Installation about vaspbandunfolding

[sfragkos]

I found a way to install it but now i get the following error:

phi = wav.wfc_r(ikpt=2, iband=72, ngrid=wav._ngrid * 2)
Traceback (most recent call last):
File "", line 1, in
File "vaspwfc.py", line 258, in wfc_r
gvec = self.gvectors(ikpt)
File "vaspwfc.py", line 195, in gvectors
(Gvec.shape[0], self._nplws[ikpt -1], np.prod(self._ngrid))
AssertionError: No. of planewaves not consistent! 6788 13576 17595

[QijingZheng]

Hi Sotirios The error says something about the syntax, it may be that you are using python3 to run the script, which could be a problem since I was using python2 when I wrote the script. Maybe you can avoid this error by explicitly using python2, e.g. python2 xxx.py Best Qijing

…

On Sun, Apr 29, 2018 at 1:05 AM, sfragkos ***@***.***> wrote: I think i managed to install it, but know i get the following error print 'Calculating IPR of #spin %4d, #kpt %4d, #band %4d' % (ispin+1, ikpt+1, iband+1) ^ SyntaxError: invalid syntax — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIINPp7yI6m-87GZF3W57ls6SqgooZdOks5ttKFugaJpZM4TreOg> .

[sfragkos]

Thank you very much for your answer.
Your code works fine when i tried it on your example. But when i try to apply it in my case, i get the following error:

phi = wav.wfc_r(ikpt=1, iband=72, ngrid=wav._ngrid * 2)
Traceback (most recent call last):
File "", line 1, in
File "/home/dimoulas/.local/lib/python2.7/site-packages/vaspwfc.py", line 258, in wfc_r
gvec = self.gvectors(ikpt)
File "/home/dimoulas/.local/lib/python2.7/site-packages/vaspwfc.py", line 195, in gvectors
(Gvec.shape[0], self._nplws[ikpt -1], np.prod(self._ngrid))
AssertionError: No. of planewaves not consistent! 6821 13642 17595

Do you know how can i fix this? what is ngrid? and what should be my KPOINTS file format?
Thanks again!!!

[QijingZheng]

Hi Sotirios Is your WAVECAR obtained from an Gamma-only VASP? Because the error message indicates that the number of G-vectors and planewaves does not match. The ngrid determines the grid size out the FFT. Best Qijing

…

On Sun, Apr 29, 2018 at 3:26 PM, sfragkos ***@***.***> wrote: Thank you very much for your answer. Your code works fine when i tried it on your example. But when i try to apply it in my case, i get the following error: *phi = wav.wfc_r(ikpt=1, iband=72, ngrid=wav._ngrid * 2)* Traceback (most recent call last): File "", line 1, in File "/home/dimoulas/.local/lib/python2.7/site-packages/vaspwfc.py", line 258, in wfc_r gvec = self.gvectors(ikpt) File "/home/dimoulas/.local/lib/python2.7/site-packages/vaspwfc.py", line 195, in gvectors (Gvec.shape[0], self._nplws[ikpt -1], np.prod(self._ngrid)) *AssertionError: No. of planewaves not consistent! 6821 13642 17595* Do you know how can i fix this? what is ngrid? and what should be my KPOINTS file format? Thanks again!!! — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIINPoZqES7xrOWB9YMfw7eyGdRr8rPvks5ttWs3gaJpZM4TreOg> .

[sfragkos]

Hi Qijing
Because I'm new in DFT and i dont understand much yet, I attach you my INCAR KPOINTS and POSCAR files. I used 2 different types of KPOINTS file as you will see. What their format should be?
I tried to turn off SOC, and this error didnt appear this time, but when i used wav.save2vesta(phi, poscar='/my_dir_to_/POSCAR') the wfc_i.vasp wfc_r.vasp file didn't appear (but it work for your example).

Thanks for your patient!

[sfragkos]

system = MoTe2
ISTART = 0
ICHARG = 1
LCHARG = .TRUE.

ENCUT = 450
ISMEAR = 0
SIGMA = 0.1
POTIM = 0.5
ISYM = 2
#ISPIN = 2

NSW = 0
ISIF = 2
EDIFF = 1E-6
EDIFFG = -0.001
IBRION = -1

#LWAVE = .TRUE.
LORBIT = 0
LNONCOLLINEAR = .TRUE.
LSORBIT = .TRUE.

LPLANE = .TRUE.
NPAR = 4
NSIM = 4

KPOINTS v1
kmesh
0
Monkhorst Pack
12 10 6
0 0 0

KPOINTS v2
kmesh
1
reciprocal
0 0 0 6.

[QijingZheng]

Hi Sotirios The error does not complain about the KPOINTS format. From your INCAR and KPOINTS V1, I guess you are doing a calculation for bulk 1T'-MoTe2? If so, then KPOINTS v1 is appropriate for the calculation, maybe you can increase the size of the k-grid. According to your INCAR, you are doing a calculation with spin-orbit coupling, which, unfortunately, was not taken into account in the wfc_r function in my code. So there will be no "*.vasp" output by the code. Best Qijing

…

On Sun, Apr 29, 2018 at 4:04 PM, sfragkos ***@***.***> wrote: *system = MoTe2* ISTART = 0 ICHARG = 1 LCHARG = .TRUE. ENCUT = 450 ISMEAR = 0 SIGMA = 0.1 POTIM = 0.5 ISYM = 2 #ISPIN = 2 NSW = 0 ISIF = 2 EDIFF = 1E-6 EDIFFG = -0.001 IBRION = -1 #LWAVE = .TRUE. LORBIT = 0 LNONCOLLINEAR = .TRUE. LSORBIT = .TRUE. LPLANE = .TRUE. NPAR = 4 NSIM = 4 *KPOINTS v1* kmesh 0 Monkhorst Pack 12 10 6 0 0 0 *KPOINTS v2* kmesh 1 reciprocal 0 0 0 6. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AIINPmaGUYcUdEKJw8yS8OBYQ8IZPE4pks5ttXQkgaJpZM4TreOg> .

[sfragkos]

Ok Qijing thank you very much!
One last question. If i want to plot the wavefuntions at the Γ point, what ikpt should be equal to?

[QijingZheng]

The ikpt is the index of the k-point that you are interested in, starting from 1.
Check OUTCAR and search the line containing "k-points in reciprocal lattice and weights:", where the k-points coordinates follows.

[sfragkos]

Thanks for everything! need any citation for using your code?

[QijingZheng]

It is not mandatory, but I am glad if you could cite my code so that more people will know it and it may help them with their problems.

[sfragkos]

Ok, how should I write the citation?
I have also one more question, the fact that SOC wasnt included in these calculations, what effect could have to the way the wavefunctions look? ( the calculations were actually for Td-MoTe2 at Γ point and not for 1T' )

[QijingZheng]

I don't know, maybe just a link to the code is ok...

By including SOC in VASP calculations, the wavefunctions then become a spinor \phi -> (\phi^+, \phi^-), i.e. a two-component function. Note that I add a plus/minus sign in the superscript, however , there are no spin up/down for a spinor.

Check these sites for more information:
https://github.com/rubel75/fold2Bloch-VASP
http://www.andrew.cmu.edu/user/feenstra/wavetrans/

[Subhendupj]

Hi Qijing,

What is unit of transition dipole moment calculated in your code? Is it in Debye?

[Subhendupj]

Another question: The "WAVECAR" used for transition dipole moment (TDM) calcualtion should be taken from "scf" calculation or from "band" calculation? I want to calculate the TDM for a particular "K " point (not at Gamma). So the KPOINT value I should give from the OUTCAR of my band calculation. Thats why my doubt is which WAVECAR I should use?

[QijingZheng]

Hi

• I think I convert the TDM in Debye, but I am not so sure about the conversion. Anyway, one usually concerns the relative intensity of the TDM.

• Either WAVECAR is OK, depending on which k-point you are interested in.

Best
Qijing