NMAXFOCKAE: Difference between revisions
No edit summary |
No edit summary |
||
Line 1: | Line 1: | ||
{{TAGDEF|NMAXFOCKAE|1{{!}}2|1} | {{TAGDEF|NMAXFOCKAE|1{{!}}2|1}} | ||
{{TAGDEF|LMAXFOCKAE|[integer] | {{TAGDEF|LMAXFOCKAE|[integer]}} | ||
{{DEF| | {{DEF|LAXMFOCKAE|-1|for DFT, Hartree-Fock | 4 | for post DFT methods}} | ||
Description: {{TAG|NMAXFOCKAE}} and {{TAG|LMAXFOCKAE}} determine whether | |||
Description: {{TAG| | the overlap densities in the Fock exchange and correlated wave function methods are accurately reconstructed on the plane wave grid. | ||
---- | ---- | ||
In the PAW method the difference between the charge density of the all-electron partial waves and | In the PAW method, the difference between the charge density of the all-electron partial waves and | ||
the pseudo partial waves | the pseudo partial waves | ||
<math> | <math> | ||
Line 16: | Line 15: | ||
is usually restored on spherical grids centered at each atom | is usually restored on spherical grids centered at each atom | ||
(one-center terms inside the PAW spheres). To describe long range electrostatic terms, the | (one-center terms inside the PAW spheres). To describe long range electrostatic terms, the | ||
the ''moments'' of the differences of the all-electron and pseudo charge density are also | the ''moments'' of the differences of the all-electron and pseudo charge density are usually | ||
also added on the plane wave grid up to a certain ''l'' quantum number (see {{TAG|LMAXFOCK}}). | |||
These augmentation charges restore the moments of the all-electron density on the plane wave | |||
grid. | |||
For the RPA, GW, and most post DFT methods, the one-center terms are presently, | For the RPA, GW, and most post DFT methods, the one-center terms are presently, | ||
Line 23: | Line 24: | ||
in particular for 3d and (to a lesser extent) 2p, 4d and 5d elements. | in particular for 3d and (to a lesser extent) 2p, 4d and 5d elements. | ||
To correct for this error, an alternative treatment is implemented | To correct for this error, an alternative treatment is implemented | ||
on the plane wave grid. This allows to restore the | on the plane wave grid. This allows to restore the all-electron charge density accurately on the plane wave grid, using the flags {{TAG|LMAXFOCKAE}} and {{TAG|NMAXFOCKAE}}. | ||
To achieve this, <math> Q_{\alpha\beta}(r) </math> is Fourier transformed | To achieve this, <math> Q_{\alpha\beta}(r) </math> is Fourier transformed | ||
Line 30: | Line 31: | ||
If the {{TAG| | If the {{TAG|LMAXFOCKAE}}=-1 (the default for DFT and Hartree-Fock calculations), only the moments of the all-electron charge density is restored on the plane wave grid. This setting is exact for density functional theory, Hartree-Fock as well | ||
as hybrid functionals, since the one-center terms are implemented. | as hybrid functionals, since the one-center terms are implemented. | ||
If | If {{TAG|NMAXFOCKAE}}=1 and {{TAG|LMAXFOCKAE}} is set, the moments of the all-electron charge density are restored on the plane wave grid. Furthermore, the all-electron charge density is restored up to a typical plane wave energy of 140 eV. This setting yields very accurate results for post DFT methods (MP2, RPA, GW, etc.) for most sp bonded materials. {{TAG|LMAXFOCKAE}} is used to specify the maximum spherical (l) quantum number up | ||
to which this more accurate treatment is used. The default for {{TAG|LMAXFOCKAE}} | to which this more accurate treatment is used. The default is {{TAG|LMAXFOCKAE}}=4, for post DFT methods. | ||
If no accurate augmentation is desired by the user, simply set {{TAG|LMAXFOCKAE}}=-1 in the INCAR file. | |||
If {{TAG|NMAXFOCKAE}}=2 is set, the charge density is restored accurately on the plane wave grid up to a typical plane wave energies of 380 eV. As before, {{TAG|LMAXFOCKAE}} can be used to specify the maximum spherical (l) quantum number up | If {{TAG|NMAXFOCKAE}}=2 and {{TAG|LMAXFOCKAE}} is set, the charge density is restored accurately on the plane wave grid up to a typical plane wave energies of 380 eV. As before, {{TAG|LMAXFOCKAE}} can be used to specify the maximum spherical (l) quantum number up | ||
to which this more accurate treatment is used. {{TAG|NMAXFOCKAE}}=2 yields very accurate results for | to which this more accurate treatment is used. {{TAG|NMAXFOCKAE}}=2 yields very accurate results for | ||
post DFT methods (MP2, RPA, GW) even for difficult 3d elements. For RPA and MP2 total energy calculations, differences between {{TAG|NMAXFOCKAE}}=1 and {{TAG|NMAXFOCKAE}}=2 are usually tiny for total energy differences. Since the absolute correlation energies might change, it is vital to use the same setting for | post DFT methods (MP2, RPA, GW) even for difficult 3d elements. For RPA and MP2 total energy calculations, differences between {{TAG|NMAXFOCKAE}}=1 and {{TAG|NMAXFOCKAE}}=2 are usually tiny for total energy differences. Since the absolute correlation energies might change, it is vital to use the same setting for | ||
{{TAG|NMAXFOCKAE}} and {{TAG|LMAXFOCKAE}}, if energy differences are calculated. | {{TAG|NMAXFOCKAE}} and {{TAG|LMAXFOCKAE}}, if energy differences are calculated. | ||
For GW calculations, increasing {{TAG|NMAXFOCKAE}} | For GW calculations, increasing {{TAG|NMAXFOCKAE}} from 1 to 2 might change QP energies by 100-200 meV for 3d and late 4d and 5d elements. | ||
If {{TAG|NMAXFOCKAE}} is used, the setting for {{TAG|LMAXFOCKAE}} should be also considered carefully. Generally, it suffices to set {{TAG|LMAXFOCKAE}} to twice the maximum ''l'' quantum number found in the {{FILE|POTCAR}} file. | |||
For instance for sp elements, {{TAG|LMAXFOCKAE}} = 2 suffices. For d elements, {{TAG|LMAXFOCKAE}} = 4 suffices | |||
(a d electron can create charge densities with l-quantum number of 4), whereas for f elements, users | |||
should test whether {{TAG|LMAXFOCKAE}} = 6 is required. | |||
In summary, usefully manual setting of {{TAG|NMAXFOCKAE}} and {{TAG|LMAXFOCKAE}} are: | |||
* {{TAG|LMAXFOCKAE}}=-1, to switch off the accurate augmentation altogether (fall back to the DFT treatment) | |||
( | * {{TAG|LMAXFOCKAE}}=4 (or larger) to force an accurate treatment for the HF part even in Hartree-Fock calculations | ||
* {{TAG|NMAXFOCKAE}}=2, to select the very accurate augmentation. Please check whether the VASP default setting for {{TAG|LMAXFOCKAE}} suffices (OUTCAR file). | |||
Revision as of 13:44, 14 March 2017
NMAXFOCKAE = 1|2
Default: NMAXFOCKAE = 1
LMAXFOCKAE = [integer]
Default: LAXMFOCKAE | = -1 | for DFT, Hartree-Fock |
= 4 | for post DFT methods |
Description: NMAXFOCKAE and LMAXFOCKAE determine whether the overlap densities in the Fock exchange and correlated wave function methods are accurately reconstructed on the plane wave grid.
In the PAW method, the difference between the charge density of the all-electron partial waves and the pseudo partial waves is usually restored on spherical grids centered at each atom (one-center terms inside the PAW spheres). To describe long range electrostatic terms, the the moments of the differences of the all-electron and pseudo charge density are usually also added on the plane wave grid up to a certain l quantum number (see LMAXFOCK). These augmentation charges restore the moments of the all-electron density on the plane wave grid.
For the RPA, GW, and most post DFT methods, the one-center terms are presently, however, not implemented. Depending on the material, this can cause sizable errors in particular for 3d and (to a lesser extent) 2p, 4d and 5d elements. To correct for this error, an alternative treatment is implemented on the plane wave grid. This allows to restore the all-electron charge density accurately on the plane wave grid, using the flags LMAXFOCKAE and NMAXFOCKAE.
To achieve this, is Fourier transformed to reciprocal space and then expanded in a set of orthogonal functions localized at each atomic site.
If the LMAXFOCKAE=-1 (the default for DFT and Hartree-Fock calculations), only the moments of the all-electron charge density is restored on the plane wave grid. This setting is exact for density functional theory, Hartree-Fock as well
as hybrid functionals, since the one-center terms are implemented.
If NMAXFOCKAE=1 and LMAXFOCKAE is set, the moments of the all-electron charge density are restored on the plane wave grid. Furthermore, the all-electron charge density is restored up to a typical plane wave energy of 140 eV. This setting yields very accurate results for post DFT methods (MP2, RPA, GW, etc.) for most sp bonded materials. LMAXFOCKAE is used to specify the maximum spherical (l) quantum number up to which this more accurate treatment is used. The default is LMAXFOCKAE=4, for post DFT methods. If no accurate augmentation is desired by the user, simply set LMAXFOCKAE=-1 in the INCAR file.
If NMAXFOCKAE=2 and LMAXFOCKAE is set, the charge density is restored accurately on the plane wave grid up to a typical plane wave energies of 380 eV. As before, LMAXFOCKAE can be used to specify the maximum spherical (l) quantum number up to which this more accurate treatment is used. NMAXFOCKAE=2 yields very accurate results for post DFT methods (MP2, RPA, GW) even for difficult 3d elements. For RPA and MP2 total energy calculations, differences between NMAXFOCKAE=1 and NMAXFOCKAE=2 are usually tiny for total energy differences. Since the absolute correlation energies might change, it is vital to use the same setting for NMAXFOCKAE and LMAXFOCKAE, if energy differences are calculated. For GW calculations, increasing NMAXFOCKAE from 1 to 2 might change QP energies by 100-200 meV for 3d and late 4d and 5d elements.
If NMAXFOCKAE is used, the setting for LMAXFOCKAE should be also considered carefully. Generally, it suffices to set LMAXFOCKAE to twice the maximum l quantum number found in the POTCAR file.
For instance for sp elements, LMAXFOCKAE = 2 suffices. For d elements, LMAXFOCKAE = 4 suffices
(a d electron can create charge densities with l-quantum number of 4), whereas for f elements, users
should test whether LMAXFOCKAE = 6 is required.
In summary, usefully manual setting of NMAXFOCKAE and LMAXFOCKAE are:
- LMAXFOCKAE=-1, to switch off the accurate augmentation altogether (fall back to the DFT treatment)
- LMAXFOCKAE=4 (or larger) to force an accurate treatment for the HF part even in Hartree-Fock calculations
- NMAXFOCKAE=2, to select the very accurate augmentation. Please check whether the VASP default setting for LMAXFOCKAE suffices (OUTCAR file).