[gmx-users] Anomalous temperature

[Mark Abraham]

On 3/03/2012 5:08 PM, zhongjin wrote:
> Dear gmx users,
>    It seems that I have found a strange temperature value after NVT 
> equilibrium for 200 ps. I simulate a CNT with two graphite sheets in 
> water box. I do not want to them to deform or change their shaps, So I 
> use a very large position restraint (1*10^6) to keep their shaps. The 
> temperature was set to 300 K, but after NVT equilibrium, the 
> tempertaure was 330K.
> .mdp content
> define                   = -DPOSRES_CNTGATE -DPOSRES_K -DPOSRES_CL
> ; OPTIONS FOR WEAK COUPLING ALGORITHMS
> ; Temperature coupling
> tcoupl                   = v-rescale
> nsttcouple               = -1
> nh-chain-length          = 1
> ; Groups to couple separately
> tc-grps                  = Other Water_and_ions
> ; Time constant (ps) and reference temperature (K)
> tau-t                    = 0.1 0.1
> ref-t                    = 300 300
> results
> Energy                      Average   Err.Est.       RMSD  Tot-Drift
> -------------------------------------------------------------------------------
> Temperature                 331.548        2.6    10.3739    15.0042  (K)
> T-Other                     436.927         12    39.4112    67.2519  (K)
> T-Water_and_ions            303.349       0.16    6.07417    1.02328  (K)
> the Other, that is CNT and two graphite sheets temperature is very high!!!
> Howerver when I freeze them, the temperture is OK
> freezegrps               = K CL Other
> freezedim                = Y Y Y Y Y Y Y Y Y
> the results is
>
> Energy                      Average   Err.Est.       RMSD  Tot-Drift
> -------------------------------------------------------------------------------
> Temperature                 299.998     0.0034    4.84173 -0.0066645  (K)
> T-Other                           0          0          0          0  (K)
> T-Water_and_ions            299.998     0.0034    4.84173 -0.0066645  (K)
> So I should freeze them or use a small postion restraint to constaint 
> them, in order to get the correct temperture?
>

By introducing position restraints, you cause local heating because 
there is a force whenever the positions deviate. Don't try to measure 
local temperature until the model might make a sensible measurement, 
i.e. after equilibration. That might mean several equilibration phases 
where you relax your position restraints gradually.

The thin nature of the sheet and tube is also somewhat unfavorable for 
the correct function of separate T-coupling groups. See manual 3.4.8, 
particularly final subsection and ref therein.

Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20120303/aa69448c/attachment.html>