[gmx-users] water temperature 10 to 20 degrees higher than specified value

[X.Periole]

> Question 1:
> What could I be doing wrong?
> 
> I'm simulating cellulose (oligomer of nine rings) in 
>water using 1.4 nm cut-off, Reaction Field, and LINCS 
>bond-constraints (generally trying to follow the recipe 
>used by H. Yu, M. Amann, T. Hansson, J. Köhler, G. Wich 
>and W. F. van Gunsteren. "Effect of methylation on the 
>stability and solvation free energy of amylose and 
>cellulose fragments: a molecular dynamics study." 
>Carbohydr. Res., 339, 1697-1709 (2004). 
>http://dx.doi.org/10.1016/j.carres.2004.05.003 
><http://dx.doi.org/10.1016/j.carres.2004.05.003> ).  The 
>cellulose and water are separately coupled to a Berendsen 
>thermostat with tau_t of 0.1.  The system contains one 
>cellulose oligomer (129 atoms) and something like 16000 
>water molecules.  I'm simulating at various temperatures 
>in the range of 283 to 363 K.  The cellulose is always 
>close to the specified temperature but the water 
>temperature is always higher by some 10 to 20 degrees. 
> True, I'm not using PME, but this large discrepancy 
>seems unacceptable to me.  
> 
The problem might be the cut-off. You should be using a
twin-range cut-off, 0.8/1.4 nm, if you don't. Also
increasing the frequency of update of the lists should
help: use 5 steps.
> 
> Question 2:
> Could someone give an update on the status of implicit 
>solvent capabilities in GROMACS?
> 
> As far as I can tell, the ever-popular "implicit 
>solvent" question has been asked twice in 2005 on 
>gmx-users without being answered:
> 
No implict solvent at this time in GROMACS.

XAvier