Data-driven and constrained optimization of semi-local exchange and nonlocal correlation functionals for materials...

Area of Science:

• Computational chemistry
• Materials science
• Surface science

Background:

• Accurate prediction of surface chemical reaction energetics is crucial for understanding chemisorption and physisorption.
• Density functional approximations (DFAs) require careful optimization for reliable energetic predictions.

Purpose of the Study:

• To develop an empirical approach to simultaneously optimize semi-local exchange and nonlocal correlation in DFAs.
• To improve the accuracy of predicting surface chemical reaction energetics.

Main Methods:

• Developed the VCML-rVV10 exchange-correlation functional through empirical optimization.
• Utilized a diverse dataset including solid bulk, surface, and gas-phase chemistry reference data.
• Incorporated physical exchange-correlation model constraints during optimization.

Main Results:

• The VCML-rVV10 functional demonstrates improved accuracy for surface chemical reaction energetics.
• Validated applicability beyond surface chemistry, providing optimized gas-phase reaction energetics.
• Achieved accurate descriptions of bulk lattice constants and elastic properties.

Conclusions:

• The VCML-rVV10 functional offers a versatile tool for computational chemistry and materials science.
• Simultaneous optimization of exchange and correlation components enhances predictive power.
• The functional's broad applicability makes it valuable for various chemical and physical simulations.