SCF Convergence Issue in MOF/H⁺-g-C₃N₄ Relaxation with and without CDFT
Hamideh Kahnouji
Dear CP2K community,
I’m a new user of CP2K and I’m currently working on a geometry relaxation of a metal-organic framework (MOF) adsorbed on protonated graphitic carbon nitride (H⁺/g-C₃N₄). Although the total system is neutral, several hydrogen atoms are formally H⁺, and I’ve used CDFT to constrain the charge distribution accordingly.
However, I’m encountering persistent SCF convergence issues. I’ve tried several strategies, including:
Increasing the electronic temperature
Reducing EPS_SCF
Adjusting mixing parameters
Despite these efforts, the SCF cycle fails to converge. To test whether the issue is related to CDFT, I also ran the same system without charge constraints, but the problem persists.
I’ve attached the input and output files for reference. If anyone has experience with similar systems or suggestions for improving convergence in this context, I would be very grateful.
Thank you for your time and support.
Best regards,
Hamideh
Example log from output (after 1400 iterations, I expect SCF not to converge:
Kerker/Diag.0.10E+00 416.5 0.59360025 -2253.4968751753 -1.02E+01
Kerker/Diag.0.10E+00 1033.6 0.59379666 -2253.6680909216 -1.04E+01
Kerker/Diag.0.10E+00 1027.0 0.59379690 -2243.2296825274 1.04E+01