Machine-Learning-Assisted Investigation on Benign Ion Migration in Metal Halide Perovskites

Area of Science:

• Materials Science
• Solid-State Physics
• Photovoltaics

Background:

• Defect-assisted ion migration is a key factor limiting the stability and efficiency of hybrid organic-inorganic metal halide perovskite solar cells.
• Instability and non-radiative losses in these devices are often linked to ion movement within the material.

Purpose of the Study:

• To investigate the interstitial-assisted iodine migration process in metal halide perovskites using advanced simulation techniques.
• To understand the dynamic behavior of trap states and their correlation with ion migration.
• To provide insights for improving the efficiency of perovskite solar cells.

Main Methods:

• Employed long-time-scale molecular dynamics (MD) simulations.
• Utilized a deep potential (DP) model to accurately capture atomic interactions.
• Analyzed the interstitial iodine migration mechanism and its impact on structural and electronic properties.

Main Results:

• Interstitial iodine migration was observed to initially reduce structural distortion and weaken electron-vibration interactions.
• The deep trap state associated with iodine trimers exhibited a dynamic "deep-shallow-deep" transition.
• This dynamic process ultimately led to an improved carrier lifetime during migration.

Conclusions:

• Ion migration, often considered detrimental, can have a beneficial effect under specific conditions in halide perovskites.
• The study highlights strong correlations between different dynamic processes within these materials.
• Findings offer fundamental insights for enhancing the performance of methylammonium lead iodide (CH3NH3PbI3) perovskite solar cells.