[CP2K-user] How to get robust convergence performance?

[Ke Zhou]

Dear All, 



Now I am simulating the strained (about 0.15, close the fracture strain) 
single-layer graphene. 

But it cannot get well convergence after running for about ~100 or ~ 1000 
steps.

I used OT method (with DIIS or CG minimizer) or direct diagonalization 
method. But both methods are not working.

Can anyone give me some suggestions?

Best,
Justin



##############
####### for OT #######
      CHOLESKY OFF
      &OT
        PRECONDITIONER FULL_ALL
        ! the most robust choice (DIIS might sometimes be faster, but not 
as stable).
        MINIMIZER DIIS ## CG
        N_DIIS 5
       ENERGY_GAP 1.0E-5
      &END OT



###### for diagonalization #######
      ADDED_MOS 100
      CHOLESKY OFF #INVERSE
      &SMEAR ON
        METHOD FERMI_DIRAC
        ELECTRONIC_TEMPERATURE [K] 300
      &END SMEAR
      &DIAGONALIZATION
        ALGORITHM STANDARD
        EPS_ADAPT 0.01
      &END DIAGONALIZATION

....
      &XC_GRID
             XC_SMOOTH_RHO NN50
             XC_DERIV NN50_SMOOTH
      &END

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20200718/69b0cee9/attachment.htm>