Dear guohuaihong,
I'm not sure that I understand what you are precisely meaning, but the potential energy at each step of an MD run is given in the file *.MDE, under the column E_KS. This means that E_KS is the potential energy of the system, understood as the total energy (given in the E_tot column) minus the kinetic energy (which is given in the form of a temperature in column T).
I hope this answers your question. I don't know either what kind of system are you simulating, but in case it is a finite system like a molecule or cluster, keep in mind that the temperature definition depends on the statistical ensemble. The T given in the *.MDE file is just one possible definition, extracted from equipartition as applied to the kinetic energy.
Best wishes, Andres Aguado On Fri, 18 Feb 2011, guohuaihong wrote:
Dear SIESTA users: I intend to get system potential energy (including electronic, ionic and pseudopotential) for each MD step as doing Nose molecular dynamics (MD) simulation in SIESTA. I have a few questions to ask: 1. How to obtain system potential? As we know, in each MD step, code only gives out E_KS and Free Energy of electrons. The total energy or the total potential won't be given until the last MD step. 2. The clue may lie in the following setting. SaveNeutralAtomPotential True SaveTotalPotential True However, according to the SIESTA-3.0-b manual, "SaveTotalPotential" is "to write the valence total effective local potential (local pseudopotential + Hartree + Vxc), at the mesh...". Does it include both electronic and ionic contributions? 3. If "SaveTotalPotential" only refers to the electronic part, I guess we have to use "SaveNeutralAtomPotential" to calculate ionic part. Is it correct? "SaveNeutralAtomPotential" is a new command in 3.0-b version compared with stable version 2.0.2 and defines "the sum of the hartree potential of a pseudo atomic valence charge plus the local pseudopotential". Then, both have local pseudopotential part, how can we avoid the double-counting? 4. How to save those potential in one file (such as *.VT or VH) in sequence like *.ANI during the MD runs. Any suggestion will be appriciated! Thank you very much! Best wishes, HH GUO
