Thermodynamic Model of Mixing and Photo-Demixing in MAPb(I,Br)3 from Machine-Learning Molecular Dynamics

Mixed-halide lead perovskites are promising materials for solar cells because of high photoconversion efficiencies and bandgaps that can be finely tuned through composition, but their application is limited by reversible light-induced halide segregation. Here, we investigate the thermodynamic origin of the segregation behaviour in MAPb(I,Br)3 by combining molecular dynamics, based on a machine-learned interatomic potential fine-tuned to r2SCAN density-functional-theory data, with a model for the electronic and elastic contributions that arise under illumination. The simulations give the free energy of mixing in the dark across the full composition range, including configurational and molecular rotational effects. Our results indicate that the mixed material is thermodynamically stable in the dark at 350 K and, by extrapolation, at room temperature, in contrast with earlier theoretical predictions of a room-temperature miscibility gap and consistent with the spontaneous remixing observed once illumination is removed. Under illumination, the accumulation of photocarriers in lower-bandgap iodide-rich regions favours compositional separation. Within this band-gap model, the carrier density at which incipient iodide-rich domains become thermodynamically favourable is within the range of 1-sun illumination for Br-rich alloys, but forming a macroscopic iodide-rich phase that occupies a few percent of the material, with the observed composition, requires carrier densities about four orders of magnitude higher. The more stable dark free energy obtained here widens this gap, and incorporating the coherency strain between the phases into the model widens it further. These results indicate that the band-gap model accounts only for the onset of photo-demixing and not for its macroscopic stage, which must involve additional physics, and demonstrate that reliable models of photo-demixing require an accurate, dynamic description of the dark thermodynamics.

Authors

Institutions

Publication Details

Journal
ChemRxiv
Published
2026-11-17
DOI
https://doi.org/10.26434/chemrxiv.15012331/v1
Primary Topic
Perovskite Materials and Applications
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Thermodynamic Model of Mixing and Photo-Demixing in MAPb(I,Br)3 from Machine-Learning Molecular Dynamics

Eduardo Menéndez‐Proupin, Ricardo Grau‐Crespo, Juan A. Anta, Ana L. Montero‐Alejo et al.
ChemRxiv
Perovskite Materials and Applications
preprint

Thermodynamic Model of Mixing and Photo-Demixing in MAPb(I,Br)3 from Machine-Learning Molecular Dynamics

Eduardo Menéndez‐Proupin, Ricardo Grau‐Crespo, Juan A. Anta, Ana L. Montero‐Alejo, Antonio Rivas-Blanco, Said Hamad, Salvador Rodríguez-Gómez Balestra
preprint en

Abstract

Mixed-halide lead perovskites are promising materials for solar cells because of high photoconversion efficiencies and bandgaps that can be finely tuned through composition, but their application is limited by reversible light-induced halide segregation. Here, we investigate the thermodynamic origin of the segregation behaviour in MAPb(I,Br)3 by combining molecular dynamics, based on a machine-learned interatomic potential fine-tuned to r2SCAN density-functional-theory data, with a model for the electronic and elastic contributions that arise under illumination. The simulations give the free energy of mixing in the dark across the full composition range, including configurational and molecular rotational effects. Our results indicate that the mixed material is thermodynamically stable in the dark at 350 K and, by extrapolation, at room temperature, in contrast with earlier theoretical predictions of a room-temperature miscibility gap and consistent with the spontaneous remixing observed once illumination is removed. Under illumination, the accumulation of photocarriers in lower-bandgap iodide-rich regions favours compositional separation. Within this band-gap model, the carrier density at which incipient iodide-rich domains become thermodynamically favourable is within the range of 1-sun illumination for Br-rich alloys, but forming a macroscopic iodide-rich phase that occupies a few percent of the material, with the observed composition, requires carrier densities about four orders of magnitude higher. The more stable dark free energy obtained here widens this gap, and incorporating the coherency strain between the phases into the model widens it further. These results indicate that the band-gap model accounts only for the onset of photo-demixing and not for its macroscopic stage, which must involve additional physics, and demonstrate that reliable models of photo-demixing require an accurate, dynamic description of the dark thermodynamics.

ChemRxiv
Queen Mary University of London (GB), Metropolitan University of Technology (CL), Universidad de Sevilla (ES), Universidad Pablo de Olavide (ES)
European Regional Development Fund, Agencia Estatal de Investigación, European Social Fund
Perovskite Materials and Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon