A well-known way to calculate thermodynamic properties of a material at finite temperatures is minimization of the Gibbs free energy, which includes “cold” static energy E(V) calculated by DFT, contributions of lattice vibrations, electron excitations, etc. But does anybody knows publications regarding the influence of number of volume points, their range relative to the equilibrium, etc. of the static energy E(V) curves on the obtained thermodynamic properties? Or may be someone has personal experience and can give a comment? My preliminary analysis shows that the above-mentioned parameters can affect the results and thus an issue of correct choice of them appears.
I have published several papers using the quasi-harmonic approximation to calculate phase transitions including temperature, as you mention it is critical the number of points calculated in the E-V curve. I used to perform the calculations from E-V with a large number of points. In the second part, I choose a small number of points to get the phonons in the whole Brillouin zone (this point is crucial, it should be the entire BZ) to observe the changes with volume. Then, with a general picture of the results, I used to perform the phonons calculations with more points to get better results.
Regards
Dear Sinhué López ,
Thank you for your answer. Can you tell how many points in calculations of E-V curve do you use and how far are the farthest points from the equilibrium (in percents)?
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Thermodynamics