Yan et al. suggested that GBs in CIGS electrically benign and not harmful to photovoltaic due to not creating deep levels [16]. On the other hand, valence band maximum at GBs
acts as hole barriers, it reduces recombination at GBs [17]. Recently, Abou-Ras et al. identified Se-Se-terminated Σ3112 twin boundaries, indicating that Cu is depleted and In is enriched in the two atomic planes next to the twin boundary by high-resolution scanning transmission electron microscopy VX-765 molecular weight and electron energy-loss spectroscopy [18]. Takahashi group in Japan also reported that downward band bending of the conduction band and broadening of the band gap near GBs are observed by photo-assisted Kelvin probe force microscopy. It accounts for photo-carriers well separate LY2157299 and suppress the recombination at GBs [19]. Therefore, we have to investigate carefully carrier transport at GB in CZTSSe thin films, which is not yet clearly identified for the role of GBs. We already reported positive potential bending of GBs on CZTSe thin films, grown by electron co-evaporation, which showed 2% to 3% of conversion efficiency [20]. In this study, we investigate sputtered CZTSSe thin-film solar cells, which exhibit better device performance than the previous samples. We report local carrier transport and surface potential of CZTSSe thin films using conductive atomic force microscopy (C-AFM) and Kelvin
probe force microscopy (KPFM), respectively. For the complete understanding of the behaviors at GBs in CIGS films, recombination at GBs is diminished also due Lenvatinib mouse to downward band bending reduced density of deep-level in-gap states (i.e.,
recombination centers) and expect relatively efficient minority-carrier collection at GBs, as shown by scanning tunneling microscopy (STM) measurements [21, 22]. Future analysis using STM can be addressed for GBs of CZTSSe thin films. Method CZTSSe thin films were grown on Mo-coated soda-lime glass substrates. The metal precursor layers were deposited by radio frequency sputtering using Cu, ZnS and SnS targets. The staking order of the precursors in this study was Cu/SnS/ZnS/Mo/glass. Thickness of each stacked layer was changed from 0.4 to 0.7 μm. After the deposition, the precursors were annealed with Se metals in a furnace at 590°C for 20 min. Thickness of the annealed CZTSSe film was 1.8 μm for this study. From X-ray diffraction, the film shows single phase of CZTSSe without any significant second phases. We obtained the final composition is Cu/(Zn + Sn) ~ 0.94 and Zn/Sn ~ 1.65 of CZTSSe thin films by energy dispersive spectrometry (EDS). S/Se ratio is estimated to be approximately 0.1. The grain size indicates 1 to 2 μm of the CZTSSe thin film investigated by field emission scanning electron microscopy (FE-SEM) (JSM-700 F). KPFM and C-AFM measurements were carried out using a commercial AFM (n-Tracer, Nanofocus Inc., Seoul, South Korea).