I will simulate PbTiO3 structure, how should I use the distance of atomic range
for the 3D model?
for both k-points grid and cut-off energy, try several settings, and choose the one that gives best convergence. You should define convergence criteria. It can be min energy difference (such as adsorption energy), total min. energy of system, or best definition of properties, such as band structure. Also note that, as you increse the both parameters, resutls tend to get better (ususally), thus definition of convergence criteria is important. An optimum point is required between the best results and minimum possible computation time.
Begin with a fairly sparse k-point grid, or even Gamma, and increase the energy cut-off in steps of 20% from a low value until the total energy of the model system stops decreasing on each step by, say 0.1 meV. Then, increase the grid density monotonically until the total energy stops changing by the same amount as chosen previously for the convergence criterion, and that's it.
Would you please show step by step to add atom model method either build crystal dope or substitue in material .
Would you definite meanig of cut-off energy and Monkhrost–Pack scheme grid k-point ?
- Badges
- Science topic