[QE-users] Seeking help for SCF calculations on possible semiconductor (Based on several error reports)

[zhangkan3 at mail.sysu.edu.cn]

Dear experts,

I am currently doing an scf calculation and have encountered several blocks here. Here is the input script:

&system
    ibrav           =   4
    celldm(1)       =  15.227846653
celldm(2)       =  1
celldm(3)       =  0.4640435926996375
celldm(4)       =  120.000
celldm(5)       =  90.000
celldm(6)       =  90.000
    nat             =   3
    ntyp            =   2
ecutwfc = 33.07
nspin = 2
tot_magnetization=0.4285714
/
 &electrons
    conv_thr        =   1.D-12
    diagonalization =  'cg'
/
ATOMIC_SPECIES
C  12.01  C_ONCV_PBE_fr.upf
N  14.01  N_ONCV_PBE_fr.upf
ATOMIC_POSITIONS {crystal}
C  0.0000000000000000    0.0000000000000000    0.9974228266286470
C  0.0000000000000000    0.0000000000000000    0.3359239109776594
N  0.0000000000000000    0.0000000000000000    0.6666531993936927
K_POINTS {automatic}
10 10 18 0 0 0




Firstly I was blocked by forgetting the system is spin-polarized and adding nspin=2; And then got stuck here:fixed occupations and lsda need tot_magnetization, here the point: I have difficulties on how to specify the magnetization related variables. Why these variables must be specified?  I got the first error in warning to add smearing, but it supposed that this system should be a semiconductor. And then, based on the first point, I have tried to execute a vasp calculation and extract the magnetization=0.4285714 from OUTCAR (I am not sure if this set is suitable). The after error report is: Error in routine card_kpoints (1):error while reading automatic k points.

I am afraid that this will be a dead cycle. So I am here to seeking some suggestions, sincerely.

Kan Zhang

from Sun Yat-sen University,China
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200415/c39acd0f/attachment.html>