DSKO: Dancing through DFTB Parametrization.

Area of Science:

• Computational materials science
• Quantum chemistry
• Condensed matter physics

Background:

• Density functional tight binding (DFTB) is a computationally efficient method.
• DFTB bridges the accuracy of density functional theory (DFT) and the speed of semiempirical models.
• The reliability of DFTB depends heavily on parameter quality, especially for the electronic interactions.

Purpose of the Study:

• To develop a novel framework, DFTB Slater-Koster Optimizer (DSKO), for generating accurate and transferable DFTB electronic parameters.
• To address the underdeveloped nature of electronic parametrization in DFTB.
• To improve the fidelity of semiempirical simulations in materials science.

Main Methods:

• DSKO incorporates robust optimization algorithms.
• Physics-informed loss functions are utilized within the DSKO framework.
• The optimization process is guided by rigorous physical constraints.

Main Results:

• DSKO generates DFTB electronic parameters that closely match DFT reference data.
• The optimized parameters yield accurate electronic properties, including density of states and band structures.
• The framework demonstrates versatility for various materials science applications.

Conclusions:

• DSKO significantly enhances the accuracy and transferability of DFTB electronic parameters.
• The developed framework enables high-fidelity semiempirical simulations.
• DSKO facilitates the broader application of DFTB in computational materials science.