Ag-Bi Charge Redistribution Creates Deep Traps in Defective Cs2AgBiBr6: Machine Learning Analysis of Density Functional Theory

Lead-free double perovskites hold promise for stable and environmentallybenign solar cells; however, they exhibit low efficiencies because defects act as chargerecombination centers. Identifying trap-assisted loss mechanisms and developing defectpassivation strategies constitute an urgent goal. Applying unsupervised machine learning todensity functional theory and nonadiabatic molecular dynamics, we demonstrate thatnegatively charged Br vacancies in Cs2AgBiBr6create deep hole traps through chargeredistribution between the adjacent Ag and Bi atoms. Vacancy electrons arefirst accepted byBi and then shared with Ag, as the trap transforms from shallow to deep. Subsequent chargelosses are promoted by Ag and Bi motions perpendicular to rather than along the Ag-Bi axis,as can be expected. In contrast, charge recombination in pristine Cs2AgBiBr6correlates mostwith displacements of Cs atoms and Br-Br-Br angles. Doping with In to replace Ag at thevacancy maintains the electrons at Bi and keeps the trap shallow.

Automated summary

Lead-free double perovskites have low efficiencies due to defects that act as charge recombination centers. By applying unsupervised machine learning, it was found that Br vacancies create deep hole traps, and doping with In can maintain shallow traps.