[gmx-users] infinite systems

[Anton Feenstra]

Yuguang Mu wrote:
> Dear David,
> I want to simulate  an infinite system, a small peptide in water
> the system is quite small, total 200 atoms.
> if I use normal cut-off, the rcut  has to be smaller than box/2 (around 5
> A), then the simulated system is nonphysical, due to the small cutoff.
> I try to use setenv GMXFULLPBC 1, as:  export GMXFULLPBC=1, then use
> rcut=1nm .
> But the grompp still say it is a error : rcut > box/2.
> What should I do ?

Use a larger system. Since we do not have an 'infinite' long range method
for Van der Waals interactions (like PME does for Coulomb), you are stuck
with cut-offs. If you take your cut-off larger than half your box, you will
be counting interactions twice, and at some point atoms will interact with
their own periodic images. GMXFULLPBC doesn't change that, it only fixes
some assumptions that you cannot make if a molecule (e.g. a polymer)
stretches across the PBC (i.e., you have an infinite topology).


-- 
Groetjes,

Anton
  _____________ _______________________________________________________
|             |                                                       |
|  _   _  ___,| K. Anton Feenstra                                     |
| / \ / \'| | | Dept. of Pharmacochem. - Vrije Universiteit Amsterdam |
|(   |   )| | | De Boelelaan 1083 - 1081 HV Amsterdam - Netherlands   |
| \_/ \_/ | | | Tel: +31 20 44 47608 - Fax: +31 20 44 47610           |
|             | Feenstra at chem.vu.nl - www.chem.vu.nl/~feenstra/       |
|             | "If You See Me Getting High, Knock Me Down"           |
|             | (Red Hot Chili Peppers)                               |
|_____________|_______________________________________________________|