[CPMD-list] Energy conservation

[WANG, YIXUAN]

Dear all,

  I do agree the point from Juerg that the problem of energy conservation 
with systems is often related to the thermostats.
  I have no experience that if the decomposition ratio in nose parameters(default 
1) is responsible much for it, however the energy conservation is really sensitive
to the set kinetic energy of electron. If it is too small, the energy usually
monotonically decrease. 

Yixuan
 


-----Original Message-----
From: cpmd-list-admin at cpmd.org [mailto:cpmd-list-admin at cpmd.org]On
Behalf Of Dr. Balasubramanian Sundaram
Sent: Wednesday, October 15, 2003 11:31 PM
Cc: Nick Cooper; cpmd-list at cpmd.org
Subject: Re: [CPMD-list] Energy conservation


Hi Nick,
   In addition, a couple of things came to my mind. Sorry if you have
thought/taken care of these before.

    -- Has the system been "equilibrated" using MD with empirical
potentials,
       prior to the CPMD run? That might help. Since the operating
temperature
       is reasonably large, I am assuming that the system is not
crystalline,
       and that you have an initial configuration that is not obtained
from
       experiments. Hence, my concern about equilibration being a
possible 
       reason for relatively poorer energy conservation. 
       
    -- Is the total potential energy (E_{KS}?) of the system too
exhibiting
       a monotonic decrease? If yes, then the monotonic decrease in the 
       conserved energy may not be from "improper" choice of
timestep/thermostat
       parameters. May be, the system is just relaxing. 

Regards,
Bala


Juerg Hutter wrote:
> 
> Hi Nick
> 
> the problem with energy conservation with systems of this
> type is often related to the thermostats.
> Instead of reducing the timestep you should try to
> increase the order of the integrator and the
> decomposition ratio for the thermostats.
> You do this by setting
> 
> NOSE PARAMETERS
>  4 4 4 6  125 4
> 
> other choices for the Suzuki integrator are
> 3,5,7,9,15,25,125,625
> 
> The decomposition ratio is by default 1 (last number)
> and I did set it to 4. I don't have too much experience
> with that.
> 
> regards
> 
> Juerg
> 
> ----------------------------------------------------------
> Juerg Hutter                   Phone : ++41 1 635 4491
> Physical Chemistry Institute   FAX   : ++41 1 635 6838
> University of Zurich           E-mail: hutter at pci.unizh.ch
> Winterthurerstrasse 190
> CH-8057 Zurich, Switzerland
> ----------------------------------------------------------
> 
> On Wed, 15 Oct 2003, Nick Cooper wrote:
> 
> > Hello all,
> >
> > I am simulating covalent materials at relatively high temperatures (3500K)
> > and I have noticed that the conserved quantity in a canonical ensemble is
> > monotonically decreasing, though admittedly the absolute change is small.
> > I have tried decreasing the timestep (from 3au down to 1au) but all this
> > does is decrease the gradient of the drop in energy. The systems I am
> > looking at have significant metallic character at this temperature, so I
> > have used a Nose-Hoover themostat for the ions and wavefunctions.
> >
> > The monotonic nature is what has me worried- am I just being paranoid?
> >
> > Any advice you can offer would be gratefully received.
> >
> > Yours sincerely,
> >
> > Nicholas Cooper
> >
> > PhD Student,
> > The University of Sydney,
> > NSW 2006 AUSTRALIA
> > _______________________________________________
> > CPMD-list mailing list
> > CPMD-list at cpmd.org
> > http://www.cpmd.org/mailman/listinfo/cpmd-list
> >
> _______________________________________________
> CPMD-list mailing list
> CPMD-list at cpmd.org
> http://www.cpmd.org/mailman/listinfo/cpmd-list