ISIF: Difference between revisions
No edit summary |
(Explanation of sign convention for stress tensor in VASP.) |
||
Line 37: | Line 37: | ||
| 8 || yes || yes || yes || no || yes | | 8 || yes || yes || yes || no || yes | ||
|} | |} | ||
In VASP the stress tensor is defined as the negative of the derivative of the energy | |||
<math> E </math> | |||
with respect to the strain tensor: | |||
<math> \epsilon_{ij} </math> | |||
For {{TAG|ISIF}}=1, only the trace of the stress tensor is calculated. This means only the total pressure is | <math> \sigma_{ij} = - \frac{\delta E} {\delta \epsilon_{ij}} </math> | ||
This might be different from other first principles codes. | |||
A positive in the diagonals means that the system is under compressive strain and wants to expand. A negative value implies that the system is under tensile strain and wants to reduce its volume. | |||
For {{TAG|ISIF}}=1, only the trace of the stress tensor is calculated. This means only the total pressure is correct and can be read off in the line: | |||
external pressure = ... kB | external pressure = ... kB | ||
The individual components of the stress tensor are not reliable in | The individual components of the stress tensor are not reliable in this case and must be disregarded. | ||
{{NB|mind|Before you perform relaxations in which the volume or the cell shape is allowed to change you must read and understand the section on [[Energy_vs_volume_Volume_relaxations_and Pulay_stress|energy vs. volume, volume relaxations, and Pulay stress]]. In general volume changes should be done only with an increased energy cutoff, ''i.e.'', {{TAG|ENCUT}}{{=}}1.3×max({{TAG|ENMAX}}) or {{TAG|PREC}}{{=}}High.}} | {{NB|mind|Before you perform relaxations in which the volume or the cell shape is allowed to change you must read and understand the section on [[Energy_vs_volume_Volume_relaxations_and Pulay_stress|energy vs. volume, volume relaxations, and Pulay stress]]. In general volume changes should be done only with an increased energy cutoff, ''i.e.'', {{TAG|ENCUT}}{{=}}1.3×max({{TAG|ENMAX}}) or {{TAG|PREC}}{{=}}High.}} |
Revision as of 07:54, 21 September 2023
ISIF = 0 | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8
Default: ISIF | = 0 | for IBRION=0 (molecular dynamics) or LHFCALC=.TRUE. |
= 2 | else |
Description: ISIF determines whether the stress tensor is calculated and which principal degrees-of-freedom are allowed to change in relaxation and molecular dynamics runs.
Mind: ISIF=8 is only available as of VASP.6.4.1. |
ISIF determines whether the stress tensor is calculated. The calculation of the stress tensor is relatively time-consuming, and therefore by default switched off for ab initio molecular dynamics runs. The forces are always calculated.
In addition ISIF determines which degrees-of-freedom (ionic positions, cell volume, and cell shape) are allowed to change.
The following table shows all combinations:
ISIF calculate degrees-of-freedom forces Stress tensor positions cell shape cell volume 0 yes no yes no no 1 yes trace only yes no no 2 yes yes yes no no 3 yes yes yes yes yes 4 yes yes yes yes no 5 yes yes no yes no 6 yes yes no yes yes 7 yes yes no no yes 8 yes yes yes no yes
In VASP the stress tensor is defined as the negative of the derivative of the energy with respect to the strain tensor:
This might be different from other first principles codes. A positive in the diagonals means that the system is under compressive strain and wants to expand. A negative value implies that the system is under tensile strain and wants to reduce its volume.
For ISIF=1, only the trace of the stress tensor is calculated. This means only the total pressure is correct and can be read off in the line:
external pressure = ... kB
The individual components of the stress tensor are not reliable in this case and must be disregarded.
Mind: Before you perform relaxations in which the volume or the cell shape is allowed to change you must read and understand the section on energy vs. volume, volume relaxations, and Pulay stress. In general volume changes should be done only with an increased energy cutoff, i.e., ENCUT=1.3×max(ENMAX) or PREC=High. |
Mind: When LHFCALC=.TRUE (for Hartree-Fock/hybrid calculations), by default ISIF=0 and the stress tensor is not calculated. |