Publication date: 11th March 2026
CsPbBr₃ is a promising perovskite solar-cell absorber due to its high stability under illumination and resistance to moisture, combined with favorable charge-transport properties. To maximize device efficiency, an electron transport layer (ETL) with suitable band alignment and high conductivity is required to enable efficient electron extraction and suppress charge recombination. SnO₂ is a widely used ETL candidate because it combines high electron mobility with excellent stability under ambient conditions. However, recombination losses associated with interfacial structure and surface chemistry at the CsPbBr₃/SnO₂ junction remain a key challenge. This study investigates the CsPbBr₃/SnO₂ interface at the atomistic and electronic scales using first-principles density functional theory (DFT) calculations and molecular dynamics (MD) simulations.
First-principles DFT calculations on CsPbBr₃/SnO₂ interface models are used as reference data for training a machine-learning interatomic potential using the Neuroevolution Potential (NEP) approach. The trained NEP enables large-scale, low-cost MD simulations of supercells, allowing temperature-dependent structural behavior to be studied across varying distances from the interface. In particular, we analyse octahedral tilting in CsPbBr₃ as a function of temperature and proximity to the interface and assess its impact on the electronic structure.
