[Pw_forum] SCF calculations for molecules in electric fields are not converged

Jibiao Li jibiaoli at foxmail.com
Tue Aug 8 08:09:14 CEST 2017


Dear QE users and developers,


I am using QE 6.1 to perform SCF calculations of a molecule in homogeneous electric fields, but the calculations are not converged with a higher lelfield value (efield_cart(1)=1). Nevertheless, when I tuned the field strength to a lower value (efield_cart(1)=0.1), the calculations are get converged. Can anyone solve the convergence problem at higher E-field strength?


Best Regards


Jibiao Li


Yangtze Normal University


&CONTROL
                 calculation = 'scf' ,
                restart_mode = 'from_scratch' ,
                      outdir = '/home/jibiao/Calc/molecules/N2/' ,
                  pseudo_dir = '/home/jibiao/Codes/pseudo/' ,
                      prefix = 'N2' ,
                     tstress = .true. ,
                     tprnfor = .true. ,
                    lelfield = .true. ,
 /
 &SYSTEM
                       ibrav = 1,
                   celldm(1) = 15.117815022,
                         nat = 2,
                        ntyp = 1,
                     ecutwfc = 29 ,
                     ecutrho = 180 ,
                 occupations = 'fixed' ,
                     degauss = 0.05D0 ,
                    smearing = 'gaussian' ,
            exxdiv_treatment = 'gygi-baldereschi' ,
 /
 &ELECTRONS
            electron_maxstep = 300,
                 mixing_beta = 0.2D0 ,
             diagonalization = 'david' ,
              efield_cart(1) = 1,
              efield_cart(2) = 0,
              efield_cart(3) = 0,
 /
ATOMIC_SPECIES
    N   14.00700  N.pbe-van_ak.UPF 
ATOMIC_POSITIONS angstrom 
    N      5.122061326    3.999999675    4.000000376    
    N      4.000000000    4.000000000    4.000000000    0  0  0 
K_POINTS automatic 
  1 1 1   0 0 0
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170808/70919e52/attachment.html>


More information about the users mailing list