[gmx-users] CHCL3 box equilibration

[Christoph Freudenberger]

Germana Paterlini wrote:
> 1) I first started with a CHCL3 solvent box (borrowed from AMBER).
> Minimiz is ok and NVT run appears ok. However, when I tried to switch to 
> NPT,
> I get the usual "pressure scaling more than 1%" error. As I understand, 
> this is a sign that
> the box is not equilibrated. I increased pressure coupling (up to 2.5), 
> but it did not help.
> 
> What can I do next?
> Note that, at the end of the NVT (up to 1 ns, 300K, Berendsen), "holes" 
> appear in the box, while other regions
> have higher density. I  increased the cutoffs to 1.7(thinking they are 
> trying to evaporate) but it did not help. What am I doing wrong?
What about the parameters for your chcl3.itp. Did you also borrow them
from amber? I tried to import paramters for ch2cl2 and mecn from an
"amber parameter database" a couple of days ago, but the parameters
are either wrong or in some weird units so the systems i created
behaved quite strange.

 > 2) I tried to build the box myself using the genbox command, but apparently genbox does not recognize
 > VdW radia for CHCL3 and genbox generate molecules on top of each other.
 > The command is : genbox -ci chcl3.gro -nmol #  (where nmol is set to approximate the density of CHCL3).
 > What should I do? Should I add the CHCL3 atom types to the vdwradii.dat file?
You sure did try to run a minimization on that?

Usually I got quite good results with a procedure similar to that.
I use genbox to create a solvent box that is 20-50% lager than that
of the disired density. When I do a minimization on this starting
structure to remove bad contacts followed be an NPT run with tau_p=2
I usally dont get problems... as long as the forcefield is ok.

regards
-- 
Christoph Freudenberger
Abt. Organische Chemie I AK Siehl - Uni Ulm -Tel: ++49-731-502-2785