[gmx-users] Some Issues on Substrates/PBC

[Justin A. Lemkul]

Michael Hirtz wrote:
> Hi Justin,
> 
>> In addition to that, you may consider using pbc = full to get an
>> infinite system and potentially avoid some of those edge effects it
>> looks like you are seeing. I know others have used that kind of
>> setup with carbon nanotubes (not my area of expertise, but I've read
>> the discussions).
> 
> Thanks a lot! That was a great tip! :) With the pbc = full setting I can use
> my topologies with bonds defined over the periodic boundaries without messing
> up the structure. I'll use the freezegrps additionally nevertheless because it
> speeds up the simulation and my substrate isn't supposed to move anyway, but
> with the 'full' option it would even work out without.
> 
> I only wonder what's the difference between 'full' and 'xyz'? In the manual
> 'full' is not even mentioned and on the list archives I found several posts
> that tell that 'xyz' is a kind of optimization that does not work for infinite
> systems (hence my problems), and it could not be used with constraints in
> earlier versions, but now it's fine. Is there anything more I should know
> about the 'full' option, or why is it not listed in the run parameters section
> of the manual?

Indeed, it's a little strange that pbc = full does not appear in any official 
documentation.  Come to think of it, I can't remember when I first actually 
learned about it myself :)  The difference is that pbc = xyz is your standard 
3-dimensional PBC option, but pbc = full is for molecules that repeat across 
periodic boundaries (i.e., an infinite CNT, or the bonds within your graphite 
sheet that are between atoms on opposite "sides" of the box).

-Justin

> 
> Best regards,
>    Michael

-- 
========================================

Justin A. Lemkul
Graduate Research Assistant
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================