|
|
| Line 3: |
Line 3: |
| Description: {{TAG|OMEGAMAX}} minimum frequency in the frequency grid. | | Description: {{TAG|OMEGAMAX}} minimum frequency in the frequency grid. |
| ---- | | ---- |
| For the frequency grid along the real and imaginary axis sophisticated schemes are used, which are based on simple model functions for the macroscopic dielectric function. The grid spacing is dense up to roughly 1.3*{{TAG|OMEGAMAX}} and becomes coarser for larger frequencies. The default value for {{TAG|OMEGAMAX}} is either determined by the outermost node in the dielectric function (corresponding to a singularity in the inverse of the dielectric function) or the energy difference between the valence band minimum and the conduction band minimum. The larger of these two values is used. Except for pseudopotentials with deep lying core states, {{TAG|OMEGAMAX}} is usually determined by the node in the dielectric function.
| |
|
| |
| For {{TAG|ACFDT calculations}}, only {{TAG|OMEGAMIN}} and {{TAG|OMEGATL}} determine the frequency grid (using a minimax algorithm).
| |
|
| |
| The defaults have been carefully tested, and it is recommended to leave them unmodified, whenever possible. The grid should be solely controlled by {{TAG|NOMEGA}}. The only other value that can be modified is the complex shift {{TAG|CSHIFT}}. In principle, {{TAG|CSHIFT}} should NOT be chosen independently of {{TAG|NOMEGA}} and {{TAG|OMEGAMAX}}: e.g. for less dense grids (smaller {{TAG|NOMEGA}}) the complex shift must be accordingly increased. The default for {{TAG|CSHIFT}} has been chosen such that the calculations are converged to 10 meV with respect to {{TAG|NOMEGA}}: i.e. if {{TAG|CSHIFT}} is kept constant and {{TAG|NOMEGA}} is increased, the QP shifts should not change by more than 10 meV; at least for {{TAG|LSPECTRAL}} = .TRUE.. This was the case for the considered test materials. For {{TAG|LSPECTRAL}} = .FALSE. this does not apply. In this case it is recommended to set {{TAG|CSHIFT}} manually and to perform careful convergence tests.
| |
|
| |
| For {{TAG|LSPECTRAL}} = .TRUE. independent convergence tests with respect to {{TAG|NOMEGA}} and {{TAG|CSHIFT}} are usually not required, and it should be sufficient to control the technical parameters via the single parameter {{TAG|NOMEGA}}. Also note that too large values for {{TAG|NOMEGA}} in combination with coarse k-point grids can cause a decrease in precision (see {{TAG|NOMEGA}}).
| |
|
| |
|
| == Related Tags and Sections == | | == Related Tags and Sections == |
| | {{TAG|OMEGAMAX}} |
| {{TAG|OMEGATL}}, | | {{TAG|OMEGATL}}, |
| {{TAG|CSHIFT}}, | | {{TAG|CSHIFT}}, |
Revision as of 09:59, 9 February 2017
