[gmx-users] PBC for ewald-geometry 3dc warning

[Lakshman Ji Verma]

Hi everyone,

I am trying to simulate a crystal slab using 3dc Ewald correction and PBC
in xyz with dimensions 3.5nm*4.4nm*3.2nm and with the box length of more
than 3xZ in the z-direction with my slab in the center of the box, as
recommended in GROMACS manual. I am getting the following warning:

"With PME and ewald_geometry = 3dc you should use pbc = xy"

I looked at different mail list I could find on internet listed below:

https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-users/2017-June/113997.html
https://www.mail-archive.com/gromacs.org_gmx-users@maillist.sys.kth.se/msg27043.html
https://mailman-1.sys.kth.se/pipermail/gromacs.org_gmx-users/2018-February/118464.html

I couldn't find any justification of why we need to used pbc in
xy direction and how it will take care of periodicity in z-direction which
was used in the original paper
https://aip.scitation.org/doi/10.1063/1.479595.

I also want to know if I can ignore this warning and what could be
the repercussions of ignoring this warning?

I have already tried using the 2018 version of GROMACS as suggested in one
of the mail lists but warning persist.

Thanks!
Lakshman