Re: [Phonopy-users] Unit of Group Velocity for QE

[Atsushi T. <atz...@gm...>]

Hi,

For your last question, I think no. As shown above, group velocity
calculation is implemented to output in [distance x
ordinary-frequency] = [angular-frequency / (reciprocal distance with
2pi)], where the definition is briefly explained at wikipedia,
https://en.wikipedia.org/wiki/Group_velocity.
If you are very careful, I think it is safer to clearly mention it,
i.e., your definition, when you write your manuscript.

I updated the document a bit,
https://phonopy.github.io/phonopy/setting-tags.html#id3

Togo

On Sat, Jul 1, 2023 at 5:28 PM Anish Das <ani...@gm...> wrote:
>
> Thanks Prof. Togo for your nice clarification.
>
> Yes, actually in phonopy.yaml it is also clearly mentioned that the units are in a.u. I missed it someway
>
> One more query, to convert THz. au to km/s, do we need to multiply it with 2.pi? As phonopy and phono3py give results in ordinal frequency.
>
> On Sat, 1 Jul 2023 at 13:34, Atsushi Togo <atz...@gm...> wrote:
>>
>> HI,
>>
>> THz. au (bohr). I checked it as below.
>> Yes, this is not well written in the documentation.
>>
>> When we doubt it is good to check acoustic mode next to the Gamma
>> point. Note in band.yaml, distance is that in reciprocal space without
>> 2 pi (can be confirmed from basis vectors). In usual case, phonon
>> frequency is in THz. Then let's check. First check is if the group
>> velocity is written to be consistent with the reciprocal distance
>> unit.
>>
>> % cd phonopy/example/Si-QE <- a = 5.4661639157319968 (A) =
>> 10.329552820956636 (Bohr)
>> % phonopy --qe -c Si.in -d --dim="2 2 2" --pa auto
>> % phonopy-load --band "0 0 0 0.5 0.5 0.5" --gv
>>
>> band.yaml
>> ...
>> - q-position: [    0.0100000,    0.0100000,    0.0100000 ]
>>   distance:    0.0016768
>>   band:
>>   - # 1
>>     frequency:    0.1494819209
>>     group_velocity: [    51.4356492,    51.4356492,    51.4356492 ]
>> ...
>>
>> % ipython
>> In [1]: import numpy as np
>>
>> n [2]: np.linalg.norm( [    51.4356492,    51.4356492,    51.4356492 ])
>> Out[2]: 89.08915773468948
>>
>> In [3]: 0.1494819209 / 0.0016768
>> Out[3]: 89.1471379413168
>>
>> % cd phono3py/example/Si-PBE <-- a = 5.46626289 (A)
>> % phono3py-load
>> % ln -s fc2.hdf5 force_constants.hdf5
>> % phonopy -c POSCAR-unitcell --dim 2 2 2 --pa auto --band "0 0 0 0.5
>> 0.5 0.5" --gv --readfc --readfc-format hdf5
>>
>> band.yaml
>> ...
>> - q-position: [    0.0100000,    0.0100000,    0.0100000 ]
>>   distance:    0.0031686
>>   band:
>>   - # 1
>>     frequency:    0.1416641221
>>     group_velocity: [    25.7956834,    25.7956834,    25.7956834 ]
>> ...
>>
>> % ipython
>> In [1]: import numpy as np
>>
>> In [2]: np.linalg.norm([    25.7956834,    25.7956834,    25.7956834 ])
>> Out[2]: 44.679434264761085
>>
>> In [3]: 0.1416641221 / 0.0031686
>> Out[3]: 44.708742693934234
>>
>> As the final checks, Bohr in Angstrom is 0.529177207423948.
>> 0.0016768 / 0.0031686 = 0.5291927034021334
>> 44.679434264761085 / 0.1416641221 / (89.08915773468948 / 0.1494819209)
>> ~= 0.5291899743850342
>>
>> Togo
>>
>> On Sat, Jul 1, 2023 at 4:20 PM Anish Das <ani...@gm...> wrote:
>> >
>> > Dear Experts
>> >
>> > I have found in the manual that the unit of the group velocity is THz. Angstrom. I am using quantum espresso. So in my case, what will be the unit THz. Angstrom or THz. au (bohr) ? As for QE the unit of 'k' is taken as 1/au and for VASP it is 1/Angstrom.
>> >
>> > Thanking You
>> > --
>> >
>> > Anish Das
>> >
>> > M.Tech. (Materials science and Engineering)
>> >
>> > Materials science centre
>> >
>> > IIT Kharagpur, India
>> >
>> > Mob. +91-9800420192
>> >
>> > email id: ani...@gm...
>> >
>> > Google Scholar: https://goo.gl/d2hW9u
>> >
>> > ResearchGate: https://goo.gl/MY55ee
>> >
>> > LinkedIn: https://goo.gl/2ckhSB
>> >
>> > _______________________________________________
>> > Phonopy-users mailing list
>> > Pho...@li...
>> > https://lists.sourceforge.net/lists/listinfo/phonopy-users
>>
>>
>>
>> --
>> Atsushi Togo
>>
>>
>> _______________________________________________
>> Phonopy-users mailing list
>> Pho...@li...
>> https://lists.sourceforge.net/lists/listinfo/phonopy-users
>
>
>
> --
>
> Anish Das
>
> M.Tech. (Materials science and Engineering)
>
> Materials science centre
>
> IIT Kharagpur, India
>
> Mob. +91-9800420192
>
> email id: ani...@gm...
>
> Google Scholar: https://goo.gl/d2hW9u
>
> ResearchGate: https://goo.gl/MY55ee
>
> LinkedIn: https://goo.gl/2ckhSB
>
> _______________________________________________
> Phonopy-users mailing list
> Pho...@li...
> https://lists.sourceforge.net/lists/listinfo/phonopy-users



-- 
Atsushi Togo