SCF not converging in CaF₂ slab geometry optimisation — advice needed

[hamid zabihi]

Hi everyone,
I am trying to optimise a CaF₂ slab with the bottom layer fixed, but the SCF is not converging. It reaches 250 inner iterations and stops with:

*** WARNING in qs_scf.F:576 :: SCF run NOT converged ***
Example of the last SCF steps: 

243–250 Broy./Diag. ... Total energy ~ -9152.002

Charge densities look fine (total ~ −4.7E-06), but the SCF oscillates and never stabilises.

I’ve attached my input file, log, and the slab .xyz coordinates.

I’d appreciate any suggestions on:

• SCF stabilisation (mixing, DIIS, level shifting, smearing, SCF_GUESS, extrapolation)

• Whether fixing the bottom layer may destabilise SCF

• Better basis/pseudopotential choices for Ca/F slabs

• Recommended cutoffs or other practical tricks for slab systems

Thanks for any guidance!

[Marcella Iannuzzi]

Hi 

The convergence is slow but not too bad 

Increasing the number of MAX_SCF should help.

It is not a good idea to let the geometry optimisation go on with not converged SCF, the forces might be very wrong. 

Regards

Marcella