First-principles Hubbard parameters with automated and reproducible workflows.

Area of Science:

• Computational Materials Science
• Condensed Matter Physics
• Quantum Chemistry

Background:

• Accurate prediction of electronic properties in strongly correlated materials requires precise Hubbard U and V parameters.
• Current methods for calculating these parameters can be computationally expensive and lack flexibility.
• Reproducibility in materials science calculations is crucial for scientific advancement.

Purpose of the Study:

• To introduce an automated and flexible computational framework, aiida-hubbard, for self-consistent calculation of Hubbard U and V parameters.
• To improve the efficiency and reproducibility of Hubbard-corrected first-principles calculations.
• To enable high-throughput screening of d and f electron compounds for various applications.

Main Methods:

• Leveraging density-functional perturbation theory for efficient parallelized computation of Hubbard parameters.
• On-the-fly definition of intersite V parameters to account for atomic relaxations and coordination environments.
• Development of a novel, code-agnostic data structure for storing Hubbard-related information and atomistic structures.

Main Results:

• Demonstrated scalability and reliability by computing U and V parameters for 115 Li-containing bulk solids.
• Revealed significant correlations between onsite U values and the oxidation state/coordination environment of atoms.
• Observed a general decay of intersite V values with increasing interatomic distance, with specific ranges for transition metal-oxygen interactions.

Conclusions:

• The aiida-hubbard framework provides an efficient and reproducible method for calculating Hubbard parameters.
• The findings offer insights into the factors influencing U and V parameters, crucial for materials design.
• This work facilitates the exploration of redox materials and accelerates the discovery of novel materials for energy storage and other technologies.