[Pw_forum] slow convergency in non-collinear calculations

Javad hashemifar jhashemifar at gmail.com
Tue Mar 20 10:20:09 CET 2007


Dear pwscf users

I am trying to calculate the non-collinear helical magnetic ground
state of the orthorhombic CrAs. The system has "No symmetry!" and I am
using ecut=30, ecutrho=300, smearing="mp",  degauss=0.05D0,
mixing_beta=0.1, mixing_mode="local-TF", please find the input file
below. Unfortunately the system does not converge and "estimated scf
accuracy" is oscillating around ~ 0.01. I would appreciate your
comments and suggestions in this regard.

Thanks in advance
Javad Hashemifar

&CONTROL
  calculation = "scf",
  prefix      = "csMP-AFh",
  restart_mode= "from_scratch",
  tstress     = .true.,
  tprnfor     = .true.,
  pseudo_dir  = "/homes/hp0070/psps",
  outdir      = "/tmp/javad/results-pwscf/CrAs-MP/explat",
/
&SYSTEM
  ibrav       = 8,
  celldm(1)   = 10.849,
  celldm(2)   = 0.6072449074,
  celldm(3)   = 3.2514148769,
  nat         = 24,
  ntyp        = 7,
  ecutwfc     = 30.0,
  ecutrho     = 400.0,
  occupations = "smearing",
  smearing    = "mp",
  degauss     = 0.05D0,
  noncolin    = .true.
  starting_magnetization(1)=0.7,
  starting_magnetization(2)=0.7,
  starting_magnetization(3)=0.7,
  starting_magnetization(4)=0.7,
  starting_magnetization(5)=0.7,
  starting_magnetization(6)=0.7,
  angle1(1)=90.0, angle1(2)=90.0, angle1(3)=90.0
  angle1(4)=90.0, angle1(5)=90.0, angle1(6)=90.0
  angle2(1)=0.0,   angle2(3)=120.0, angle2(5)=240.0
  angle2(2)=100.0, angle2(4)=220.0, angle2(6)=340.0
/
&ELECTRONS
  electron_maxstep = 200,
  mixing_beta = 0.1,
  mixing_mode = "local-TF",
  conv_thr    = 1.0d-8,
/
ATOMIC_SPECIES
  Cr1  51.9960   Cr.pbe-sp-van.UPF
  Cr2  51.9960   Cr.pbe-sp-van.UPF
  Cr3  51.9960   Cr.pbe-sp-van.UPF
  Cr4  51.9960   Cr.pbe-sp-van.UPF
  Cr5  51.9960   Cr.pbe-sp-van.UPF
  Cr6  51.9960   Cr.pbe-sp-van.UPF
  As   74.9220   As.pbe-n-van.UPF
ATOMIC_POSITIONS  (alat)
 As    0.20000000    0.15181123    0.08670384
 Cr1   0.49000000    0.45543368    0.20592220
 Cr2   0.99000000    0.45543368    0.33598027
 As    0.70000000    0.15181123    0.45519864
 As    0.30000000    0.45543368    0.62860632
 Cr1   0.01000000    0.15181123    0.74782468
 Cr2   0.51000000    0.15181123    0.87788275
 As    0.80000000    0.45543368    0.99710112
 As    0.20000000    0.15181123    1.17050880
 Cr3   0.49000000    0.45543368    1.28972716
 Cr4   0.99000000    0.45543368    1.41978523
 As    0.70000000    0.15181123    1.53900359
 As    0.30000000    0.45543368    1.71241128
 Cr3   0.01000000    0.15181123    1.83162964
 Cr4   0.51000000    0.15181123    1.96168771
 As    0.80000000    0.45543368    2.08090607
 As    0.20000000    0.15181123    2.25431376
 Cr5   0.49000000    0.45543368    2.37353212
 Cr6   0.99000000    0.45543368    2.50359019
 As    0.70000000    0.15181123    2.62280855
 As    0.30000000    0.45543368    2.79621624
 Cr5   0.01000000    0.15181123    2.91543460
 Cr6   0.51000000    0.15181123    3.04549267
 As    0.80000000    0.45543368    3.16471103
K_POINTS (automatic)
 8  8  2    1 1 1




======================================
Seyed Javad  Hashemifar,    Ph.D.
[current:]
Tel:+49-203-3794743        Fax:+49-203-3794742
Fachbereich Physik (AG Kratzer), Universitat Duisburg-Essen
47048 Dusiburg, Germany
[permanent:]
Tel:  +98-311-3912375       Fax: +98-311-3912376
Physics Department, Isfahan University of Technology
84156 Isfahan, Iran
---------------------------------------------------------------------------



More information about the Pw_forum mailing list