[QE-users] Pseudo potential

Mon Aug 31 20:42:42 CEST 2020

Good evening sir,

I was actually trying to do a relax calculation with PAW pseudopotential
and revPBE XC functional. I observed that the no of kohn sham states are
more than no of electrons. When I tried setting nbnd=308 it gave error
about not able to set the Fermi level.
And I used nspin = 4 and in the output, it says "Noncolinear calculation
without spin orbit"

Can you please explain if I am doing something wrong here?

I will be highly obliged.

Regards
Yuvam Bhateja

[image: Screenshot (141).png]

On Mon, Aug 31, 2020 at 5:44 PM Yuvam Bhateja <yuvamb16 at gmail.com> wrote:

> Thanks for the clarification sir,
>
> I feel so relieved now.
>
> I'll try to learn more about how DFT actually works.
>
> Thank you for the help.
>
> Regards
> Yuvam Bhateja
>
>
> On Mon, 31 Aug 2020, 5:42 pm Pietro Delugas, <pdelugas at sissa.it> wrote:
>
>> The shape of the DOS does not depend on the value of the Fermi level you
>> give in input, so the DOS should be right.
>>
>> dont get too fixated with definitions, for insulators text books indicate
>> the (intrinsic )  Fermi Level in the middle of the HOMO-LUMO gap.
>>
>> For the program the fermi level is just any value in the range between
>> the HOMO and the LUMO, so it is ok to use any value in this range.
>>
>>
>> On 31/08/20 13:53, Yuvam Bhateja wrote:
>>
>> Thank you sir, I got your point now.
>>
>> But when I used to do DOS of my system, I used to get Fermi energy at the
>> edge of HUMO and sometimes inside the band (non zero). But since band gap
>> was coming very big, I assumed that it is acting as an insulator.
>>
>> Does that means my DOS is wrong?
>> In most of the publication either there is no mention of Fermi energy or
>> the same result i.e., at the edge of HUMO.
>>
>> I employed DFT+U too but there was no fractional occupation in it.
>>
>> Regards
>> Yuvam Bhateja
>>
>> On Mon, 31 Aug 2020, 5:17 pm Pietro Delugas, <pdelugas at sissa.it> wrote:
>>
>>> look at line 1064, there is an occupation = 0.5 which means the that
>>> state is very close to the Fermi energy.
>>>
>>> if all occupations are either 1 or 0, then yes it is an insulator.
>>> Inspecting the scf output is not the most clever way to check though. What
>>> one should do is to compute the density of states around the Fermy energy
>>> if it is vanishing it is an insulator, if it is different than 0 is a
>>> metal.
>>>
>>>
>>> On 31/08/20 13:37, Yuvam Bhateja wrote:
>>>
>>> Thank you sir for this valuable information.
>>>
>>> But as you mentioned fractional, but all I can see is 1.0000 and 0.0000.
>>>
>>> And the occurrence of 1.000 and 0.000 means it is an insulator? Because
>>> In my scf calculation, I observed the same and it is shown to be a AFM
>>> insulator - Cr2O3.
>>>
>>> Regards
>>> Yuvam Bhateja
>>>
>>> On Mon, 31 Aug 2020, 4:50 pm Pietro Delugas, <pdelugas at sissa.it> wrote:
>>>
>>>> Yes Occupation Numbers. For each   k-point are first printed all the
>>>> computed  eigenvalues and then occupation for each KS state.
>>>>
>>>>
>>>> On 31/08/20 11:57, Yuvam Bhateja wrote:
>>>>
>>>> Dr. Pietro,
>>>>
>>>> Sorry for putting another question in a thread.
>>>> When you said fractional occupation, from where are you reading the
>>>> data?
>>>>
>>>> Are you saying on the basis of data written below
>>>> "k = 0.0000 0.0000-1.0000 (  4508 PWs)   bands (ev):"
>>>> or the "Occupation numbers."
>>>>
>>>> I apologize in advance if it is a stupid question, I am very new in DFT.
>>>>
>>>> Regards
>>>> Yuvam Bhateja
>>>>
>>>> On Mon, 31 Aug 2020, 1:47 pm Pietro Delugas, <pdelugas at sissa.it> wrote:
>>>>
>>>>> Hi
>>>>>
>>>>>  from the results of your calculation you have fractional
>>>>> occupations.  Does the literature report this material as insulator or as
>>>>> metal ?
>>>>>
>>>>>  It might be that this fractional occupations are caused by the wrong
>>>>> dispersion of the some atomic states. This a very well known problem of LDA
>>>>> or PBE with localized d or f atomic states. One way to correct this
>>>>> behavior is using LDA+U approach.
>>>>>
>>>>> I hope this helps
>>>>>
>>>>> regards
>>>>>
>>>>> Pietro
>>>>> On 31/08/20 09:46, tanmay chaki wrote:
>>>>>
>>>>> Dear sir,
>>>>> As your suggestions I have changed the required things. But I got
>>>>> total magnetization zero. but this compound has a magnetic moment which is
>>>>> reported in the journal. I have attached the input and output file. please
>>>>> help me to do correct calculations.
>>>>> Thanking you.
>>>>>
>>>>> Tanmay Chaki
>>>>> Research Scholar
>>>>> N.B.U
>>>>>
>>>>>
>>>>> On Fri, Aug 28, 2020 at 11:16 PM tanmay chaki <tanmaychaki at gmail.com>
>>>>> wrote:
>>>>>
>>>>>> Thanks for your suggestions
>>>>>>
>>>>>> On Fri, 28 Aug, 2020, 10:43 pm Yuvam Bhateja, <yuvamb16 at gmail.com>
>>>>>> wrote:
>>>>>>
>>>>>>> Hey,
>>>>>>>
>>>>>>> As per my knowledge, there's no specific pseudopotential for spin
>>>>>>> calculation.
>>>>>>>
>>>>>>> You need to use commands like nspin and starting_magnetization in
>>>>>>> order to start your spin calculation.
>>>>>>>
>>>>>>> Your choice of XC Functional can affect the results in your spin
>>>>>>> calculation.
>>>>>>>
>>>>>>> Let me know if it helps.
>>>>>>>
>>>>>>> Regards
>>>>>>> Yuvam Bhateja
>>>>>>>
>>>>>>> On Fri, 28 Aug 2020, 10:38 pm tanmay chaki, <tanmaychaki at gmail.com>
>>>>>>> wrote:
>>>>>>>
>>>>>>>> Hello QE users,
>>>>>>>> Is there any specific Pseudo potential for spin polarized
>>>>>>>> calculation?
>>>>>>>> _______________________________________________
>>>>>>>> Quantum ESPRESSO is supported by MaX (
>>>>>>>> www.max-centre.eu/quantum-espresso)
>>>>>>>> users mailing list users at lists.quantum-espresso.org
>>>>>>>> https://lists.quantum-espresso.org/mailman/listinfo/users
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Quantum ESPRESSO is supported by MaX (
>>>>>>> www.max-centre.eu/quantum-espresso)
>>>>>>> users mailing list users at lists.quantum-espresso.org
>>>>>>> https://lists.quantum-espresso.org/mailman/listinfo/users
>>>>>>
>>>>>>
>>>>> _______________________________________________
>>>>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
>>>>> users mailing list users at lists.quantum-espresso.orghttps://lists.quantum-espresso.org/mailman/listinfo/users
>>>>>
>>>>> _______________________________________________
>>>>> Quantum ESPRESSO is supported by MaX (
>>>>> www.max-centre.eu/quantum-espresso)
>>>>> users mailing list users at lists.quantum-espresso.org
>>>>> https://lists.quantum-espresso.org/mailman/listinfo/users
>>>>
>>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200901/c3a449b3/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screenshot (141).png
Type: image/png
Size: 187701 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200901/c3a449b3/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: single.opt_4.58_xdm2.in
Type: application/octet-stream
Size: 4522 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20200901/c3a449b3/attachment.obj>