The earth-abundant semiconductor Cu3BiS3 (CBS) exhibits promising photovoltaic properties and is often considered analogous to the solar absorbers copper indium gallium diselenide (CIGS) and copper zinc tin sulfide (CZTS) despite few device reports. The extent to which this is justifiable is explored via a thorough X-ray photoemission spectroscopy (XPS) analysis: spanning core levels, ionization potential, work function, surface contamination, cleaning, band alignment, and valence-band density of states. The XPS analysis overcomes and addresses the shortcomings of prior XPS studies of this material. Temperature-dependent absorption spectra determine a 1.2 eV direct band gap at room temperature; the widely reported 1.4-1.5 eV band gap is attributed to weak transitions from the low density of states of the topmost valence band previously being undetected. Density functional theory HSE06 + SOC calculations determine the band structure, optical transitions, and well-fitted absorption and Raman spectra. Valence band XPS spectra and model calculations find the CBS bonding to be superficially similar to CIGS and CZTS, but the Bi3+ cations (and formally occupied Bi 6s orbital) have fundamental impacts: giving a low ionization potential (4.98 eV), suggesting that the CdS window layer favored for CIGS and CZTS gives detrimental band alignment and should be rejected in favor of a better aligned material in order for CBS devices to progress.
Keywords: CuBiS; XPS; band gap; density functional theory; ionization potential; photoemission.