Abstract
Depositing CuInSe2 absorbers by sputtering from a single compound target has been proposed over 30 years ago, as old as the current mainstream processes, co-evaporation and post-selenization. Single target sputtering is a simple and straightforward process. Uniform CuInSe2 or Cu(In,Ga)Se2 (CIGS) thin films can be deposited over a large area without the need of toxic gases or post-selenization. However, while both co-evaporation and post-selenization processes have evolved and reached efficiency over 20%, single target sputtering seems to be abandoned, with little progress made in the past thirty years. This dissertation aims to improve the efficiency of the CIGS absorbers deposited by sputtering from a CIGS target in order to fully utilize the advantages of the process. In the dissertation, we point out that the key drawback of the quaternary-sputtering process is the limited Se supply, and several approaches are proposed to resolve the Se deficiency. The first approach is to supply extra Se by sputtering an Sb2Se3 target, which is a stable and up-scalable method to provide extra Se during sputtering. The influences of extra Se supply before, during and after CIGS sputtering are investigated. Secondly, we show that the Se deficiency can also be compensated by extra Na doping, which significantly improves cell efficiency without the need of extra Se supply. Finally, we show that Na and K co-doping further improves the quality of quaternary-sputtered CIGS absorbers, and the possible mechanisms are proposed. With these approaches to minimizing the impact of Se deficiency and improving cell performance, an efficiency of over 14% can be obtained by sputtering from a single CIGS target without post-selenization or introducing compositional grading in the absorbers. The feasibility of the quaternary-sputtering process is proved, and, at the current stage, we do not see any limitation of the quaternary-sputtering process and expect for higher cell efficiency in the near future.