[gmx-users] Viscosity units

[Rini Gupta]

Dear Gromacs Users,

 I am trying to learn viscosity calculations using non-equilibrium method in Gromacs 4.5.4. I have gone through Hess paper and several posts on viscosity calculation.
I have setup a system of pure water (32000 molecules) and run till 10ns using NPT and setting
cos-acceleration=0.1 nm/ps^2 in .mdp file.

Then, I have used following command to get viscosity from energy file.
g_energy_d -f .edr -s .tpr -o energy.xvg

and selected 1/viscosity from one of the options and I get,

Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
1/Viscosity                 1.64374     0.0016  0.0341615 0.00789197  (m s/kg)

This gives Viscosity=1/1.64374= 0.6083 kg/ms=608.3 cP

I have some doubts regarding this result:

1) Can anyone please tell me what are the units of viscosity here.
It shows from screen output (as well from energy.xvg) that 1/viscosity is in m-s/kg units. But, the experimental value for pure water is 0.854 cP =0.000854 kg/m-s. 
So, I have a doubt regarding units coming from GROMACS it is in cP or kg/ms as printed. If kg/ms are the units then, this value of viscosity for pure water is not correct ? 
Please tell me what I am doing wrong here.

2) Is this a right way of calculating viscosity using NEMD or I should
do it by using 2CosZ*Vel-X option. Please tell me how can I get viscosity from this term.  



Thanks and Regards,
Dr. Rini Gupta
UBC, Vancouver
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110727/fd86dc5a/attachment.html>