[CPMD-list] Energy components in PIMD

[Łukasz Walewski]

Dnia 2008-04-29, wto o godzinie 07:26 -0400, Axel Kohlmeyer pisze:
> 
> dear lukasz,
> 
> can your input conserve energy for a non-PIMD run?
> ...and then you have to check whether you can maintain
> adiabacity with the massive thermostat on the ions only.
> quite a few times, the no-thermostat CP-dynamics is only
> meta-stable and when you switch to massive N-H chains you
> excite DOFs that will make the CP dynamics lose it.

I am trying to understand which DOF you mean - electronic ? ionic ? I
assume that after 10ps of NVT simulation in 300K those DOF are already
excited - that's why the dynamics is not adiabatic. Now, what will
happen if I turned out the termostat on the ions and restart the
simulation ? If the cause of non-adiabaticity was accidentally inactive
DOFs then the non-adiabaticity should remain unchanged also in NVE
simulation. If the restarted NVE run was stable I would assume that the
cause is different - perhaps connected with the thermostat itself.

Going even further isn't it that the electrons in NVT simulation tend to
achieve the energy equipartition, i.e. the kinetic energy of the
electronic DOFs rises to achieve the level of 0.5kT per DOF ? And since
we allow the energy gain from the ionic thermostat the electrons just
absorb it. This would mean that there is a strong coupling between the
electronic and ionic DOF in my simulation which is exactly the CPMD
catastrophe. In this case decreasing the electron mass should improve
this.

Could you please comment on that, is there a seed of reason or did I
lost the point totally ?

Regards,
Lukasz