[Pw_forum] problem about calculation with ph.x

[lzh-ming]

Dear Prof. Eyvaz Isaev, 
  Thanks for your answering and instruction.
 
  I haved downloaded examples_disp.tgz through the site www.quantum-espresso.org and learn how to calculate phonon dispersion curves with that routine. But since I cann't run the programs in a parallel style,it will be time-consuming calculating with that way.(But I can submit many jobs. I cann't compile the source in a parallel style.)
 
  So I wonder if I can calculate the  F(q) for each required q-points then get the files dyn1:10 together, and run q2r.x, matdyn.x to get the phonon dispersions.
  Actually I have runned in this way with the ph.in files setted such as
    ldisp=.false.
   /
    0.0 0.0 0.0 (other points)
 
and I got the dyn files, the CdCNi3-444.fc file and the CdCNi3-444.freq file in the right format. But there are negative frequencies in them. 
 
So I wonder if there are some underlying settings different between ldisp=.true. and ldisp=.false. , so I tried ph.in files in the former one. And I get the reported problem.
 
Right yesterday I learned from the site www.quantum-espresso.org(F.A.Q.) that negative frequencies may be resulted from a non-converged groundstate (Ecut, #k, ...). Maybe I set the ecut and k-grid too small. But the reported problem still be. So is my settings wrong or there is a flaw in the ph.x ?
 
Best regards,
 
li-ming-fan 
 
 
 


在2008-12-29，"Eyvaz Isaev" <eyvaz_isaev at yahoo.com> 写道：
>Dear li-ming-fan,
>
>In order to calculate phonon dispersion curves you need to calculate real space Interatomic Force Constants matrix F(R). This can be obtained via the FFT using the dynamical matricies F(q). 
>
>In order to get F(q) set, you should use only nq1,nq2,nq3, but not iq1,iq2,iq3 which are used to calculate F(q) for a given q-point(s).
>
>After F(q)=> F(R) transformation you have to supply a list of q-points along high symmetry directions and calculate phonon modes for these points.
>
>Bests,
>Eyvaz.
>
>
>-------------------------------------------------------------------
>Prof. Eyvaz Isaev, 
>Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, 
>Department of Physics, Chemistry, and Biology (IFM), Linkoping University, Sweden 
>Condensed Matter Theory Group, Uppsala University, Sweden 
>Eyvaz.Isaev at fysik.uu.se, isaev at ifm.liu.se, eyvaz_isaev at yahoo.com
>
>
>--- On Mon, 12/29/08, lzh-ming <lzh-ming at 163.com> wrote:
>
>> From: lzh-ming <lzh-ming at 163.com>
>> Subject: [Pw_forum] problem about calculation with ph.x
>> To: "Pw_forum" <pw_forum at pwscf.org>
>> Date: Monday, December 29, 2008, 2:46 PM
>> Hello everyone,
>>   I have tried to calculate the phonon disperation of
>> CdCNi3.
>> I wrote in the file CdCNi3.ph.in as follow:
>>   
>>   ldisp=.true.,
>>   nq1=4, nq2=4, nq3=4,
>>   iq1=2, iq2=1, iq3=1,
>>  
>> but in the file CdCNi3.dyn0 , I got 
>>   4 4 4
>>   1
>>   0.0 0.0 0.0 
>>  
>> I changed the value of iq1,iq2,iq3 to 2,2,1, I still got
>> 0.0 0.0 0.0.
>> I set the value as 3 3 2, I got 0.25 0.25 0.25.
>> When the value is 3 3 3, I got -0.5 -0.5 -0.5.(This is
>> right.)
>>  
>> I tried all the 10 unequivalent points of 4*4*4 grid(simple
>> cube),
>> I only got 3 different outputs.
>>  
>> What's the problem? Can you help me?
>>  
>> Thanks a lot!
>>  
>> li-ming-fan
>>  
>> 
>> 
>>  _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>
>
>      
>_______________________________________________
>Pw_forum mailing list
>Pw_forum at pwscf.org
>http://www.democritos.it/mailman/listinfo/pw_forum
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20081230/1bab7f5a/attachment.html>