Tight-Binding Approximation-Enhanced Global Optimization

Area of Science:

• Materials Science
• Computational Chemistry
• Solid State Physics

Background:

• Predicting atomic structures computationally is crucial for materials discovery but often limited by high computational costs.
• First-principles methods, while accurate, require significant computational resources for exploring vast structural landscapes.
• Existing methods struggle with the efficiency needed for complex materials and large search spaces.

Purpose of the Study:

• To develop a computationally efficient method for atomic structure prediction.
• To overcome the limitations of traditional first-principles calculations in exploring structural search spaces.
• To enable faster and more reliable identification of stable crystal structures.

Main Methods:

• Utilized density functional tight-binding (DFTB) for efficient electronic structure calculations.
• Implemented an automatic adaptive parametrization scheme for repulsive pair potentials.
• Employed a genetic algorithm for structure optimization, guided by the efficient DFTB method.

Main Results:

• Successfully applied the method to optimize the crystal structures of bulk carbon, titanium dioxide, palladium oxide, and calcium hydroxide.
• Demonstrated the efficiency of the DFTB approach in accelerating the search for stable atomic structures.
• Assessed the stability of the previously unknown crystal structure of palladium hydroxide.

Conclusions:

• The developed DFTB-based approach significantly enhances the efficiency of atomic structure prediction.
• This method provides a viable pathway for accelerating materials discovery and understanding material properties.
• The technique is effective for optimizing known structures and exploring unknown crystal phases.