Optimization of Close Space Sublimation and Post Deposition Routes for Antimony Chalcogenide Solar Cells

Abstract

This thesis explores the influence of close space sublimation (CSS) growth conditions on antimony selenide solar cells as well as the possible benefits of post-growth processing approaches including an assessment of protective layer annealing. Single and two-step growth approaches involving the use of seed layers to modify film coverage and grain structure were investigated as a way of improving the solar cell performance. Both isolated layers and complete device structures were fabricated to allow investigation of the interrelation of preferred ribbon orientation with device efficiency. It was identified that, whilst the use of a seed layer was an important step to achieve good film coverage and grain morphology, the ribbon orientation appeared to have minimal influence on performance. The developed CSS growth approaches were then expanded to produce antimony sulfoselenide films and devices for a single phase source material. It was demonstrated that the approach was feasible, allowing the formation of material with a notably higher bandgap than for the base selenide. This indicated that material did not completely degrade during sublimation with the resulting devices achieve >4% efficiency and with a notably higher open circuit voltage than selenide counterparts. There were however significant issues with the formation of large oxide phase regions within the absorber. These served to reduce the device performance with the cause being attributed to sulphur loss and reaction with oxygen, the growth ambient during deposition. Post growth annealing approaches to improve antimony selenide solar cell efficiency were systematically investigated. Air, selenium, and nitrogen environments were initially compared across a broad temperature range. The results highlighted the degree of sensitivity of the material to post growth annealing with bot air annealing and selenization causing minimal changes to film morphology but drastic performance loss. Nitrogen annealing appeared more favourable with some minor open circuit voltage increases, both again the overall trend was a decrease in cell performance. To overcome these limitations the nitrogen ambient annealing approach was expanded to a protective layer annealing approach. A series of capping layers CdS, ZnO and P2O5 were deposited on the back surface prior to the annealing process to protect the antimony selenide layers and then etched off prior to device completion. The CdS capping layer was found to protect the surface from oxidation but frustratingly still resulted in performance decreases. There was however one “outlier” device series which showed a marked improvement for all device parameters. This result was not reproducible despite many attempts but seemed to indicate the potential of the approach so other materials were investigated. ZnO was considered but it was quickly determined it was unsuited as a capping layer. P2O5 however was tested and despite the limited number of samples being able to be prepared, it was found to notably improve device performance even with short time and low temperature anneals. Secondary ion mass spectrometry analysis showed significant quantities of phosphorus had been incorporated in the film during annealing. This finding demonstrates there is high potential from the protective layer annealing approach and indicates additional work in this area could leave to improved device efficiencies.