Electricity produced by cadmium telluride (CdTe) photovoltaic modules is the lowest-cost electricity in the solar industry, and now undercuts fossil fuel-based sources in many regions of the world. This is due to recent efficiency gains brought about by alloying selenium into the CdTe absorber, which has taken cell efficiency from 19.5% to its current record of 22.1%. Although the addition of selenium is known to reduce the bandgap of the absorber material, and hence increase the cell short-circuit current, this effect alone does not explain the performance improvement. Here, by means of cathodoluminescence and secondary ion mass spectrometry, we show that selenium enables higher luminescence efficiency and longer diffusion lengths in the alloyed material, indicating that selenium passivates critical defects in the bulk of the absorber layer. This passivation effect explains the record-breaking performance of selenium-alloyed CdTe devices, and provides a route for further efficiency improvement that can result in even lower costs for solar-generated electricity.

An 800 nm CdSeTe layer was added to the CdTe absorber used in high-efficiency CdTe cells to increase the current and produce an increase in efficiency. The CdSeTe layer employed had a band-gap near 1.41 eV, compared with 1.5 eV for CdTe. This lower band-gap enabled a current density increase from approximately 26 to over 28 mA/cm2. The open-circuit voltage obtained in the high-efficiency CdTe-only device was maintained and the fill-factor remained close to 80%. Improving the short-circuit current density and maintaining the open-circuit voltage lead to device efficiency over 19%. External quantum efficiency implied that about half the current was generated in the CdSeTe layer and half in the CdTe. Cross-sectional STEM and EDS showed good grain structure throughout. Diffusion of Se into the CdTe layer was observed. This is the highest efficiency polycrystalline CdTe photovoltaic device demonstrated by a university or national laboratory.

Thin film solar cells based on cadmium telluride (CdTe) are complex devices which have great potential for achieving high conversion efficiencies. Lack of understanding in materials issues and device physics slows down the rapid progress of these devices. This paper combines relevant results from the literature with new results from a research programme based on electro-plated CdS and CdTe. A wide range of analytical techniques was used to investigate the materials and device structures. It has been experimentally found that n-, i- and p-type CdTe can be grown easily by electroplating. These material layers consist of nano- and micro-rod type or columnar type grains, growing normal to the substrate. Stoichiometric materials exhibit the highest crystallinity and resistivity, and layers grown closer to these conditions show n → p or p → n conversion upon heat treatment. The general trend of CdCl₂ treatment is to gradually change the CdTe material’s n-type electrical property towards i-type or p-type conduction. This work also identifies a rapid structural transition of CdTe layer at 385 ± 5 °C and a slow structural transition at higher temperatures when annealed or grown at high temperature. The second transition occurs after 430 °C and requires more work to understand this gradual transition. This work also identifies the existence of two different solar cell configurations for CdS/CdTe which creates a complex situation. Finally, the paper presents the way forward with next generation CdTe-based solar cells utilising low-cost materials in their columnar nature in graded bandgap structures. These devices could absorb UV, visible and IR radiation from the solar spectrum and combine impact ionisation and impurity photovoltaic (PV) effect as well as making use of IR photons from the surroundings when fully optimised.