New strontium titanate polymorphs under high pressure

Area of Science:

• Materials Science
• Computational Chemistry
• Solid State Physics

Background:

• Strontium titanate (SrTiO3) is a technologically important perovskite oxide.
• Understanding its structural phases under pressure is crucial for its applications.
• Previous studies have explored limited pressure ranges or structures.

Purpose of the Study:

• To discover and characterize new dynamically stable structures of SrTiO3.
• To investigate the structural behavior of SrTiO3 across a wide pressure range (0-200 GPa).
• To provide simulated X-ray diffraction patterns for experimental validation.

Main Methods:

• Structure prediction using the Minima Hopping method.
• Machine learning potential (Charge Equilibration via Neural Network Technique - CENT) for energy calculations.
• Validation using Density Functional Theory (DFT).
• Phonon and molecular dynamics (NVT) calculations for stability analysis.

Main Results:

• Identification of six novel, dynamically stable SrTiO3 structures.
• Structures stable at pressures from 0 to 200 GPa.
• Successful validation of machine learning potential against DFT.
• Simulated X-ray diffraction patterns for predicted structures.

Conclusions:

• The study expands the known structural landscape of SrTiO3 under pressure.
• The developed machine learning potential (CENT) offers an efficient route for materials discovery.
• The predicted structures and diffraction patterns can guide future experimental investigations.