Generating Reference Data and Controlling Accuracy in DFT and Hybrid DFT Simulations

Explore a comprehensive lecture on generating reference data and controlling accuracy in Density Functional Theory (DFT) and hybrid DFT simulations. Delve into all-electron approaches for validating pseudopotentials and evaluating basis set convergence. Examine methods for accelerating hybrid DFT simulations, including localization properties of orbitals and error control in truncating localized orbitals. Learn about first-principles molecular dynamics, solving Kohn-Sham equations, and the use of maximally localized Wannier functions. Discover techniques for reducing computational costs in hybrid DFT, such as recursive subspace bisection and truncation of Wannier functions. Gain insights into PBEO MD simulations of water and the broader context of improving DFT accuracy through Jacob's Ladder and the Escher view.

Acknowledgements
Context: DFT-Molecular Dynamics
Improving DFT Accuracy: Jacob's Ladder
Improving DFT Accuracy (the Escher view)
First-Principles Molecular Dynamics
Solving the Kohn-Sham equations
All-electron atom
Pseudopotential: definition
Pseudopotentials: a recipe
Norm-conserving potentials
All-electron plane wave calculations (AEPW) • Qbox code 1.74.1 (GPL)
Diamond
Silicon (Z=14)
Reducing the cost of hybrid DFT: Using localized orbitals
Maximally localized Wannier functions
Truncation of Wannier functions
Recursive Subspace Bisection
Localization of orbitals in inhomogeneous systems
Truncation error due to bisection
PBEO MD simulations of water
Summary