Dear lammps users,
I am a very new user of lammps and I want to calculate specific heat (Cv) for the metal clusters using lammps. I have performed some NVT run and it is observed that the specific heat continuously goes on decreasing with temperature. Now my question is whether I should run the NVT calculation for this purpose, ? Or should I first run the NVE calculation and then go for NVT?
Thanks in advance.
With Regards,
Bedamani
that looks more like a personal choice to me. if you want to compute Cv as dE/dT, then (in my limited experience) both NVE and NVT would yield good results (please let me know if im wrong).
anyhow, i think that the advantage of NVE is that you can definitely use the Cv ~ variance(E), which is the method i’d choose for Cv calculation. I don’t know if the canonical ensemble relations can be use that easily with NH thermostat.
that looks more like a personal choice to me. if you want to compute Cv as
dE/dT, then (in my limited experience) both NVE and NVT would yield good
results (please let me know if im wrong).anyhow, i think that the advantage of NVE is that you can definitely use the
Cv ~ variance(E), which is the method i'd choose for Cv calculation. I don't
know if the canonical ensemble relations can be use that easily with NH
thermostat.
the canonical (NVT) ensemble describes a bulk system in contact with
an infinite reservoir at a given temperature T. what exactly does a
cluster couple to and how?
also, NVE assumes a bulk system, since a constant volume doesn't
really happen for a cluster. in fact, at the nano scale most of the
(bulk) statistical thermodynamics considerations are only weakly
applicable. those are usually derived and strictly applicable only for
macroscopic systems.
axel.