[Pw_forum] question about the example2

[Shaorui Sun]

Dear all:
     The example2 is about the phonon mode computation of Gamma and X points for the silicon. In the example, the k-points
were set as:
 K_POINTS
  10
   0.1250000  0.1250000  0.1250000   1.00
   0.1250000  0.1250000  0.3750000   3.00
   0.1250000  0.1250000  0.6250000   3.00
   0.1250000  0.1250000  0.8750000   3.00
   0.1250000  0.3750000  0.3750000   3.00
   0.1250000  0.3750000  0.6250000   6.00
   0.1250000  0.3750000  0.8750000   6.00
   0.1250000  0.6250000  0.6250000   3.00
   0.3750000  0.3750000  0.3750000   1.00
   0.3750000  0.3750000  0.6250000   3.00
-----------------------------------------------------
the results are reasonable agree with the experimental results. But when I use those K-points for the K(0.75,0.75,0) and L(0.5,0.5,0.5) points, the results are not good (L(TA)) which are listed in the 2nd column of the table . 

    I also use k-points automatic generated with :
K_POINTS automatic
  8 8 8   1 1 1
the wavelengths of the phonon modes for the four points (listed in the 3rd column of the table)are all different from the experimental results, especially for the low frequency modes, such as the X(TA) and L(TA).


***********************************************************************************************
------------------------------------------------------------------------------------------
             experimental(cm-1)      special points        automatic generation
                                  computational(cm-1)     computational(cm-1)
Gamma(TO)      517                    510                     501
Gamma(LO)      517                    510                     501
X(TA)          150                    140                     212
X(LA)          410                    408                     427
X(TO)          463                    458                     468
X(LO)          410                    408                     427
L(TA)          114                    190                     190
L(LA)          378                    402                     398
L(TO)          487                    509                     498
L(LO)          417                    437                     424
---------------------------------------------------------------------------------------
************************************************************************************************
the experimental result get from the paper: P.R.B. 43, 7231 Paolo Giannozzi,Stefano de Gironcoli, Pasquale Pavone and Stefano Baroni 


the pw input file for the X point with k-poins generated automatically is shown as follows:
  &control
    calculation='phonon'
    prefix  = 'Si-X'
    pseudo_dir = '/home/wien2k/pw/pseudo/',
    outdir='/biology/wien2k/pwwork/Si/v0/X/'
 /
 &system    
    ibrav=  2, celldm(1) =10.263107, nat=  2, ntyp= 1,
    ecutwfc =18.0, 
 /
 &electrons
    conv_thr =  1.0d-8
 /
 &phonon
    xqq(1)=1.0, xqq(2)=0.0, xqq(3)=0.0
 /
ATOMIC_SPECIES
 Si  28.086  Si.vbc.UPF
ATOMIC_POSITIONS
 Si 0.00 0.00 0.00 
 Si 0.25 0.25 0.25 
K_POINTS automatic
  8 8 8   1 1 1
-----------------------------------------

while the ph files for the X point is shown as follows :
 &inputph
  tr2_ph=1.0d-14,
  prefix='Si-X',
  amass(1)=28.08,
  outdir='/biology/wien2k/pwwork/Si/v0/X/',
  fildyn='Si-X.dynX',
 /
1.0 0.0 0.0
------------------------------------------------------
Why the k-points generated with automatic mode  could not get the right results, or Is there something wrong in my input files? Is there deeply difference between the two ways of setting k-points? Must I produce different k-points for different points in reciprocal space by hand, and how to do that for the K and L points?
Any direction,comments or examples are appreciated!    


 
 




Best Regards!
                        S. R. Sun
---------------------------------------------
Beijing Synchrotron Radiation Facility
Institute of High Energy Physics
Chinese Academy of Sciences
P. O. Box 918, 100049 Beijing
P. R. China
Tel: 0086+10 88236710
email:sunsr at ihep.ac.cn