Crystallographic and optical properties and band diagrams of CuGaS2 and CuGa5S8 phases in Cu-poor Cu2S–Ga2S3 pseudo-binary system

Abstract

We synthesized Cu-poor Cu–Ga–S samples such, as CuGaS2 and CuGa5S8 with the composition of (1 − x)Cu2S–(x)Ga2S3 with 0.5 ≤ x ≤ 1.0, by a mechanochemical process and sequential heating. The crystal structure changes from tetragonal chalcopyrite-type CuGaS2 (0.5 ≤ x ≤ 0.55) to tetragonal stannite-type CuGa5S8 (x = 0.8). For samples with 0.60 ≤ x ≤ 0.75, the diffraction peaks were identified to be those of a mixed phase of the chalcopyrite- and stannite-type structures. The band-gap energies of Cu-poor Cu–Ga–S samples increase in a stepwise manner with increasing x. The band-gap energy of CuGa5S8 (x = 0.8) with the tetragonal stannite-type structure is approximately 2.66 eV, which is wider than that of chalcopyrite-type CuGaS2 (2.45 eV). The energy levels of valence band maxima (VBMs) were estimated from the ionization energies measured by photoemission yield spectroscopy (PYS). The energy levels of the VBM and conduction band minimum (CBM) of the Cu-poor Cu–Ga–S samples decrease significantly with increasing x (decreasing Cu/Ga ratio). The energy level of the VBM of CuGaS2 (−5.8 eV) is considerably deeper than those of CuInSe2 (−5.2 eV) and CuInS2 (−5.5 eV). The VBM of stannite-type CuGa5S8 with x = 0.8 (−6.4 eV) is much deeper than that of chalcopyrite-type CuGaS2 (−5.8 eV) and stannite-type CuIn3Se5 (−5.6 eV). In order to understand the band structures of chalcopyrite-type CuGaS2 and stannite-type CuGa5S8, we performed first-principles calculations using the Heyd–Scuseria–Ernzerhof (HSE06), nonlocal screened hybrid density functional method. The theoretical band-gap energy of stannite-type CuGa5S8 (2.2 eV) is wider than that of chalcopyrite-type CuGaS2 (2.0 eV). Both the theoretical and experimental band gaps of stannite-type CuGa5S8 are about 0.2 eV wider than those of chalcopyrite-type CuGaS2.