I am working through the section in Allen-Tildesley on calculating
properties from RMS fluctuations. In particular, specific heat,
Cv, from fluctuations in potential (or kinetic) energy.
This works fine in the NVE ensemble, however I am attempting to
use Berendsen's external bath technique to generate a constant
pressure and temperature. According to the 1984 article the
ensemble is undefined and therefore the fluctuation data is
invalid. (Which I verified, neither Cv or Cp were correct)
So what I ended up doing is using the coupling bath to establish
the target temperature/pressure then turn off the bath in order to
generate correct RMS fluctuation data. Unfortunately the pressure
(and temperature) tend to wander away from the target values.
Could someone point me to more recent literature which might have
a better way to do this. Ideally, I would like to get the specific
heat value without having to turn off the external bath.
Thanks in advance.