posted on 2021-11-30, 15:02authored byLewis Wright
Research grade photovoltaic devices made from inorganic compounds such as CdTe and Cu(In,Ga)Se2 are steadily progressing towards the thermodynamic limit, and as such have moved into commercial production. However, these compounds contain elements that are either toxic, scarce, or used in other high-demand products, making them potentially problematic for long-term mass production of photovoltaic modules. This thesis investigates three different compounds made from earth abundant elements as potential alternatives to the established thin film technologies.
The kesterite Cu2ZnSnSe4 is widely lauded for its abundance, low toxicity, and ideal/tuneable optoelectronic properties. Here, to build on its compositional advantages, an industrially scalable spray pyrolysis deposition technique is used to produce Cu2ZnSnSe4 absorber layers for photovoltaic devices. In the first work chapter this leads to a brand-new water-based solution, utilising the amine-thiol mixture solvent system, which produced an in-house record 6.8% device efficiency. This not only shows the promise of spray deposition for cheap-to-produce devices, but that objectively non-toxic solvents can be used to dissolve precursors, another boon to mass production.
Despite being an in-house record efficiency this Cu2ZnSnSe4 device suffered from the open-circuit voltage deficit (Voc-deficit) characteristic of kesterites. The source of this deficit is unknown and prevents not only efficiency improvements but also interest in mass-production. In the second work chapter a hypothesis is presented for the root cause of the Voc-deficit, specifically that Sn’s multivalence is responsible for defects deep in the bandgap that limit the Voc and ultimately the efficiency. To test this hypothesis is the first recorded attempt at forming the kesterite Cu2ZnZrSe4, replacing multivalent Sn with monovalent Zr. Combinatorial sputtering is used to quickly traverse this new, large, quaternary compositional phase space.
While attempting to produce a new quaternary material, the pseudo-binary phase Cu-doped ZnSe was found to reliably form, with a favourable 1.3eV bandgap and photoluminescence an order of magnitude larger than the inhouse record kesterite device. In the third work chapter is a preliminary investigation of this Cu:ZnSe phase, again with combinatorial sputtering to quickly traverse the phase space.
Funding
EPSRC Centre for Doctoral Training in New and Sustainable PV
Engineering and Physical Sciences Research Council