Hi,
I want to compute dielectric properties of systems containing TIP4P water. I need the total dipole time correlation function, so would like to avoid post-processing trajectories. I’ve been doing this with compute property/atom to get the coordinates and charges and then just summing up coord x charge every few time steps to get the total dipole. But this doesn’t quite work with TIP4P since I need the coordinate of the massless charge site instead of the oxygen site.
As far as I can tell, I don’t think I can just do something like “compute c1o waterox property/atom xm ym zm” to get the massless site coordinates? Can I simply hack the "dump custom” code to add an option for outputting the massless site? Or is there more to it than that? I’d rather not have to deal with changing the source code but if it’s just a matter of adding the extra option maybe I can figure out how to do that without too many headaches 
Thanks!
Chris Daub
NTNU
Hi,
I want to compute dielectric properties of systems containing TIP4P water. I need the total dipole time correlation function, so would like to avoid post-processing trajectories. I’ve been doing this with compute property/atom to get the coordinates and charges and then just summing up coord x charge every few time steps to get the total dipole. But this doesn’t quite work with TIP4P since I need the coordinate of the massless charge site instead of the oxygen site.
As far as I can tell, I don’t think I can just do something like “compute c1o waterox property/atom xm ym zm” to get the massless site coordinates? Can I simply hack the "dump custom” code to add an option for outputting the massless site? Or is there more to it than that? I’d rather not have to deal with changing the source code but if it’s just a matter of adding the extra option maybe I can figure out how to do that without too many headaches
since the direction of the dipole vector will not change between using
the position of the O atom and the location of the M point, all you'd
have to do is to compute the corresponding scaling factor from the
incorrect dipole to that of TIP4P. a bigger problem is going to be
that you likely will have to apply minimum image conventions for each
water molecule, i.e. you would have to make sure you start from a
proper initial configuration and then use unwrapped coordinates.
axel.
I already am using the unwrapped coordinates so that should be OK. But you’re right, it is trivial to correct each individual molecular dipole, and if I’m not mistaken I can equally well correct the total system dipole at the end by the same factor. Saves me some headaches! Thanks for the help, much appreciated. Chris.