Researchers from the Tyndall National Institute at University College Cork, Ireland, have made significant strides in accurately predicting the electronic properties of photovoltaic materials using a refined computational approach. The team, led by Dr. Andrew C. Burgess and including Lórien MacEnulty, Ethan D’Arcy, David Gavin, and Professor David D. O’Regan, has developed a method that enhances the precision of bandgap predictions for kesterite materials, which are promising candidates for thin-film solar cells.
The study, published in the journal Physical Review Materials, focuses on Cu2ZnSnS4 and Cu2ZnGeS4 kesterite materials. The researchers employed Density Functional Theory (DFT) plus Hubbard U technique (DFT+U), a computational method that corrects for electronic correlations in materials. The key innovation lies in the evaluation of corrective parameters via minimum-tracking linear response, which ensures reliable bandgap predictions without the need for empirically tuned parameters. This approach also maintains low computational overhead, making it practical for widespread use in materials research.
The researchers found that applying Hubbard U corrections to all atomic subspaces that dominate the conduction and valence band edges, rather than just the conventional 3d and 4f atomic states, significantly improved the accuracy of bandgap predictions. Interestingly, the inclusion of Hund’s J corrections via the extended DFT+U+J functional worsened the results, but this issue was mitigated by using the Burgess-Linscott-O’Regan (BLOR) flat-plane based Hubbard U plus Hund’s J functional. The refined method’s predictions were even found to marginally outperform those from the self-consistent GW approach, a more computationally intensive method.
The practical implications for the energy sector are substantial. Accurate bandgap predictions are crucial for the development of efficient photovoltaic materials. The streamlined DFT+U method developed by the Tyndall National Institute team can expedite the discovery and optimization of new materials for solar cells, potentially lowering costs and improving the performance of solar energy technologies. Additionally, the method was used to predict defect-induced changes to the bandgap and associated formation energies in large supercells, providing valuable insights into the stability and performance of these materials under real-world conditions.
This research represents a significant advancement in the computational tools available for materials science, particularly in the field of photovoltaics. By providing a reliable, efficient, and parameter-free method for predicting electronic properties, the team has opened new avenues for innovation in solar energy technology.
This article is based on research available at arXiv.