CuIn_1-xGa_xSe₂ film has excellent photovoltaic performance due to its large optical absorption coefficient and direct-band gap. It was prepared by spin-coating and chemical co-reduction method which is a simple and easy way with low cost in this work. The surface morphology of the product film was observed using scanning electron microscope (SEM). The absorbance curves are measured by visible spectrophotometer. The phases of the samples were characterized by X-ray diffraction (XRD). It was found by phase analysis that prolonging the reaction time and increasing the reaction temperature were beneficial to the sample crystallization. With the doping concentration increasing, the surface morphology of CuIn_1-xGa_xSe₂ films changed with a tendency from spherical crystals to rods. The effect of doping concentration on the resistivity is not particularly obvious. As the doping concentration increases, the resistivity will increase slightly; When x=1, the resistivity changes greatly, which may be due to the poor film continuity. Their estimated band gaps of CuIn_1-xGa_xSe₂ films are 1.25 eV, 1.3 eV, 1.33 eV, 1.38 eV and 1.4 eV respectively.

Keywords: CuIn_1-xGa_xSe₂, morphology, band gap, photovoltaic, solar cell

The optimal band gap for a single junction solar cell is about 1.4 eV. Pure CuInSe₂ with a gap 1.04 eV has excellent photoelectric performance [1,2]. Doping is an important means to regulate the properties of materials [3-5], the enhancement of the thermoelectric properties of the SrTiO₃ by doping with Zr were investigated [6]. To increase the width of CuInSe₂ band gap, indium can be partially replaced by gallium and then it can obtain the compound of copper indium gallium diselenide (CIGSe). CIGSe polycrystalline thin film is recently being developed as an absorber material for thin-film photovoltaic solar cells, it is a multi-crystalline semiconductor of p-type with high-quality characteristics, such as its large optical absorption coefficient and direct-band gap [7]. Light trapping was studied in ultrathin CuIn_1-xGa_xSe₂ solar cells by dielectric nanoparticles [8]. CuInSe₂ nanowire arrays with core-shell structure were electrodeposited at various duty cycles into anodic alumina templates [9]. The influence of process parameters were investigated on the gallium composition of a CuIn_1-xGa_xSe₂ solar cell [10]. It reported the effects of substrate temperatures in the three-stage growth of CuIn_1-xGa_xSe₂ thin films and their photovoltaic performances [11]. The structural and optical properties of Cu-poor CuIn_1-xGa_xSe₂ films with different gallium contents prepared by co-evaporated technique were studied [12]. The effect of growth conditions on the properties of sputtered precursor thin films for CuIn_1-xGa_xSe₂ (CIGS) absorber layers was reported [13]. Single phase polycrystalline copper indium gallium diselenide thin-films for solar photovoltaic applications were fabricated by an economical two-stage method of pulsed current electrodeposition [14]. Other research works include simulations about CuIn_1-xGa_xSe₂ alloys [15], 3-stage deposition of CuIn_1-xGa_xSe₂ on Mo-coated glass and stainless steel substrates [16], CuIn_1-xGa_xSe₂ nanopowders or nanoparticles produced by solvothermal method [17, 18] and CuIn_1-xGa_xSe₂ prepared by electrodeposition [19]. It has reported our work about CuInSe₂ films prepared from chlorides under different conditions [20]. Quaternary compound CuIn_1-xGa_xSe₂ was prepared by spin-coating and chemical co-reduction method in this work.

The corresponding amounts of raw materials were weighed according to the stoichiometric ratio of CuIn_1-xGa_xSe₂, the precursor solution with the highest concentration was prepared, and the mixture was shaken by the ultrasonic cleaner to make it uniformly mixed. The CuIn_1-xGa_xSe₂ precursor film was prepared by spin-coating with 3,000 rpm for 10 seconds, and then the sample was placed in a reaction vessel containing hydrazine hydrate for heat treatment at a certain temperature. Then the CuIn_1-xGa_xSe₂ film sample was obtained after soaking in ionized water for 24 h and drying at room temperature. The surface morphology of the product film was observed using scanning electron microscope (SEM). The absorbance curves are measured by Model 723PC visible spectrophotometer made by Shanghai Precision Instrument Co., Ltd. According to the light absorption characteristics of the absorbance curve in the visible light region, the band gap widths of the CuInSe₂ film samples were estimated using extrapolation. The phases of the samples were characterized by X-ray diffraction (XRD).

CuIn_1-xGa_xSe₂ films were prepared by spin-coating and chemical co-reduction method
The phases of CuIn_1-xGa_xSe₂ films prepared by different reaction time
Fig. 1 shows the XRD patterns of CuIn_1-xGa_xSe₂ obtained with different reaction time of 5 h, 10 h and 20 h at 220 °C. It can be seen that the samples with long reaction time have higher XRD peak intensities, and the reaction time extension is favorable for the crystallization of the film samples. Compared with the standard PDF card with No.89-5649, the XRD patterns of the CuIn_0.9Ga_0.1Se₂ film samples in Fig.1a show three high peaks with 2θ angles at 26.67°, 44.32° and 52.47° that correspond to the (112), (220) and (312) crystal planes respectively; According to the standard PDF card with No.88-2043, the XRD peak of impurity phase Cu₂Se appears in Fig. 1(a), where the 2θ angle 27.12° corresponds to the (111) crystal plane. Fig. 1(b) indicates that the CuIn_0.8Ga_0.2Se₂ film samples reacted with 5 h and 10 h have no XRD peaks corresponding to the target phase while the XRD intensity are very low. The target XRD peaks occur in the sample with the reaction time 20 h; Impurity CuSe appeared in the sample obtained with reaction time 10 h. Fig. 1(c) and (d) show the XRD patterns of CuIn_0.7Ga_0.3Se₂ and CuIn_0.6Ga_0.4Se₂ film samples, compared with the standard PDF card with No.89-5649, the XRD peak positions with 2θ angles are slightly shifted to the right, while the impurity phase Cu₂Se also appeared in the sample reacted for 20 h in Fig. 1(c). It can be seen from Fig. 1(e) that the XRD intensities for the target phase of the CuIn_0.5Ga_0.5Se₂ sample are lower than others and increase with the reaction time prolongation. CuIn_0.4Ga_0.6Se₂ film samples with good crystallinity are obtained under different reaction conditions as shown in Fig. 1(f). It can be seen from Fig. 1(g) that the CuIn_0.3Ga_0.7Se₂ sample with reaction time 5 h has no XRD peaks. As the reaction time increases to 10 h and 20 h, the CuIn_0.3Ga_0.7Se₂ sample crystallizes better and better, but impurity phase Cu₂Se appeared when 20 h. Fig. 1(h) shows that the CuIn_0.2Ga_0.8Se₂ samples did not show the XRD peaks of the target phase with the reaction time 5 h and 10 h, and the XRD peak appeared when the reaction time extended to 20 h. Compared with the standard PDF card No.89-5649 in Fig. 1(i), the 2θ angle positions of XRD peaks for the CuIn_0.1Ga_0.9Se₂ samples are shifted to the right as a whole. The XRD peaks of the CuGaSe₂ sample correspond to the standard PDF card with No.31-456, and the 2θ angles with 27.7°, 46.11° and 54.25° correspond to (112), (204) and (312) crystal planes of CuGaSe₂ respectively as shown in Fig. 1(j).
 
Phases of CuIn_1-xGa_xSe₂ films prepared at different reaction temperatures
Fig. 2 shows the XRD patterns of CuIn_1-xGa_xSe₂ samples reacted at different temperatures for 20 h, compared with the standard PDF card with No.89-5649, the XRD peaks of 2θ angles of 26.67°, 44.32° and 52.47° correspond to (112), (220) and (312) crystal planes respectively. It can be seen from Fig. 2(a) that CuIn_0.9Ga_0.1Se₂ film samples are all with better crystallinity obtained at three temperatures, but impurity phase Cu₂Se also appears, among which CuIn_0.9Ga_0.1Se₂ sample prepared at 200 °C has best crystallinity. Fig. 2(b) indicates that CuIn_0.8Ga_0.2Se₂ film samples have impurity phase Cu₂Se while the sample with best crystallinity was obtained at 220 °C. Fig. 2(c) and Fig. 2(d) show that the XRD peak intensity and sharpness of CuIn_0.7Ga_0.3Se₂ and CuIn_0.6Ga_0.4Se₂ samples prepared at 220 °C are significantly higher than others, but Fig. 2(d) shows a higher XRD peak for impurity phase Cu₂Se. Fig. 2(e) and Fig. 2(f) show that the XRD peak intensities of CuIn_0.5Ga_0.5Se₂ and CuIn_0.4Ga_0.6Se₂ samples prepared at 220 °C are significantly higher than those obtained at 180 °C and 200 °C while impurity phase Cu₂Se appears at 220 °C. Fig. 2(g) shows that the XRD intensity of the CuIn_0.3Ga_0.7Se₂ sample prepared at 180 °C is very low and become greatly high for the sample obtained at 220 °C. Fig. 2(h) shows that there are no obvious XRD peaks for CuIn_0.2Ga_0.8Se₂ sample prepared at 180 °C, which indicates that no target phase appears, when the reaction temperatures were raised to 200 °C and 220 °C respectively, the XRD peaks for target phase appeared, the film sample prepared at 220 °C has higher XRD peak intensity and better crystallinity. Fig.2i shows that CuIn_0.1Ga_0.9Se₂ film sample obtained at 220 °C has higher XRD peak intensity and better crystallinity.
 
Comparison of XRD patterns of CuIn_1-xGa_xSe₂ prepared at 220 °C for 20 h
Fig. 3 shows the XRD patterns of CuIn_1-xGa_xSe₂ with doping concentrations of 0.2, 0.4, 0.6 and 0.8 and 1.0 prepared at 220 °C for 20 h. It can be seen that, as the Ga doping concentration increases, the characteristic XRD peaks of the samples gradually shift toward the large diffraction angles. For example, as shown along the (112) crystal plane, the 2θ angle of CuGaSe₂ is about 0.57° larger than that of CuIn_0.8Ga_0.2Se₂.

Analysis of electrical properties of CuIn_1-xGa_xSe₂ films
The resistivities of CuIn_1-xGa_xSe₂ films are shown in Fig. 4, and the influence of doping concentration on the resistivity are not obvious. As the doping concentration increases, the resistivity will increase slightly, and when x=1, the resistivity changes greatly, it may be due to poor continuity of the film sample.

Morphology analysis of CuIn_1-xGa_xSe₂ thin films
Fig. 5 shows the SEM images of CuIn_0.8Ga_0.2Se₂, CuIn_0.6Ga_0.4Se₂, CuIn_0.4Ga_0.6Se₂, CuIn_0.2Ga_0.8Se₂ and CuGaSe₂ film samples prepared at 220 °C for 20 h respectively. It can be seen that the CuIn_0.8Ga_0.2Se₂ and CuIn_0.6Ga_0.4Se₂ films are dense and continuous and composed of spherical crystals with diameters of about 0.2 to 0.3 μm, CuIn_0.6Ga_0.4Se₂ film has a non-uniform particle size. The CuIn_0.4Ga_0.6Se₂ film is composed of rod-like crystals with lengths of about 0.8 to 1.0 μm, and its shape is similar to rod shape gathered together with crystal grains. The CuIn_0.2Ga_0.8Se₂ film consists of small rod-like crystals with lengths of about 0.3 to 0.5 μm, and its length and size are smaller than those of the CuIn_0.4Ga_0.6Se₂ film. The CuGaSe₂ film is composed of rod-like crystals with lengths of about 0.8 to 1.0 μm and irregular sheet crystals.
 
Composition analysis of CuIn_1-xGa_xSe₂ film by EDS
Fig. 6 shows the EDS spectrums of CuIn_0.2Ga_0.8Se₂ film and CuGaSe₂ film prepared by spin-coating and co-reduction method. It can be seen from Fig. 6(a) that Cu, Se, In and Ga elements appear from left to right, it confirmed that the above four elements are contained in the product film, although the element ratio is not necessarily stoichiometric. Fig. 6(b) indicates that CuGaSe₂ film sample consists of Cu, Se and Ga elements.
 
Estimation of the band gaps of CuIn_1-xGa_xSe₂ thin films
The absorption spectrums of the film samples were tested using a UV-Vis method on a visible spectrophotometer. Fig. 7 shows the absorbance of CuIn_1-xGa_xSe₂ films with different doping concentrations prepared at 220 °C for 20 h. The band gaps of the film samples were then estimated by extrapolation as shown in Fig. 8. It can be seen that the estimated band gaps of CuIn_0.8Ga_0.2Se₂, CuIn_0.6Ga_0.4Se₂, CuIn_0.4Ga_0.6Se₂, CuIn_0.2Ga_0.8Se₂ and CuGaSe₂ films are 1.25 eV, 1.3 eV, 1.33 eV, 1.38 eV and 1.4 eV, respectively.

The CuIn_1-xGa_xSe₂ films were prepared by spin-coating and co-reduction method. It was found by phase analysis that longer reaction time and higher reaction temperature were beneficial to the sample crystallization, the better experimental conditions for preparing CuIn_1-xGa_xSe₂ films are at 220 °C for 20 h. With the increase of doping concentration, the surface morphology of CuIn_1-xGa_xSe₂ films changed with a tendency from spherical crystals to rods. The effect of doping concentration on the resistivity is not particularly obvious. As the doping concentration increases, the resistivity will increase slightly. When x=1, the resistivity changes greatly, which may be due to the poor continuity of the film. According to the absorbance of CuIn_1-xGa_xSe₂ films, their estimated band gaps are 1.25 eV, 1.3 eV, 1.33 eV, 1.38 eV and 1.4 eV, respectively.

This work was financially supported by the National Natural Science Foundation of China (No.51272140) and the Innovation Team of the Co-Innovation Center for Green Building of Shandong Province in Shandong Jianzhu University