[QE-users] puzzle on homo higher than lumo

[Paolo Giannozzi]

If you set occupancies to "fixed", the code assumes that the system is 
an insulator and computes the HOMO as the highest N-th Kohn-Sham state, 
the LUMO as the lowest (N+1)-th KS state (N=half the number of 
electrons). If the system does not want to be an insulator, the "HOMO" 
might turn out to have higher energy than the "LUMO".

Paolo

On 09/03/2024 03:38, wangzongyi via users wrote:
> Dear all
> I am trying to calculate metallic oxide(CoO). However, I have some 
> puzzle in writing vcrelax.in file.
> As the experience shows that CoO is insulator, my vcrelax.in file is 
> showed in the following lines, I calculate it as insulator
> * &control*
> *    calculation='vc-relax'*
> *    restart_mode='from_scratch',*
> *    prefix='CoO'*
> *    pseudo_dir = '../../files/pseudo'*
> *    outdir='./tmp'*
> *    verbosity='high'*
> *    etot_conv_thr = 1.0D-4*
> *    forc_conv_thr = 1.0D-3*
> * /*
> * &system*
> *    ibrav = 0,*
> *    celldm(1) = 8.00,*
> *    nat = 4,*
> *    ntyp = 3,*
> *    ecutwfc = 50.0*
> *    ecutrho = 400.0*
> *    nspin = 2*
> *    tot_magnetization=0.00*
> *    occupations = 'fixed',*
> *    nbnd = 28*
> * /*
> * &electrons*
> *    conv_thr =  1.d-10*
> *    startingwfc ='file'*
> *    startingpot ='file'*
> * /*
> * &ions*
> *    ion_dynamics = 'bfgs'*
> * /*
> * &cell*
> *    cell_dynamics = 'bfgs'*
> * /*
> *ATOMIC_SPECIES*
> * Co1  58.933194  co_pbesol_v1.2.uspp.F.UPF*
> * Co2  58.933194  co_pbesol_v1.2.uspp.F.UPF*
> * O    15.999     O.pbesol-n-kjpaw_psl.0.1.UPF*
> *ATOMIC_POSITIONS {crystal}*
> *Co1      0.00000000     0.00000000     0.00000000*
> *Co2      0.50000000     0.50000000     0.50000000*
> *O        0.25000000     0.25000000     0.25000000*
> *O        0.75000000     0.75000000     0.75000000*
> *CELL_PARAMETERS {alat}*
> *   0.570726115   0.570726115   1.031099100*
> *   0.570726115   1.031099100   0.570726115*
> *   1.031099100   0.570726115   0.570726115*
> *K_POINTS {automatic}*
> * 6 6 6 0 0 0*
> *HUBBARD {ortho-atomic}*
> *U Co1-3d  6.7553*
> *U Co2-3d  6.7553*
> however, the outfile shows that
> *highest occupied, lowest unoccupied level (ev):    14.3247   13.7975*
> I am puzzled, what happened, why is homo higher than lumo? is my 
> calculation correect? Or should I regard the system as matel when 
> calculate varelax?
> Thank you very much
-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 206, 33100 Udine Italy, +39-0432-558216