** Next:** Stochastic boundary conditions
** Up:** Performing the simulations
** Previous:** Langevin dynamics in NVT
** Contents**
** Index**

*N.B. This document is no longer maintained, please visit our wiki.*
####

Parrinello-Rahman (NpT) dynamics with Langevin thermostat

The Parrinello-Rahman dynamics is currently available only in connection
with Langevin thermostat 6.62.5. The geometric constraints and metadynamics are not
supported in the current version of VASP. See Sec. 6.62.6 and 6.62.5 for brief
description of the Parrinello-Rahman dynamics and Langevin thermostat, respectively.
- Use the same setup as for Langevin dynamics in NVT ensemble (see Sec. 6.62.2)
but set
`ISIF=3` to allow for the cell volume and
cell shape variations. At the moment, dynamics with fixed volume + variable shape (`ISIF=4`)
and fixed shape + variable volume (`ISIF=7`) are not available.
- Use
`LANGEVIN_GAMMA_L` to set friction coefficient for lattice degrees of freedom
(see 6.62.3)
- Set mass for the lattice degrees
of freedom using the parameter using the parameter
`PMASS` 6.62.3
- Optionally, external pressure (in kB) can be defined using the parameter
`PSTRESS`

Note that the temperatures listed in the file `OSZICAR` are
computed using the kinetic energy including contribution from both
atomic and lattice degrees of freedom. The external pressure
for a simulation can be computed as one third of trace of stress-tensor
corrected for kinetic contribution listed in OUTCAR. This can be achieved
e.g. by using the following command:
grep "Total+kin" OUTCAR| awk 'BEGIN {p=0.} {p+=($2+$3+$4)/3.} END {print "pressure (kB):",p}'

IMPORTANT: In Parrinello-Rahman dynamics, components of stress tensor are used to define
forces acting on lattice degrees of freedom (see Ref. [79,80] for details).
In order to achieve a reasonable quality of sampling (or even to avoid numerical problems),
it is essential to eliminate Pulay stress. Unfortunately, this usually requires rather
large value of `ENCUT`.
The setting with `PREC=low` frequently used in
NVT MD is not recommended for molecular dynamics with variable cell volume.
For more details on the Pulay stress see 7.6.

N.B. Requests for support are to be addressed to: vasp.materialphysik@univie.ac.at