[Wien] seclr4.f

[morteza rafiee]

Dear Stefaan, Georg, Florent, and Laurence,
  Thank you for your contributions and suggestions. I would report back to the list now. I changed the eval in *.in2 to 0.1 according to Florent's suggestion as follows:
   
  TOT             (TOT,FOR,QTL,EFG,FERMI)
        -9.0      80.0 0.50 0.05                EMIN , NE , ESEPERMIN, ESEPER0
  GAUSS    0.1          (GAUSS,ROOT,TEMP,TETRA,ALL      eval)
   
  And used the in1new 3 just to check whether the jump in DIS is occurred in the 4th  cycle or not. It appears that reducing the smearing of the Gaussian method of calculating Fermi energy postpone the jump by only one cycle more. But it could not cause to be disappeared the jump on :DIS completely to reach to a stable convergence. The result is as follows:
   
  **  testerror: Error in Parallel LAPW2
  'l2main' - QTL-B.GT.15., Ghostbands, check scf files
  'l2main' - QTL-B.GT.15., Ghostbands, check scf files
   
  :DIS  :  CHARGE DISTANCE       0.1735704
  :DIS  :  CHARGE DISTANCE       0.1532537
  :DIS  :  CHARGE DISTANCE       0.8629821
  :DIS  :  CHARGE DISTANCE       0.5734159
  :DIS  :  CHARGE DISTANCE       1.5234070
  :DIS  :  CHARGE DISTANCE       1.9057531
  :DIS  :  CHARGE DISTANCE       2.2626862
   
  So at this point, I changed according to Laurence's suggestion the mixing parameter in *.inm to 0.1 and ran the scf from scratch using suggested commands:
  run_lapw –cc 0.025 –ec 0.025 –in1orig
  run_lapw –cc 0.00001 –ec 0.00001 –in1new 1 –NI –i 99
  The results are as follows: 
  
   
  :DIS  :  CHARGE DISTANCE       0.1786970
  :DIS  :  CHARGE DISTANCE       0.1677096
  :DIS  :  CHARGE DISTANCE       0.0550953
  :DIS  :  CHARGE DISTANCE       0.0383360
  :DIS  :  CHARGE DISTANCE       0.0243402
  :DIS  :  CHARGE DISTANCE       0.0202661
  :DIS  :  CHARGE DISTANCE       0.0165780
  :DIS  :  CHARGE DISTANCE       0.0123664
  :DIS  :  CHARGE DISTANCE       0.0086933
  :DIS  :  CHARGE DISTANCE       0.0052818
  :DIS  :  CHARGE DISTANCE       0.0043145
  :DIS  :  CHARGE DISTANCE       0.0028704
  :DIS  :  CHARGE DISTANCE       0.0011820
  :DIS  :  CHARGE DISTANCE       0.0008102
  :DIS  :  CHARGE DISTANCE       0.0004644
  :DIS  :  CHARGE DISTANCE       0.0003812
  :DIS  :  CHARGE DISTANCE       0.0001443
  :DIS  :  CHARGE DISTANCE       0.0001032
  :DIS  :  CHARGE DISTANCE       0.0000458
  :DIS  :  CHARGE DISTANCE       0.0000407
  :DIS  :  CHARGE DISTANCE       0.0000254
  :DIS  :  CHARGE DISTANCE       0.0000129
  :DIS  :  CHARGE DISTANCE       0.0000124
  :DIS  :  CHARGE DISTANCE       0.0000087
  :DIS  :  CHARGE DISTANCE       0.0000053
   
  So it looks that the Laurence’s an Georg's suggestions could successfully solve the problem. 
   
  Our calculation to check the Stefaan’s suggestion is still running, as we have increased the number of k-points from 16x16x1 to 25x25x2. We will report the result as soon as our calculations are finished.
   
  Thank you again,
  Your,
  Moretza Rafiee.

 
---------------------------------
Everyone is raving about the all-new Yahoo! Mail beta.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20061125/a1177a92/attachment.html