I would give a shot to local-TF, mixing-beta not too small (~ 0.3).

is it a metal ? does increasing degauss help ?
is this an instability related to the occupation of the U-projectors ?
does it converge if you if you remove U ?
does the fcc cell with 10 atoms converge ?

stefano

On 09/10/2015 09:03, 庞瑞(PANG Rui) wrote:
Dear Zhao
The most efficient way is reducing the mixing_beta, you may try 0.1 -0.01 or even smaller if you can afford the CPU time. Is 50 enough for O? You may check it, the smallest value usually is given in the PP file. A personal experience is that do not use semi-core PP. I do not understand the reason but I seldom get convergence when using semi-core PPs in periodical cases with PWSCF.
Good luck




------------------


        庞瑞 (PANG Rui)

South University of Science and Technology of China/Department of Physics

No.1088,Xueyuan Road, Shenzhen,Guangdong

------------------ Original ------------------
*From: * "Shuai"<zhao-sh...@edu.life.kyutech.ac.jp>;
*Date: * Fri, Oct 9, 2015 12:00 PM
*To: * "PWSCF Forum"<pw_forum@pwscf.org>;
*Subject: * [Pw_forum] Convergence NOT achieved for Sr2MnMoO6 using PAW pp
Dear User:

I am doing the SCF calculation of Sr2MnMoO6 using PAW Potentials but it
cannot converged.

I checked different Ecut (30Ry, 40Ry or 50Ry), mixing mode (TF or
plain), diagonalization (david or cg) and mixing_beta (0.3, 0.7), but it
always not converged.

The SCF input is:

  &control
   calculation   = 'scf'
   prefix           = 'Sr2MnMoO6'
   restart_mode  = 'from_scratch'
   pseudo_dir    = '/'
   outdir           = '/'
  /
  &system
   ibrav            = 1
   celldm(1)     = 15.27295
   nat               = 40
   ntyp             = 4
   ecutwfc        = 50
   ecutrho        = 600
   occupations   = 'smearing'
   smearing      = 'mp'
   degauss        = 0.01
   lda_plus_u     = .true.
   Hubbard_U(2)  = 4.0
  /
  &electrons
   conv_thr         = 1e-8
   mixing_mode   = 'TF'
   mixing_beta   = 0.7
   diagonalization = 'david'
  /
ATOMIC_SPECIES
   Sr    87.620   Sr.pbe-spn-kjpaw_psl.1.0.0.UPF
   Mn   54.938   Mn.pbe-spn-kjpaw_psl.1.0.0.UPF
   Mo   95.960   Mo.pbe-spn-kjpaw_psl.1.0.0.UPF
   O     15.999   O.pbe-n-kjpaw_psl.1.0.0.UPF
ATOMIC_POSITIONS (crystal)
   Sr    0.0000   0.0000   0.0000
   Sr    0.5000   0.0000   0.0000
   Sr    0.0000   0.5000   0.0000
   Sr    0.5000   0.5000   0.0000
   Sr    0.0000   0.0000   0.5000
   Sr    0.5000   0.0000   0.5000
   Sr    0.0000   0.5000   0.5000
   Sr    0.5000   0.5000   0.5000
   Mn   0.2500   0.2500   0.2500
   Mo   0.7500   0.2500   0.2500
   Mo   0.2500   0.7500   0.2500
   Mn   0.7500   0.7500   0.2500
   Mo   0.2500   0.2500   0.7500
   Mn   0.7500   0.2500   0.7500
   Mn   0.2500   0.7500   0.7500
   Mo   0.7500   0.7500   0.7500
   O     0.2500  -0.0140   0.2500
   O    -0.0140   0.2500   0.2500
   O     0.2500   0.2500  -0.0140
   O     0.7500   0.0140   0.2500
   O     0.5140   0.2500   0.2500
   O     0.7500   0.2500   0.0140
   O     0.2500   0.5140   0.2500
   O     0.0140   0.7500   0.2500
   O     0.2500   0.7500   0.0140
   O     0.7500   0.4860   0.2500
   O     0.4860   0.7500   0.2500
   O     0.7500   0.7500  -0.0140
   O     0.2500   0.0140   0.7500
   O     0.0140   0.2500   0.7500
   O     0.2500   0.2500   0.5140
   O     0.7500  -0.0140   0.7500
   O     0.4860   0.2500   0.7500
   O     0.7500   0.2500   0.4860
   O     0.2500   0.4860   0.7500
   O    -0.0140   0.7500   0.7500
   O     0.2500   0.7500   0.4860
   O     0.7500   0.5140   0.7500
   O     0.5140   0.7500   0.7500
   O     0.7500   0.7500   0.5140
K_POINTS (automatic)
   4 4 4 1 1 1

Thanks in advance for your suggestions.

Best regards,

Shuai Zhao

--
PhD candidate
Graduate School of Life Science and Systems Engineering
Kyushu Institute of Technology, Japan

_______________________________________________
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum


_______________________________________________
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

_______________________________________________
Pw_forum mailing list
Pw_forum@pwscf.org
http://pwscf.org/mailman/listinfo/pw_forum

Reply via email to