tuning SOC in PAW calculation

[zjuer]

Hello all,
In my PAW calculation I want to tune the spin-orbit coupling (SOC) effect from 0% (nonrelativistic) to 100%(relativistic) gradually. How can I realize this goal? I have read the tutorial of abinit about PAW calculation, I guess I have to modify the atompaw code to change the ratio of SOC effect and generate PAW pseudopotential with different SOC ratios, but am not sure. Is there anyone familair with this topic can help out? Many thanks.

[jzwanzig]

In abinit, PAW SOC calculations are done with the same PAW datasets as non-SOC calculations. There is a parameter, spnorbscl, that does what you want. See the input variables documentation for abinit.