[Octopus-users] Potential Energy fluctuation during Ehrenfest-TDDFT MD simulation

[matthieu verstraete]

It looks like your kinetic energy peaks when the center of mass has
largest speed, whereas the O2 vibrations probably induce a much
smaller oscillation of the total energy (need to zoom in a bit).

You will need to test the parameters yourself, to know if the box is
large enough and time step small enough. How about 1 atom? There is
even less reason to move the center of mass, there. Is the COM
initially at 0,0,0?

Matthieu

2009/8/3 Yasumitsu Suzuki <ysuzuki at tcl.t.u-tokyo.ac.jp>:
> Dear octopus developers and users,
>
> Hello everyone.
> Today I want to ask you about the fluctuations of energies during Ehrenfest-TDDFT molecular dynamics simulations using octopus.
> I have performed several Ehrenfest-TDDFT MD simulations with octopus so far, and I found that in most cases of mine, total energies were not conserved and potential energies fluctuated largely, with the width around 0.1 Hartree.
> So to investigate the reason of this, I reduced the system to simple oxygen molecule and performed Eherenfest -TDDFT MD simulations about it with no initial velocities (0 K).
> However, I noticed that potential energies fluctuated largely with 0.1 Hartree even in the case of simple oxygen molecule.
> My inp file like this:
>
> CalculationMode = td
> Dimensions = 3
> PeriodicDimensions =0
> BoxShape=parallelepiped
> Lsize = 13.0
> spacing=0.40
> %Coordinates
> "O" | 0 | 1.1397 | 3.78
> "O" | 0 | -1.1397 | 3.78
> %
> ExtraStates = 1
> %Occupations
> 1 | 1
> 0 | 0
> %
> Mixing=0.1
> SpinComponents = spin_polarized
> TDEvolutionMethod= aetrs
> TDMaximumIter=50000
> TDTimeStep=0.04134
> MoveIons = vel_verlet
> TDDynamics=ehrenfest
>
>
> And I attached the figures that shows the changes of obtained energies and oxygen atom - oxygen atom distances and z coordinate of O2 molecule during the simulation.
> >From this figure, it is found that potential energy ([Ion-Ion]+[Electronic]) fluctuated largely, and it had no connection with O-O distances.
> So I can't understand why this large fluctuations happened.
> Is my TDTimeStep (which nearly equals to 1 attosecond) large? Or is my simulation box small?
> Or is it a limit of aetrs and Ehrenfest method?
> Please give me advices about this matter.
> I will greatly appreciate any kind of comment of you.
>
> Yasumitsu Suzuki
> Department of Chemical System Engineering
> Graduate School of Engineering
> The University of Tokyo
>
>
> _______________________________________________
> Octopus-users mailing list
> Octopus-users at tddft.org
> http://www.tddft.org/mailman/listinfo/octopus-users
>
>



-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Matthieu Verstraete

European Theoretical Spectroscopy Facility (ETSF)
Dpto. Fisica de Materiales,
U. del Pais Vasco,
Centro Joxe Mari Korta, Av. de Tolosa, 72,   Phone: +34-943018393
E-20018 Donostia-San Sebastian, Spain        Fax  : +34-943018390

Mail : matthieu.jean.verstraete at gmail.com
http://www-users.york.ac.uk/~mjv500