[gmx-users] Understanding the gmx dipoles and gmx dielectric GROMACS functionality

[Patel, Lara Anne]

To whom this may concern,


I have been trying to find documentation for the gmx dipoles and gmx dielectric functionalities in GROMACs and have found that what is out there is remarkably sparse.


I would really appreciate some clarification on what each of these functionalities actually do.  As far as I can tell, the gmx dipoles outputs two files that are of interest if you want to calculate the static dielectric constant:

        Epsilon(0) = 1 + 4 Pi/3 * (<M^2> - <M>^2)/(V k_B T)

where M is the systems total dipole moment, V is the system volume, k_B is the Boltzmann constant, and T is the system temperature.


The first file is aver.xvg which outputs the averages for the total dipole moment M. Is this a cumulative average of the systems (based on your index selection) total dipole moment and related values?


The second file is epsilon.xvg.

  1.  Is the epsilon value in this file a cumulative average as well?
  2.  Is it the static dielectric constant or something else?
  3.  If it is the static dielectric constant, what Volume is used? Is it the systems average volume up to that point (which becomes important in NPT simulations) or is it the systems average volume over the entire simulation?  Although rigorously the dielectric should be calculated on simulations in the NVT simulation, it is of interest to know how the volume is handled when in an NPT simulation and how significant volume fluctuations are.

The gmx dielectric functionality is also a bit confusing since it is not entirely clear what is being calculated there.  Is gmx dielectric's aim to calculate the frequency dependence of the dielectric constant? And if so, how is that executed?

Best Regards,

Lara Patel