Electric field simulation: eFieldNormalized

From: Carlo Guardiani (carlo.guardiani_at_dsf.unica.it)
Date: Tue Nov 28 2017 - 13:53:57 CST

Dear NAMD experts,
I would like to run constant electric field
simulations of an ion channel embedded in
its membrane. At first I tried to run my
simulation at constant pressure using the
commands:

eFieldOn yes
eField 0.0 0.0 -0.134
eFieldNormalized yes

The simulation however, aborted immediately
with the following error message:

"FATAL ERROR: Periodic cell has become too
small for original patch grid!
Possible solutions are to restart from a
recent checkpoint, increase margin, or
disable useFlexibleCell for liquid
simulation."

I therefore tried to rerun the job at
constant volume and the simulation
could run without problems. However,
I don't know what to do with the
instruction

eFieldNormalized yes

As far as I understand, this statement
is intended to avoid excessive pressure
in an NPT simulation in the presence of
a constant electric field. Do I have to
keep this instruction also when running
in the NVT ensemble ? And in such a case,
which voltage would I be actually applying ?

In the absence of scaling of the electric field
the voltage should be computed as:

V = Ez*Lz*43.17

where V is the voltage in mV, Ez is the electric
field in kcal/(mol*Ang*e), Lz is the dimension of
the simulation box in Angstrom and 43.17 is the
conversion factor.

If the eField vector is scaled by the reciprocal
lattice vectors at each timestep, the effective Ez
is no longer the nominal one declared with the
eField < electric field vector > instruction in the
input file. Which formula should I use then to compute
the effective potential I am applying ?

Many thanks for your help and best wishes,

Carlo Guardiani

This archive was generated by hypermail 2.1.6 : Sun Dec 31 2017 - 23:21:48 CST