[CPMD-list] thermostats in CP vs. BO

[Isaac Tamblyn]

I'm not sure if this worked the first time I tried to post, so apologies if this
is a repeat:

I have done a series of runs using the Nose thermostat for both CP and BO MD. My
system consists of 128 H atoms, Rs ~ 1.85. The parameters in both cases were as
follows:

NOSE IONS MASSIVE

 3000 4400

4400 comes from the oscillation frequency of H2.

In the case of CP, I find that the temperature oscillates between ~2500 K to
3500 K. The fictitious electron mass was set at 35, and the timestep at 1 A.U.
<- I could not get the fictitious electron kinetic energy to remain low at
higher timesteps, I worry about lowering the mass much more.

My question is about what happens when I try to use the same thermostat using
BOMD. I find that the temperature varies greatly, up to 6000 K. Does this make
any sense? I am confused how a thermostat acting on the ions would care how
their positions are generated (CP vs. BO). Any suggestions on how to apply
thermostats properly on BO runs would be appreciated.

Also, is it a standard technique to employ a thermostat on the fictitious
kinetic energy of the electrons, and are there not consequences to this?

Thank you in advance,

Isaac