LFOCKSTD: Difference between revisions
(Created page with "{{DISPLAYTITLE:LFOCKSTD}} {{TAGDEF|LFOCKSTD|[logical]|.FALSE.}} Description: {{TAG|LFOCKSTD}} uses exact one-centre density terms for the exact exchange part of the total energy and RPA forces. ---- {{NB|mind|Avaliable as of 6.5.2.}} This feature is availabe for low-scaling ACFDT/random-phase-approximation (RPA) and GW calculations, i.e. {{TAG|ALGO}}=ACFDTR, RPAR, EVGW0R, GWR. If selected, exact one-center terms in the electronic densit...") |
(Somewhat more pedagogical description.) |
||
| Line 2: | Line 2: | ||
{{TAGDEF|LFOCKSTD|[logical]|.FALSE.}} | {{TAGDEF|LFOCKSTD|[logical]|.FALSE.}} | ||
Description: {{TAG|LFOCKSTD}} | Description: {{TAG|LFOCKSTD}} applies to RPA and GW calculations. It forces VASP to evaluate the exact exchange fully consistent with the standard treatment in HF calculations. | ||
---- | ---- | ||
{{NB|mind|Avaliable as of 6.5.2.}} | {{NB|mind|Avaliable as of 6.5.2.}} | ||
This feature is availabe for low-scaling [[ACFDT/RPA_calculations| ACFDT/random-phase-approximation (RPA)]] and [[GW calculations]], i.e. {{TAG|ALGO}}=ACFDTR, RPAR, EVGW0R, GWR. | This feature is availabe for low-scaling [[ACFDT/RPA_calculations| ACFDT/random-phase-approximation (RPA)]] and [[GW calculations]], i.e. {{TAG|ALGO}}=ACFDTR, RPAR, EVGW0R, GWR. | ||
Usually VASP uses shape restoration (see NMAXFOCKAE and LMAXFOCKAE) for the RPA correlation contributions as well as for the evaluation of the exact exchange during RPA/GW calculations. This however results in signficant noise which is particularly pronounced for the exact exchange. To mitigate this issue, LFOCKSTD was introduced: it forces VASP to use the standard HF-treatment for the exact exchange, but continues to use shape restoration for the correlation contributions. This not only reduces the noise in the force, but also leads to an exact exchange energy that is fully compatible wiht the exact exchange contribution in standard HF calculations (that is the energy " HF-free energy FHF" in the RPA calculations is idential to the "free energy TOTEN " for when reading the WAVECAR file and performing a single step total energy calculation ( ALGO = Eigenval; LFHCALC = .TRUE. ; AEXX = 1.0 ; NELM = 1). This means that this restores the agreement of <code>FHF</code> with the result obtained in the [[ACFDT/RPA_calculations#EXX-inline-ref|step wise computation of the total energy]]. | |||
This restores the agreement of <code>FHF</code> with the result obtained in the [[ACFDT/RPA_calculations#EXX-inline-ref|step wise computation of the total energy]]. | |||
It is strongly recommended to activate LFOCKSTD for all GW and RPA calculations starting from version 6.5.2. | It is strongly recommended to activate LFOCKSTD for all GW and RPA calculations starting from version 6.5.2. | ||
Revision as of 12:03, 4 July 2025
LFOCKSTD = [logical]
Default: LFOCKSTD = .FALSE.
Description: LFOCKSTD applies to RPA and GW calculations. It forces VASP to evaluate the exact exchange fully consistent with the standard treatment in HF calculations.
| Mind: Avaliable as of 6.5.2. |
This feature is availabe for low-scaling ACFDT/random-phase-approximation (RPA) and GW calculations, i.e. ALGO=ACFDTR, RPAR, EVGW0R, GWR.
Usually VASP uses shape restoration (see NMAXFOCKAE and LMAXFOCKAE) for the RPA correlation contributions as well as for the evaluation of the exact exchange during RPA/GW calculations. This however results in signficant noise which is particularly pronounced for the exact exchange. To mitigate this issue, LFOCKSTD was introduced: it forces VASP to use the standard HF-treatment for the exact exchange, but continues to use shape restoration for the correlation contributions. This not only reduces the noise in the force, but also leads to an exact exchange energy that is fully compatible wiht the exact exchange contribution in standard HF calculations (that is the energy " HF-free energy FHF" in the RPA calculations is idential to the "free energy TOTEN " for when reading the WAVECAR file and performing a single step total energy calculation ( ALGO = Eigenval; LFHCALC = .TRUE. ; AEXX = 1.0 ; NELM = 1). This means that this restores the agreement of FHF with the result obtained in the step wise computation of the total energy.
It is strongly recommended to activate LFOCKSTD for all GW and RPA calculations starting from version 6.5.2.
In earlier versions, one-center density contributions of EXX can be treated only with shape and electrostatic multipole reconstruction using LMAXFOCKAE.