Dynamical Disorder in the Mesophase Ferroelectric HdabcoClO₄: A Machine-Learned Force Field Study.

Area of Science:

• Materials Science
• Condensed Matter Physics
• Computational Chemistry

Background:

• Hybrid molecular ferroelectrics offer lead-free alternatives with tunable properties.
• HdabcoClO4 exhibits ultrafast room-temperature ferroelectric switching.
• Understanding dynamical disorder in these materials is limited.

Purpose of the Study:

• To investigate the nature of dynamical disorder in HdabcoClO4.
• To develop and utilize a machine-learned force field (MLFF) for molecular dynamics (MD) simulations.
• To correlate simulated behavior with experimental observations.

Main Methods:

• Density functional theory (DFT) calculations to generate training data.
• Training a neural network (NeuralIL) to create an MLFF.
• Performing MLFF-MD simulations to study phase transitions and thermal expansion.

Main Results:

• MLFF-MD simulations accurately reproduced experimental phase transitions and thermal expansion.
• Identified low-temperature phase transition linked to ClO4- orientational disorder.
• Observed high-temperature phase transition involving Hdabco+ and ClO4- rotation.
• Detected proton transfer even in the low-temperature phase, increasing with temperature.

Conclusions:

• The study elucidates the complex dynamical disorder in HdabcoClO4, including proton transfer and chain disorder.
• MLFF-MD simulations provide a powerful tool for studying hybrid molecular ferroelectrics.
• Findings advance the understanding of lead-free ferroelectric materials for future applications.