Hybrid+U and time-dependent hybrid+U

[ashwin_r]

Hello,

Is the current implementation of time-dependent DFT (TDHF) in VASP able to handle a hybrid+U calculation correctly?

While the hybrid+U question has been raised a couple of times before on the forum, the responses have generally argued against it on physical grounds -- I understand this. Nevertheless, regardless of whether hybrid+U is physically meaningful or nonsensical, from a purely technical/implementational perspective, will the code work correctly?

Thank you!

[manuel_engel1]

Hi,

Thanks for reaching out. The code will most likely run, but DFT+U combined with hybrids is untested in this context. We highly recommend to conduct some tests to see if the results make sense.

[ashwin_r]

Thank you. It's good to know that there isn't anything within the implementation that would inherently preclude such a calculation. We will, of course, do our due diligence.