Dear All, I want to simulate the infrared spectrum of a molecular crystal. As the phonon calculations usually take weeks to finish, I'd like to ask, wether someone of you has propositions on proper treshold values?
I'm using rpb-paw US pseudopotential (revised PBE, as suggested for vdW-potential) and gamma point (also said to be sufficient for molecular crystals). Does the ecutrho of the preceding SCF calculation influence the speed of the phonon calculation? What is a sufficient low value for the tr2_ph to get a reasonable (comparable to experimental values) spectrum? Thanks a lot, Guntram
