Electronic and optical properties of Cu2XSnS4 (X = Be, Mg, Ca, Mn, Fe, and Ni) and the impact of native defect pairs

Abstract

Reducing or controlling cation disorder in Cu2ZnSnS4 is a major challenge, mainly due to low formation energies of the anti-site pair (Cu−Zn + Zn+Cu) and the compensated Cu vacancy (V−Cu + Zn+Cu). We study the electronic and optical properties of Cu2XSnS4 (CXTS, with X = Be, Mg, Ca, Mn, Fe, and Ni) and the impact of defect pairs, by employing the first-principles method within the density functional theory. The calculations indicate that these compounds can be grown in either the kesterite or stannite tetragonal phase, except Cu2CaSnS4 which seems to be unstable also in its trigonal phase. In the tetragonal phase, all six compounds have rather similar electronic band structures, suitable band-gap energies Eg for photovoltaic applications, as well as good absorption coefficients α(ω). However, the formation of the defect pairs (CuX + XCu) and (VCu + XCu) is an issue for these compounds, especially considering the anti-site pair which has formation energy in the order of ∼0.3 eV. The (CuX + XCu) pair narrows the energy gap by typically ΔEg ≈ 0.1–0.3 eV, but for Cu2NiSnS4, the complex yields localized in-gap states. Due to the low formation energy of (CuX + XCu), we conclude that it is difficult to avoid disordering from the high concentration of anti-site pairs. The defect concentration in Cu2BeSnS4 is however expected to be significantly lower (as much as ∼104 times at typical device operating temperature) compared to the other compounds, which is partly explained by larger relaxation effects in Cu2BeSnS4 as the two anti-site atoms have different sizes. The disadvantage is that the stronger relaxation has a stronger impact on the band-gap narrowing. Therefore, instead of trying to reduce the anti-site pairs, we suggest that one shall try to compensate (CuX + XCu) with (VCu + XCu) or other defects in order to stabilize the gap energy. © 2017 AIP Publishing