[gpaw-users] Rutile TiO2 bulk -- d-orbitals

[José C. Conesa]

Hi,

I am not user of gpaw itself, therefore I am not sure how the pdos is 
obtained in that case. What I meant is: if when computing the pdos the 
definition of the x, y and z axes is made coincident with the a, b and c 
crystal cell axes, one must take into account that in the rutile 
structure the axes of the individual TiO6 octahedra, which define the 
t2g and eg orbitals, are not at all parallel to the a,b and c 
crystallographic axes, and therefore the pdos analysis, if done in the 
said manner, will mix both manifolds.


El 20/09/2017 a las 17:42, Ask Hjorth Larsen escribió:
> Hi,
>
> 2017-09-20 8:39 GMT+02:00 Taha Salavati-Fard via gpaw-users
> <gpaw-users at listserv.fysik.dtu.dk>:
>> Hi,
>>
>> Thanks for the message. I am not sure how it could change the pdos results.
>> This is what I did:
>> I built the structure using crystallographic data. Then I relaxed it with
>> VASP and took the optimized structure and ran a single point energy
>> calculation with gpaw with the same level of theory etc.  to get top.gpw.
>> Finally, I used the script in the manual to get pdos on atomic orbitals.
>> I'd appreciate it if you elaborate how the axes could be the source of
>> errors.
>> Thanks.
> Would you be able to try with setups='sg15' (requires installing the
> sg15 setups, see e.g. 'gpaw install-data')?
>
> Then we can see whehter sg15 and PAW agree.
>
> Best regards
> Ask
>
>>
>>
>> On Sep 20, 2017 2:31 AM, "José C. Conesa via gpaw-users"
>> <gpaw-users at listserv.fysik.dtu.dk> wrote:
>>> Hi,
>>>
>>> Did you take into account that the axes of the Ti octahedra do not
>>> coincide with the crystallographic axes?
>>>
>>>
>>> El 20/09/2017 a las 3:37, Taha Salavati-Fard via gpaw-users escribió:
>>>
>>> Dear GPAW users,
>>> I have come across a problem with pdos on atomic orbitals. I have taken
>>> the famous rutile TiO2 and calculated its dos and pdos onto atomic orbitals.
>>> It is well known that eg and t2g must be separated. However, when I use pdos
>>> method given in the manual and add up d_xy,d_xz and d_yz for t2g and also
>>> d_x^2-y^2 and d_3z^2-r^2 for eg, the result is very strange: eg and t2g do
>>> not split and yet worse, t2g is higher in energy that eg. I have attached
>>> the calculations.
>>> Also please see Fig. 2 in PHYSICAL REVIEW B 82, 115109 (2010) for TiO2
>>> pdos.
>>> I was wondering if anyone can make a comment on this issue.
>>> Thanks-Taha
>>>
>>> --
>>> Taha Salavati-fard
>>> PhD Candidate in Physics
>>> Department of Physics and Astronomy
>>> Department of Chemical and Biomolecular Engineering
>>> Catalyst Center for Energy Innovation (CCEI)
>>> University of Delaware
>>> Room 305, 217 Sharp Lab
>>> Newark, Delaware 19716, USA
>>> Phone:    (+1) 302-831-6381
>>>
>>>
>>> _______________________________________________
>>> gpaw-users mailing list
>>> gpaw-users at listserv.fysik.dtu.dk
>>> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>>>
>>>
>>> --
>>> José C. Conesa
>>> Instituto de Catálisis y Petroleoquímica, CSIC
>>> Marie Curie 2, Cantoblanco
>>> 28049 Madrid, Spain
>>> Tel. (+34)915854766
>>>
>>>
>>> _______________________________________________
>>> gpaw-users mailing list
>>> gpaw-users at listserv.fysik.dtu.dk
>>> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users
>>
>> _______________________________________________
>> gpaw-users mailing list
>> gpaw-users at listserv.fysik.dtu.dk
>> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users

-- 
José C. Conesa
Instituto de Catálisis y Petroleoquímica, CSIC
Marie Curie 2, Cantoblanco
28049 Madrid, Spain
Tel. (+34)915854766

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20170920/6f0ec0f9/attachment.html>