[gmx-developers] Re: [gmx-users] Thermal Conductivity

[Graham Smith]

> > Any of you know about code to compute the Thermal Conductivity?
> > I didn't see this in Gromacs...

> Hmm. That sounds like you will need energies per atom or per
> molecule, or perhaps per coordinate. These are not readily
> available, and I am actually unsure if it is even possible to
> calculate them based on atom-pair interaction energies (which are,
> basically, the only real energies that are used; all others are
> derived from this). 

If you do need the energies per atom (sum of pairwise non-bonded
interactions for each atom), then I have a modified version of the
inner loop code generator and fnbf.c that returns an extra array
e[0..natom-1] containing these energies. I'm happy to send this to you
if you want. My code doesn't the same for the long-range ES energies
or bonded energies though. 

(all the pairwise energies seem to be calculated in the inner loops,
but they are summed into energies for the defined energy groups and
only these sums are returned to the higher-level routines ... I don't
know if the developers think it might be worth while to include the
energy-per-particle array in later versions, even if it's not passed
out of do_fnbf, but just present there so that people who want to
modify the code can get their hands on it easily?) 

Cheers, 

Graham 

########################################################################

Dr. Graham R. Smith, 		       Department of Biological Sciences,  
Biomolecular Modelling Laboratory,     Biochemistry Building,              	
Cancer Research UK, 		       Imperial College of Science,        	
44 Lincoln's Inn Fields, 	        Technology & Medicine,             	
London WC2A 3PX,		       London SW7 2AZ,                     	
U.K. 				       U.K.                                	
Tel: +44-(0)20 7269 3348 	       Tel: +44-(0)20 7594 5737            	
email: graham.smith at cancer.org.uk	
URL: http://www.bmm.icnet.uk/~smithgr