Re: [ESPResSo-users] Periodic boundary condition.

[Axel Arnold]

On Tuesday 08 November 2011 22:03:26 Joe wrote:
> Hello developers,
> 
> I have a question about the periodic boundary condition.
> Followed the 2.3. Creating the first simulation script part from the user's
> guide, I set the MD box as the following:  setmd box_l $box_l $box_l
> $box_l . I tried to run it without  'setmd periodic 1 1 1'. I didn't find
> any difference except for negative coordinates in vtf files. When box_l is
> positive, should the coordinates always be equal to or greater then zero?

Hi,

first, if you don't setmd periodic, the default is 1 1 1, so no wonder you do 
not see a difference. If you want non-periodic boundaries, you need to setmd 
periodic 0 0 0.

Second, the coordinates that Espresso writes out are unfolded coordinates, and 
therefore by diffusion take any values, positive or negative. You can call 
writevtf with the parameter "folded" to get folded coordinates, which are then 
between 0 and box_l, if you use periodic boundary conditions. If not, the 
particles can always have negative positions, as long as you don't put 
constraint walls into the system.

Cheers,
Axel

-- 
JP Dr. Axel Arnold
ICP, Universität Stuttgart
Pfaffenwaldring 27
70569 Stuttgart, Germany
Email: address@hidden
Tel: +49 711 685 67609