Dear QE Users, I am attempting to perform convergence tests on an 001 slab of TiO2 (anatase) (1x1 with 1 layer of TiO2 in the c direction). While I am not getting any errors when attempting a ‘relax’ calculation, I am finding the the slab never really reaches a global minimum. I have decreased my convergence threshold and implemented Gaussian smearing but am still not reaching convergence. I have a ~ 15 angstrom vacuum layer implemented in the CELL_PARAMETERS block. Can someone who has more experience with surface slabs assist? This will eventually be used with an adsorbate for further studies.
I am including my input file: &CONTROL calculation = 'relax', restart_mode = 'from_scratch', prefix='ANA', pseudo_dir='/work/04703/wjid/Pseudo', tstress=.true., tprnfor=.true., / &SYSTEM ibrav=0, nat=12, ntyp=2, ecutwfc=60, ecutrho=500, occupations='smearing', smearing='gaussian', degauss=0.01 / &ELECTRONS conv_thr = 1.0D-6, mixing_mode = 'local-TF', mixing_beta = 0.3d0, diagonalization = 'cg', / &IONS ion_dynamics='bfgs', / &CELL cell_dynamics = 'bfgs', press=0.1d0, / ATOMIC_SPECIES Ti 47.88d0 Ti_PBE.UPF O 16.00d0 O_PBE.UPF CELL_PARAMETERS 7.18366994d0 0.0d0 0.0d0 0.0d0 7.18366994d0 0.0d0 0.0d0 0.0d0 41.94037d0 ATOMIC_POSITIONS crystal Ti 0.40773000000000 0.40773000000000 0.14705900000000 Ti 0.59227000000000 0.59227000000000 0.61764700000000 Ti 0.40773000000000 0.59227000000000 0.38235300000000 Ti 0.59227000000000 0.40773000000000 0.85294100000000 O 0.40773000000000 0.40773000000000 0.33529400000000 O 0.59227000000000 0.59227000000000 0.80588200000000 O 0.40773000000000 0.59227000000000 0.57058800000000 O 0.59227000000000 0.40773000000000 0.10000000000000 O 0.59227000000000 0.40773000000000 0.66470600000000 O 0.40773000000000 0.59227000000000 0.19411800000000 O 0.59227000000000 0.59227000000000 0.42941200000000 O 0.40773000000000 0.40773000000000 0.90000000000000 K_POINTS automatic 4 4 1 0 0 0 Thanks, Will DeBenedetti Doctoral Candidate Cornell University
_______________________________________________ Pw_forum mailing list [email protected] http://pwscf.org/mailman/listinfo/pw_forum
