Re: [Phonopy-users] Last step of VASP_DFPT method, please help, Luke Soule

[邵和助 <hz...@ni...>]

Dear Luke

       I found that there's some problem in your file of FORCE_CONSTANTS. You have not generate the right FORCE_CONSTANTS. Have you done the right steps which the manuel provides?

1. Prepare unit cell structure named, e.g., POSCAR-unitcell.

2. Prepare a perfect supercell structure from POSCAR-unitcell, e.g.,phonopy -d --dim="2 2 2" -c POSCAR-unitcell

3. Rename SPOSCAR created in (2) to POSCAR (POSCAR-{number} and disp.yaml files will never be used.)
% mv SPOSCAR POSCAR

4. Calculate force constants of the perfect supercell by running VASP with IBRION = 8 and NSW = 1. An
example of INCAR for insulator may be such like (just an example!):

PREC = Accurate
ENCUT = 500

IBRION = 8
EDIFF = 1.0e-08
IALGO = 38
ISMEAR = 0; SIGMA = 0.1
LREAL = .FALSE.
ADDGRID = .TRUE.
LWAVE = .FALSE.
LCHARG = .FALSE. 


5 After finishing the VASP calculation, confirm vasprun.xml contains hessian elements, and then create
FORCE_CONSTANTS:
% phonopy --fc vasprun.xml.

Please notice here there is two '-'..

Here you will generate the right FORCE_CONSTANTS file.

6. Run phonopy with the original unit cell POSCAR-unitcell and setting tag FORCE_CONSTANTS = READ
or --readfc option, e.g., as found in example/NaCl-VASPdfpt
% phonopy --dim="2 2 2" -c POSCAR-unitcell band.conf

If you properly follow the above steps, you should got the right dispersion of phonon.

 

Hezhu

 

 

-----原始邮件-----
发件人: "Luke Soule" <ls...@gm...>
发送时间: 2016年1月20日 星期三
收件人: pho...@li...
抄送:
主题: Re: [Phonopy-users] Last step of VASP_DFPT method, please help, Luke Soule



Yes, that would be very helpful.

 

Thanks

 

Luke

 

 

From:邵和助 [mailto:hz...@ni...]
Sent: Monday, January 18, 2016 5:07 PM
To:pho...@li...
Subject: Re: [Phonopy-users] Last step of VASP_DFPT method, please help, Luke Soule

 

Luke

     You could attach your FORCE_CONSTANTS, band.conf, and POSCAR and I try to check these files for you.

hezhu

-----原始邮件-----
发件人: "Luke Soule" <ls...@gm...>
发送时间: 2016年1月19日 星期二
收件人:pho...@li...
抄送:
主题: Re: [Phonopy-users] Last step of VASP_DFPT method, please help, Luke Soule

I apologize, I stupidly didn’t realize I could merely alter my supercell number in my FORCE_CONSTANTS file. Another error arose though and this one I am clueless on.

 

Can someone help a beginner to phonopy? 

 

File "/usr/bin/phonopy", line 638, in <module>

    fc = file_IO.parse_FORCE_CONSTANTS(filename="FORCE_CONSTANTS")

  File "/usr/lib/python2.7/dist-packages/phonopy/file_IO.py", line 197, in parse_FORCE_CONSTANTS

    force_constants[i, j] = np.array(tensor)

ValueError: could not broadcast input array from shape (3,0) into shape (3,3)

 

Thanks, Luke

 

From:邵和助 [mailto:hz...@ni...]
Sent: Thursday, January 14, 2016 5:41 PM
To:pho...@li...
Subject: Re: [Phonopy-users] Last step of VASP_DFPT method, please help, Luke Soule

 

hi

have you run the example given in Phonopy? Maybe you could find some clues to resolve your problem.

hezhu

-----原始邮件-----
发件人: "Luke Soule" <ls...@gm...>
发送时间: 2016年1月15日 星期五
收件人:pho...@li...
抄送:
主题: [Phonopy-users] Last step of VASP_DFPT method, please help, Luke Soule

So I have followed the VASP-DFPT manual and have generated FORCE_CONSTANTS. Yet it is giving me this error when I type “phonopy POSCAR-unitcell -c –dim=”2 2 2” –readfc”

 

“The error Force constants are read from FORCE_CONSTANTS.

 

Number of atoms in supercell is not consistent with the matrix shape of

force constants read from FORCE_CONSTANTS.

Please carefully check DIM, FORCE_CONSTANTS, and POSCAR.”

 

Is there any way to fix this?

 

Also, can I have some pointers on syntax? What are the steps to develop a density of state plot from the VASP-DFPT?

                

From:邵和助 [mailto:hz...@ni...]
Sent: Wednesday, January 13, 2016 4:54 PM
To:pho...@li...
Subject: Re: [Phonopy-users] Help with generating FORCE_SETS

 

hi

I think there is something wrong with your incar file, check it.

hezhu

-----原始邮件-----
发件人: "Luke Soule" <ls...@gm...>
发送时间: 2016年1月13日 星期三
收件人:pho...@li...
抄送:
主题: Re: [Phonopy-users] Help with generating FORCE_SETS

The file does not end with those three lines, it ends with

<I type=”logical” name=”LHYPERFINE”> F </i>

</incar>

 

I’ll try to re run a vasp calculation and see if I get a different result. Thanks,

 

Luke

 

 

From:邵和助 [mailto:hz...@ni...]
Sent: Tuesday, January 12, 2016 6:02 PM
To:pho...@li...
Subject: Re: [Phonopy-users] Help with generating FORCE_SETS

 

hi

Maybe your vasprun.xml file is not completed. You could check the file, see if the final lines are like this "  </varray>
 </structure>
</modeling>"

 If so, you should rerun a vasp calculation to get right vasprun.xml.

-----原始邮件-----
发件人: "Luke Soule" <ls...@gm...>
发送时间: 2016年1月13日 星期三
收件人:pho...@li...
抄送:
主题: [Phonopy-users] Help with generating FORCE_SETS

Hello, I am having difficulty producing a FORCE_SETS file. When I type

 

“phonopy -f vasprun.xml”

 

Instead of spiting out

 

“FORCE_SETS has been created”

 

I am getting

 

File "/usr/bin/phonopy", line 514, in <module>

    log_level)

  File "/usr/lib/python2.7/dist-packages/phonopy/interface/vasp.py", line 105, in create_FORCE_CONSTANTS

    fc_and_atom_types = read_force_constant_vasprun_xml(filename)

  File "/usr/lib/python2.7/dist-packages/phonopy/interface/vasp.py", line 463, in read_force_constant_vasprun_xml

    return get_force_constants_vasprun_xml(vasprun)

  File "/usr/lib/python2.7/dist-packages/phonopy/interface/vasp.py", line 468, in get_force_constants_vasprun_xml

    for event, element in vasprun:

  File "/usr/lib/python2.7/dist-packages/phonopy/interface/vasp.py", line 441, in _iterparse

    for event, elem in etree.iterparse(fname):

  File "<string>", line 107, in next

cElementTree.ParseError: no element found: line 40, column 0

 

 

 

Can anyone help?

 

Thanks,

 

Luke Soule

New Mexico Institute of Mining and Technology