[gmx-users] Dielectric is essentially 1?

[Steven Spronk]

Hello all,

I have a protein-membrane system in which there are some charges in the 
membrane.  I am using ffgmx with PME long-range electrostatics, and I 
haven't done anything with epsilon_r in my mdp files.

Would it be correct to say that in nonpolarizable force fields, 
membranes have dielectric constants of 1, as opposed to 2 or 4, which 
people often quote as a typical dielectric for membrane or protein 
environment?  Or course, in a real system, the Coulombic energy between 
two charges is reduced when they are in some medium.  But this effect 
arises from polarization in the intervening molecules and/or 
reorientation of intervening dipoles (such as water), right?  So in a 
system in which atoms are not polarizable (at least for ffgmx), and in 
a bilayer that doesn't have water nearby for screening, the dielectric 
is essentially 1?  The energy between two charges in the membrane, as 
calculated by the force field, is *not* any different than if they were 
in a vacuum, right?  I'm just making sure that I really understand the 
force field terms.

Thanks,
Steve