fix pafi command


fix ID group-ID pafi compute-ID Temp Tdamp seed keyword values...
  • ID, group-ID are documented in fix command

  • pafi = style name of this fix command

  • compute-ID = ID of a compute property/atom that holds data used by this fix

  • Temp = desired temperature (temperature units)

  • Tdamp = damping parameter (time units)

  • seed = random number seed to use for white noise (positive integer)

  • keyword = overdamped or com

    overdamped value = yes or no or 1 or 0
      yes or 1 = Brownian (overdamped) integration in hyperplane
      no or 0 = Langevin integration in hyperplane
    com value = yes or no or 1 or 0
      yes or 1 = zero linear momentum, fixing center or mass (recommended)
      no or 0 = do not zero linear momentum, allowing center of mass drift


compute pa all property/atom d_nx d_ny d_nz d_dnx d_dny d_dnz d_ddnx d_ddny d_ddnz
run 0 post no
fix hp all pafi pa 500.0 0.01 434 overdamped yes


Perform Brownian or Langevin integration whilst constraining the system to lie in some hyperplane, which is expected to be the tangent plane to some reference pathway in a solid state system. The instantaneous value of a modified force projection is also calculated, whose time integral can be shown to be equal to the true free energy gradient along the minimum free energy path local to the reference pathway. A detailed discussion of the projection technique can be found in (Swinburne).

This fix can be used with LAMMPS as demonstrated in examples/PACKAGES/pafi, though it is primarily intended to be coupled with the PAFI C++ code, developed at https://github.com/tomswinburne/pafi, which distributes multiple LAMMPS workers in parallel to compute and collate hyperplane-constrained averages, allowing the calculation of free energy barriers and pathways.

A compute property/atom must be provided with 9 fields per atom coordinate, which in order are the x,y,z coordinates of a configuration on the reference path, the x,y,z coordinates of the path tangent (derivative of path position with path coordinate) and the x,y,z coordinates of the change in tangent (derivative of path tangent with path coordinate).

A 4-element vector is also calculated by this fix. The 4 components are the modified projected force, its square, the expected projection of the minimum free energy path tangent on the reference path tangent and the minimum image distance between the current configuration and the reference configuration, projected along the path tangent. This latter value should be essentially zero.


When com=yes/1, which is recommended, the provided tangent vector must also have zero center of mass. This can be achieved by subtracting from each coordinate of the path tangent the average x,y,z value. The PAFI C++ code (see above) can generate these paths for use in LAMMPS.


When overdamped=yes/1, the Tdamp parameter should be around 5-10 times smaller than that used in typical Langevin integration. See fix langevin for typical values.

Restart, fix_modify, output, run start/stop, minimize info

No information about this fix is written to binary restart files. None of the fix_modify options are relevant to this fix.

This fix produces a global vector each timestep which can be accessed by various output commands.


This fix is part of the EXTRA-FIX package. It is only enabled if LAMMPS was built with that package. See the Build package page for more info.


The option defaults are com = yes, overdamped = no

(Swinburne) Swinburne and Marinica, Physical Review Letters, 120, 1 (2018)