Category:Ionic minimization: Difference between revisions

By virtue of the Born-Oppenheimer approximation, the electronic and ionic degrees of freedom are treated separately in VASP. Using the Hellmann-Feynman theorem, VASP can approximate the forces on each ion due to the electronic ground state. The most straightforward approach to ionic minimization is to move the ionic positions such that the force at each site vanishes. This is also known as structure optimization.

Alternatively to the Hellmann-Feynman theorem, VASP can machine learn force fields and, thus, obtain forces approximately 1000 times faster compared to performing an electronic minimization. However, VASP first needs to train on the density-functional-theory (DFT) solutions of similar structures to obtain a machine-learned force field, so it is still necessary to perform DFT calculations. Therefore, this approach is beneficial for large supercells, where it is possible to train on a smaller system.

Finally, ionic minimization does not generally follow the physical path of an ion. For instance, the quasi-Newton RMM-DIIS algorithm (IBRION=1) and conjugate-gradient algorithm (IBRION=2) aim to minimize the total energy without regarding any equation of motion. In contrast, there are various algorithms based on molecular dynamics that can be used to tackle the ionic-minimization problem.

Theory

• Forces: Forces.

How to

• Structure optimization
• Effect of Pulay stress on volume optimizations: Energy vs volume Volume relaxations and Pulay stress.