Origin of Reduced Efficiency in Cu(In,Ga)Se $_2$ Solar Cells With High Ga Concentration: Alloy Solubility Versus Intrinsic Defects

It is well known that adding Ga to CuInSe₂ forming CuIn_1-xGa_xSe₂ (CIGS) alloy can significantly improve the solar cell efficiency, but adding too much Ga will lead to a decline of the solar cell efficiency. The exact origin of this puzzling phenomenon is currently still under debate. It is especially unclear whether it is caused by either structural or electronic issues. In this paper, we conclude that the defect issue, especially antisite defects M_Cu (M = In, Ga), rather than the alloy solubility is the key problem for the reduced efficiency in CIGS. The deep levels that are induced by M_Cu defects can pin the open-circuit voltage (V_oc) of CIGS. Self-compensation in CIGS, which forms 2V_Cu + M Cu defect complexes, is found to be beneficial to quenching the deep-trap levels induced by M_Cu in CIGS. Unfortunately, the density of isolated M_Cu is quite high and cannot be largely converted into 2V_Cu + M_Cu complexes under thermal equilibrium condition. Thus, nonequilibrium growth conditions or low growth temperature that can suppress the formation of the deep-trap centers M_Cu will be necessary to improve the efficiency of CIGS solar cells, especially with high Ga concentrations.