should be related to their +1 oxidation state. [ 9 ] Therefore, all alkali metals with the +1 oxidation state are expected to have positive effect on CZTSSe solar cells. However, no systematic studies have been reported on the effects of alkali metals dopants so far.In this paper, a systematic research of alkali-metals doping effects on CZTSSe solar cells performances was reported. Alkali metal-containing precursors were utilized to investigate various infl uences of alkali metals on CZTSSe solar cells. K-doped CZTSSe solar cells showed the best device performance in our research. Various thicknesses of CdS fi lms were deposited on K-passivated CZTSSe surface in order to further improve the solar cell effi ciency. Finally, over 8% PCE of K-doped CZTSSe solar cell with around 35 nm CdS has been reached.The major X-ray diffraction (XRD) peaks of the undoped CZTSSe fi lm shown in Figure 1 a can be indexed to zincblende CZTSe . [ 10 ] All (112) diffraction peaks of each of the alkali-metal-doped samples shifted to small diffraction angles comparing with the undoped sample (Figure 1 b). Because the reduction of diffraction angle represents the lattice constant increase, this means that we could raise lattice constants of CZTSSe by doping alkali metal. The larger lattice constant is attributed to the alkali metal atoms occupying Cu vacancy ( V Cu ) or substituting positions of CZTSSe atoms. Interestingly, the variation of diffraction angle is not linearly related to the radius of the alkali-metals. (112) peaks left-shift from an order of Li, Na, and then K-doped sample, whereas the (112) peak positions of Rb and Cs-doped samples shift back to larger diffraction angles. This indicates that the relatively smaller alkali-metal atoms could occupy the V Cu or substitute CZTSSe atoms, but it becomes diffi cult to replace CZTSSe atoms for Rb and even more so for Cs because of their larger atoms radii. (112) peak shift of Li-doped samples is the smallest among the alkali-metal-dopants. The Li ion radius (0.59 Å) is the same as Cu and Zn ion (0.60 Å), so the small shift should be mainly attributed to Li ions occupying the V Cu . Raman spectra of the selenized fi lms with and without doping alkali metals were shown in Figure S1 (Supporting Information). Intense Raman shifts at 172, 195, and 238 cm −1 attributed to Cu 2 ZnSnSe 4 phase were observed.[ 11 ] According to this observation, Raman spectra have proved that there is no secondary phase introduced when we intentionally doped alkali elements into CZTSSe fi lms. Besides, as the primary characteristic peaks in the Raman spectra do not change, the S/Se ratio is considered as constant for undoped and each of the alkali-doped samples, which further supports the fact that the shift of XRD peaks is not induced by the dissimilar degree of the selenization. Figure 2 shows the scanning electron microscopy (SEM) images of the alkali-metal-doped and undoped samples. TheCopper based direct band gap semiconductors used as thinfi lm photovoltaic materials have been investigated for a long...