A possibility of CP2K implementation of r2SCAN-3c?

30 views

Skip to first unread message

Ji Woong Yu

unread,

Oct 22, 2025, 5:36:26 AM (7 days ago) Oct 22

to cp2k

Dear all,

I am trying to build a CP2K version of the r2SCAN-3c composite method. As I understand it, r2SCAN-3c = r2SCAN + D4 dispersion + gCP, combined with a modified mTZVPP basis and refitted parameters for gCP and D4. CP2K already supports r2SCAN via LibXC, and provides D4 and gCP as separate corrections.

Has anyone implemented or tested an r2SCAN-3c-like setup in CP2K using these components? In particular, has anyone verified whether the original mTZVPP-based gCP and D4 parameters can be used as-is, or whether they require refitting for CP2K’s basis/pseudopotential framework?

If there are ongoing efforts or reports of results for CP2K with r2SCAN + D4 + gCP (without refitting), I would be very interested in any pointers.

Thanks in advance

Jiwoong

Reply all

Reply to author

Forward

0 new messages