[Pw_forum] strange results for phonon frequencies.

[Elie M]

Dear all,
Sorry to ask so many questions about computing phonon frequencies but recently i have got some strange results for one of the systems I am working on. My work involves computing phonon frequencies for a 42 carbon at the Gamma point so in total 126 irreducible representations. in a previous calculations, my input was:
SCF FILE         

 /

 &system   


    ibrav= 0, celldm(1) =1.88972599, nat=42,
ntyp= 1, ecutwfc =32, ecutrho=128, occupations='smearing', smearing='mp',degauss=0.01

 

/

 &electrons

    conv_thr=1.D-6,     

    mixing_beta=0.7D0,

    diago_david_ndim=2,
   

ATOMIC_SPECIES

 C 
12.0107  C.pbe-rrkjus.UPF

 ...
PH FILE:
Phonon
calculation for GB60 at gamma point

 &inputph

  tr2_ph=1.0d-15,

  prefix='GrBo42ph',

  fildvscf='GB60dvscf',

  elph=.true.,.....
The results obtained were:
omega(
1) =      -4.450324 [THz] =    -148.446823 [cm-1]

     omega( 2) =      -3.144068 [THz] =    -104.874835 [cm-1]

     omega( 3) =       2.963314 [THz] =      98.845504 [cm-1]

     omega( 4) =       3.454797 [THz] =     115.239637 [cm-1]

     omega( 5) =       3.908880 [THz] =     130.386191 [cm-1]...     omega(**) =      47.380063 [THz] =    1580.428796 [cm-1]

     omega(**) =      47.573318 [THz] =    1586.875087 [cm-1]

     omega(**) =      47.821300 [THz] =    1595.146853 [cm-1]

     omega(**) =      48.119632 [THz] =    1605.098137 [cm-1]
Where the 2 negative frequencies,as professor Gionazzi repeated several times, could be due to the violation of the ASR rule.
For my own piece of mind I repeated the calculation with a better accuracy where ecutrho=256 (8 times ecutwfc which should be the case if an ultrasoft potential is used)and tr2_ph =10^(-17) although mixing_beta was set equal to 0.3 this time. I expected better results but what I got was the following:
omega( 
1 -  1) =  -77614121.2 
[cm-1]   --> A'              I+R

    
omega(  2 -  2) = 
-13157144.1  [cm-1]   --> A'              I+R

    
omega(  3 -  3) =  
-5577066.3  [cm-1]   --> A'              I+R

    
omega(  4 -  4) =  
-4332492.8  [cm-1]   --> A'              I+R

    
omega(  5 -  5) =  
-4094962.7  [cm-1]   --> A'              I+R

    
omega(  6 -  6) =  
-3368158.6  [cm-1]   --> A'              I+R

    
omega(  7 -  7) =  
-2781643.0  [cm-1]   --> A'              I+R

    
omega(  8 -  8) =  
-2403295.2  [cm-1]   --> A'              I+R

    
omega(  9 -  9) =  
-1777809.3  [cm-1]   --> A'              I+R

    
omega( 10 - 10) =        
82.5  [cm-1]   --> A''             I+R

    
omega( 11 - 11) =        
85.8  [cm-1]   --> A''             I+R...     omega(120 -120) =   
2301243.8  [cm-1]   --> A'              I+R

    
omega(121 -121) =   
2494732.0  [cm-1]   --> A'              I+R

    
omega(122 -122) =   
2788571.0  [cm-1]   --> A'              I+R

    
omega(123 -123) =   
3135326.3  [cm-1]   --> A'              I+R

    
omega(124 -124) =   
3630869.3  [cm-1]   --> A'              I+R

    
omega(125 -125) =   
8808450.4  [cm-1]   --> A'              I+R
What is the reason for those LARGE negative values at the beginning and the LARGE ones at the end? Is it because mixing_beta was changed?
Thanks for the help
Elie KoukaesUniversity of NottinghamNG7 2RDUK

 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20120902/ea1e7fd9/attachment.htm