Problems with SCF convergence and question regarding SCF/MIXING

[Martin Otto]

Dear CP2K developers,

i want to investigate the interaction between an oxygen atom and a Au-111 surface. I encountered problems that my SCF calculations didn't converge and i'm quite certain that my problem is charge sloshing. Some calculations did eventually converge and i noticed that the partial charge of the oxygen was larger than expected. Thus, i want to include dipole correction in my SCF calculations but i can't converge these SCF calculations at all. Any recommendations are highly appreciated.

Additionally, i have two questions about the input:

1) In FORCE_EVAL/DFT/SCF/MIXING: If i use the Broyden mixing method and i give the input a value for BETA (example input as below), is Kerker dampening/ preconditioning automatically activated? Or do i need to use the Kerker mixing mode?

Example Input:

&MIXING

      METHOD BROYDEN_MIXING

      ALPHA 0.01

      BETA 1.5

      NBUFFER 10

&END MIXING

2) Apparently, i can optimize the BETA value by checking the mean of " eigenvalues of (default mixing * dielectric matrix)". Any possibility i can print this information?