When I tried including SOC during the optimization of InBi lattice constants, the lattice constant increased from 4.6 Å (without SOC) to more than 6 Å (with SOC), which seems like an 'incorrect' result when including SOC.

It seems that most papers only include SOC when calculating electronic properties, but rarely using SOC during geometry optimization. Is this because SOC yields 'strange' structures, or because including SOC generally does not significantly affect the structure of most materials? Thank you very much in advance!

  • $\begingroup$ I've seen very detailed studies of phonons with spin-orbit coupling in other codes, implying that it gets structures right... have you looked at the vasp forum for similar issues? $\endgroup$ Commented May 22 at 18:27
  • $\begingroup$ Which XC functional are you using? Is the material magnetic? $\endgroup$ Commented May 22 at 22:22
  • $\begingroup$ Hi @PhilHasnip I was using PBE functional, I don't think InBi is magnetic. $\endgroup$
    – James
    Commented May 22 at 23:48


You must log in to answer this question.