposited film layers which may result in an out-diffusion effect [27] . This may be due to deposition temperature. Hence, O contribution from the precursor to the stoichiometry of both doped and undoped ZnO is greatly masked by the diffused O from the substrate. The thickness of the film is estimated to range from 101 to 150 nm as obtained from the RBS spectrum. The fit in Figure (4b) gives a stoichiometry of Zn:Cr:O to be 53:24:45.

The I-V characteristic curves are shown in figures in Figure 8(a) and Figure 8(b). The figure reveal that the I-V curves are linear for all the doped and undoped thin films with resistivity ranging from 1.134 × 10⁻² Ω cm to 1.24 × 10² Ω cm at 300 K. The resistivity is observed to decrease with increase in Cr doping concentration up till 3% Cr doped. Further increase in Cr concentration at 5% Cr doped leads to increase in the resistivity. The initial increase may be due to improved crystallinity and decrease in the grain size with increase in Cr concentration which in turn may affect the degree of carrier mobility. The electrical conductivity also increase with increase in measured temperature following the above trend.

The gas sensing response of undoped and doped ZnO film is shown in Figure 9(a) and Figure 9(b). It is a plot of change in electrical resistance of ZnO:Cr thin film as a function of operating temperature for different concentration of Cr for a fixed concentration of ammonium in the gas chamber. Figure 9(a) and Figure 9(b) show that addition of Cr as a dopant enhanced the gas sensing response of the ZnO thin films. While Figure 9(b) revealed that the optimum operating temperature is around 200˚C irrespective of the doping concentration. At higher temperatures, the electrical resistance shows no significant variation with Cr doping concentration. This may be attributed to electrical stabilization of the surface with increase in Cr content and operating temperature which result in saturation of the surface thereby limiting the absorption-desorption reactions.