[QE-users] SCF not converging in QE6.3 (but it does so in QE6.0)

[Guido Fratesi]

Dear all,

The problem I'm facing here is the missing SCF convergence with the 6.3 
and develop releases, while older 6.0 release convergences smoothly. I 
am computing a three-layer Au(111) slab, that I replicated from the 
(1x1) relaxation to a (4x3) supercell (and also other sizes). To my 
experience this system should not pose specific difficulties.

In detail:


* I'm running (HPC system marconi at cineca, Italy) the precompiled version 
of QE6.3, and the SCF loop is not converging, regardless of changing 
smearing, mixing mode & parameter, pseudopotential. The input is given 
at the end of this message. E.g., "grep accuracy 
Au111_4x3_3L_K3x4.rlx.out" gives:
      estimated scf accuracy    <       1.89873147 Ry
      estimated scf accuracy    <       0.12553445 Ry
      estimated scf accuracy    <       0.01337648 Ry
      estimated scf accuracy    <       0.00966317 Ry
      estimated scf accuracy    <       0.01041836 Ry
      estimated scf accuracy    <       0.00627208 Ry
      estimated scf accuracy    <       0.01265435 Ry
      estimated scf accuracy    <       0.00686049 Ry
      estimated scf accuracy    <       0.02148796 Ry
      estimated scf accuracy    <       0.00703012 Ry
      estimated scf accuracy    <       0.00905125 Ry
      estimated scf accuracy    <       0.01597448 Ry
      estimated scf accuracy    <       0.00370159 Ry
      estimated scf accuracy    <       0.00418866 Ry
and so on without reducing further.


* As a comparison, I run older version of QE (QE6.0 also build by cineca 
staff on marconi) with the same input, and convergence is smooth and 
quick as expected:

grep accuracy qe6.0.rlx.out
      estimated scf accuracy    <       1.93398261 Ry
      estimated scf accuracy    <       0.15421538 Ry
      estimated scf accuracy    <       0.00644341 Ry
      estimated scf accuracy    <       0.00382849 Ry
      estimated scf accuracy    <       0.00265796 Ry
      estimated scf accuracy    <       0.00056037 Ry
      estimated scf accuracy    <       0.00022539 Ry
      estimated scf accuracy    <       0.00001359 Ry
      estimated scf accuracy    <       0.00000159 Ry
      estimated scf accuracy    <       0.00000009 Ry
      estimated scf accuracy    <       0.00000415 Ry
      estimated scf accuracy    <       0.00000089 Ry
      estimated scf accuracy    <       0.00000004 Ry


* I downloaded today the develop version from 
https://github.com/QEF/q-e/archive/develop.zip recompiling on the same 
machine and still the SCF does not converge.


* Important to say, if I place a molecule on one side of the slab, the 
SCF converges (also with QE6.3).


Here is my input (all files are available in this online folder: 
https://drive.google.com/drive/folders/1072PbthHTvbqNTCXqHKeAJUyPOwi_2Ez)

  &control
     calculation = 'relax' ,
     restart_mode = 'from_scratch' ,
     prefix='Au111_5x4_3L' ,
     pseudo_dir = '../0-pseudo.d/' ,
     outdir='./tmp/' ,
     max_seconds   = 1200
     iprint = 1
     verbosity = 'high'
  /
  &system
     ibrav=  12 ,
     a = 11.8022910005134168d0 ,
     b =  8.8517182503850626d0 ,
     c = 20.0d0 ,
     cosab = -.5 ,
     nat =  36 ,
     ntyp = 1 ,
     ecutwfc = 36 ,
     ecutrho = 220 ,
     occupations = 'smearing' ,
     degauss = 0.02 ,
     smearing = 'mv' ,
  /
  &electrons
     mixing_mode = 'local-TF'
     conv_thr = 1e-7
  /
  &ions
  /
ATOMIC_SPECIES
  Au  0.0 Au.pbe-dn-rrkjus_psl.0.1.UPF
ATOMIC_POSITIONS Angstrom
Au      1.475286375063      0.851756985775     -4.818265124511
Au      4.425859125193      0.851756985775     -4.818265124511
Au      7.376431875324      0.851756985775     -4.818265124511
Au     10.327004625454      0.851756985775     -4.818265124511
Au      0.000000000000      3.407027943102     -4.818265124511
Au      2.950572750130      3.407027943102     -4.818265124511
Au      5.901145500260      3.407027943102     -4.818265124511
Au      8.851718250391      3.407027943102     -4.818265124511
Au     -1.475286375063      5.962298900428     -4.818265124511
Au      1.475286375067      5.962298900428     -4.818265124511
Au      4.425859125197      5.962298900428     -4.818265124511
Au      7.376431875328      5.962298900428     -4.818265124511
Au      0.000000000000      1.703513971551     -2.409132562256
Au      2.950572750130      1.703513971551     -2.409132562256
Au      5.901145500260      1.703513971551     -2.409132562256
Au      8.851718250391      1.703513971551     -2.409132562256
Au     -1.475286375063      4.258784928877     -2.409132562256
Au      1.475286375067      4.258784928877     -2.409132562256
Au      4.425859125197      4.258784928877     -2.409132562256
Au      7.376431875328      4.258784928877     -2.409132562256
Au     -2.950572750126      6.814055886203     -2.409132562256
Au      0.000000000004      6.814055886203     -2.409132562256
Au      2.950572750134      6.814055886203     -2.409132562256
Au      5.901145500265      6.814055886203     -2.409132562256
Au      0.000000000000      0.000000000000      0.013283696000
Au      2.950572750130      0.000000000000      0.013283696000
Au      5.901145500260      0.000000000000      0.013283696000
Au      8.851718250391      0.000000000000      0.013283696000
Au     -1.475286375063      2.555270957326      0.013283696000
Au      1.475286375067      2.555270957326      0.013283696000
Au      4.425859125197      2.555270957326      0.013283696000
Au      7.376431875328      2.555270957326      0.013283696000
Au     -2.950572750126      5.110541914652      0.013283696000
Au      0.000000000004      5.110541914652      0.013283696000
Au      2.950572750134      5.110541914652      0.013283696000
Au      5.901145500265      5.110541914652      0.013283696000
K_POINTS automatic
  3 4 1  1 1 0

Thank you for your kind attention,

Guido Fratesi



-- 
Guido Fratesi

Dipartimento di Fisica
Universita` degli Studi di Milano
Via Celoria 16, 20133 Milano, Italy