Efficient preparation of TiO₂ nanoparticle models using interatomic potentials.

Area of Science:

• Materials Science
• Computational Chemistry
• Nanotechnology

Background:

• Computational modeling is crucial for understanding nanoparticle properties.
• Density functional theory (DFT) calculations for large nanoparticles are computationally intensive.
• Efficient methods are needed to reduce the cost of nanoparticle simulations.

Purpose of the Study:

• To investigate the efficiency of interatomic potentials (IPs) for preparing nanoparticle structures.
• To quantify the computational cost reduction for subsequent DFT calculations.
• To explore the structural and energetic stabilization of nanoparticles using IP-based methods.

Main Methods:

• Comparison of direct DFT optimization with IP preoptimization for faceted TiO2 nanoparticles.
• Evaluation of computational time savings for DFT energy and structure optimizations.
• IP-based molecular dynamics annealing for spherical TiO2 nanoparticles.

Main Results:

• IP preoptimization significantly reduces DFT computational costs (3x-10x speedup for energy evaluations, 2x for structure optimizations).
• IP-based molecular dynamics annealing leads to significant structural reconstruction and energetic stabilization.
• IP methods are 3-4 orders of magnitude faster than DFT tight-binding methods.

Conclusions:

• IPs provide a highly efficient route for preparing and designing large sets of nanoparticles.
• IP preoptimization accelerates DFT convergence and improves accuracy.
• IP-based methods are a valuable tool for computational materials science and nanotechnology.