Crystal structure optimisation using an auxiliary equation of state.

Area of Science:

• Computational Materials Science
• Solid-State Physics
• Crystallography

Background:

• Standard crystal structure optimization requires extensive energy and force calculations, often involving fitting equations of state to energy-volume curves.
• This process is computationally demanding, especially for complex, low-symmetry structures and advanced electronic structure methods like non-local functionals where analytical gradients are unavailable.

Purpose of the Study:

• To develop a computationally efficient method for optimizing crystal structures.
• To reduce the number of energy calculations required for determining equilibrium cell volumes.

Main Methods:

• A novel approach based on a known equation of state is presented.
• The method predicts the equilibrium volume using a single-point energy calculation and refines it iteratively if necessary.

Main Results:

• The proposed method significantly reduces computational expense compared to traditional energy-volume curve fitting.
• Validation was performed on binary compounds (PbS, PbTe, ZnS, ZnTe) using nine density functionals.
• The approach was successfully applied to complex materials like Cu2ZnSnS4 and HKUST-1.

Conclusions:

• The presented method offers a substantial improvement in the efficiency of crystal structure optimization.
• This technique is particularly beneficial for computationally intensive electronic structure methods and complex crystal systems.
• The approach facilitates faster exploration of materials properties and accelerates materials discovery.