Hello, Sonu

There are several codes for extracting the anharmonic force constants. Two codes that are based on DFT supercell calculations are the temperature dependent effective potential (TDEP) method (Phys. Rev. B 88 (14) (2013) 144301) and Phono3py (Phys. Rev. B 91 (9) (2015) 094306). Your choice of code could depend on whether you need an effective hamiltonian for renormalizing the phonon modes.

Thank you Rasmus Tranås for your reply. Your answer is complete and helps me a lot.

I am becoming more curious to know if there exists an alternative to the supercell method for 3rd order IFC calculation, like the density-functional perturbation theory (DFPT) is used for 2nd order IFC calculation?

Thank you in advance.

No problem, it is good that my answers could help. Supercell calculations are needed for the IFC calculations for the codes i mentioned, and they are very accurate. In my experience the IFCs themselves can be very sensitive to inaccuracies in the force calculations and supercell size (and thus also force constant cutoffs), so there are few shortcuts.

In your main question it seems that you need the phonon thermal conductivity. If you need fast estimates (without doing supercell calculations) for the lattice thermal conductivity, you have the possibilities of using other methods, e.g. Slacks (Article NonmetalHc Crystals with High Thermal Conductivity

Article Model for Lattice Thermal Conductivity at Low Temperature

Several machine learning methods have been shown to be quite accurate for predicting the lattice thermal conductivity as well. The codes are not always published with the works, but it is always possible to ask the authors (J. Phys. Chem. A 2021, 125, 1, 435–450 ) (Article Charting Lattice Thermal Conductivity for Inorganic Crystals...

Dear Rasmus Tranås,

Thank you for your valuable suggestions. Sure, my concern was related to the accuracy of the calculation. Now I will look at the references you suggested.

Thanks.

Sorry, this isn't my field.