[gmx-developers] Re: Writing the electric field (and its gradient) on particular atoms during an MD simulation

[John Chodera]

As an addendum to this:

I believe that the electric field defined in this manner can be
computed from the force on atom i due to only the electrostatic
contribution.

The gradient of the electric field only needs to be computed with
respect to the (atomic) position at the field is measured -- this
forms a 3x3 symmetric matrix. This could either be analytically
computed or could be extracted from a finite-difference approximation
to this difference -- the accuracy of finite-difference would be fine
for our purposes.

We would need to write out this information once every 20 fs, where
the simulation is usually run with a timestep of 1 fs, if this is
necessary to consider for reasons of efficiency.

Cheers,

- John

--
Dr. John D. Chodera <jchodera at gmail.com>      | Mobile    : 415.867.7384
Postdoctoral researcher, Pande lab            | Lab phone : 650.723.1097
Department of Chemistry, Stanford University  | Lab fax   : 650.724.4021
http://www.dillgroup.ucsf.edu/~jchodera