Optimizing special quasirandom structure (SQS) models for accurate functional property prediction in disordered 2D...

Area of Science:

• Materials Science
• Computational Materials Science
• Condensed Matter Physics

Background:

• Two-dimensional (2D) materials, including MXenes, offer tunable properties for diverse applications.
• Alloying 2D materials creates solid-solutions (disordered alloys) with unique characteristics.
• Predicting disordered alloy properties requires computationally tractable models that mimic random atomic arrangements.

Purpose of the Study:

• To assess the accuracy of special quasirandom structures (SQSs) for predicting properties of disordered 2D MXene alloys.
• To determine the optimal supercell size of SQSs for reliable property predictions.
• To investigate the influence of SQS size on formation energies, elastic properties, and structural parameters.

Main Methods:

• Density Functional Theory (DFT) calculations.
• Cluster Expansion (CE) method.
• Systematic variation of SQS supercell size (N=6-8) for binary MXene alloys (e.g., Ti-M'CO2).

Main Results:

• SQSs with N=6-8 provide converged predictions for formation energies, elastic properties, and structural parameters of disordered MXene alloys.
• Formation energies are reproduced within ~2.5 meV for SQSs with N>4.
• Structural and elastic properties for TiNbCO2 converge at N~6, correlating with electronic structure convergence.

Conclusions:

• Modest sized SQSs are sufficient to accurately model disordered MXene alloys.
• The findings provide guidance for selecting appropriate SQS sizes in computational studies of 2D alloy properties.
• This work facilitates structure-property relationship investigations in other disordered 2D materials.