[Pw_forum] Spin-Orbit Calculation

[Ajit Kumar Jena]

Thanks a lot Roberta for your informative reply.

___
Ajit
On 10-Sep-2015 10:30 pm, "roberta pigliapochi" <rpigliapochi at gmail.com>
wrote:

> Dear Ajit,
> please bear in mind that I AM NOT AN EXPERT and I would instead really
> appreciate comments from more experienced users. I will try to give some
> advice but take it with a pinch of salt. I would also like to follow up on
> this by asking:
>
> can noncolinear calculations and DFT+U go together now?
>
> Referring to your points:
>
> 1)the AFM initial guess is set up fine. The system might still converge to
> a FM situation, though. If instead you'd rather force it to stay AFM I'd go
> for 'tot_magnetization'
> 2)yes. You may also need to specify lda_plus_u=.true. beforehand.
> 3)I'd say so, and I also think that the 'angle' isn't a fixed value but
> rather will be optimised during the scf. Instead if you'd prefer the two
> (or total) magnetizations to be kept fixed along y I'd go for
> 'constrained_magnetization'
> 4)I think that the default 'domag' (whatever it means) is set to false,
> hence there should be no need for your flag to be there. Also, 'nspin=4'
> may be redundant considering that you have set 'noncolin=.true.'
>
> hope some more hard-core expert will comment on this.
>
> good luck (:
> Roberta
>
> Roberta Pigliapochi
> PhD Student | Grey Research Group
> University of Cambridge
>
>
> On 8 September 2015 at 12:48, Ajit Kumar Jena <jenapsi at gmail.com> wrote:
>
>> Dear All,
>>             I wish to do a *spin-orbit calculation* of our system
>> (LiFePO4) which involves iron (Fe) as a magnetic element. Her*e *I am
>> providing my input variables containing the system name lists and atomic
>> species.
>>
>> My questions are the following :
>>
>> 1) As it is mentioned in the input, should I consider two different types
>> of spins ( up and down ) to do a anti-ferromagnetic calculation ?
>> 2) The way I have incorporated Hubbard 'U' is appropriate or not ?
>> 3) If I am giving the inputs in this way, will it take the starting spin
>> quantization axis along Y-direction or not ?
>> 4) Is there any problem if i will not include " starting_spin_angle =
>> .true. " ?
>>
>> Any kind of help is highly appreciated. I am extremely sorry if my
>> questions are very trivial.
>>
>> *Input*
>> &system
>>     ibrav=  8,
>>     celldm(1) = 19.326236,
>>     celldm(2) = 0.587151,
>>     celldm(3) = 0.458766,
>>     nat=  28,
>>     ntyp= 7,
>> !    nspin = 4,
>>     starting_magnetization(2) = 0.7,
>>     starting_magnetization(3) = -0.7,
>>     ecutwfc =30.0,
>>     ecutrho = 250.0,
>>     occupations = 'smearing' ,
>>     degauss = 0.001 ,
>>     smearing = 'gaussian' ,
>>     noncolin = .true.
>>     lspinorb = .true.
>>     starting_spin_angle = .true.
>>     angle1(2) = 90.0
>>     angle2(2) = 90.0
>>     angle1(3) = 90.0
>>     angle2(3) = 90.0
>>     lda_plus_u_kind = 0,
>>     Hubbard_U(2)=3.0,
>>     Hubbard_U(3)=3.0,
>>
>> ATOMIC_SPECIES
>>  Li1  6.94      Li.pbe-s-van_ak.UPF
>>  Fe1  55.845    Fe.rel-pbe-spn-rrkjus_psl.0.2.1.UPF
>>  Fe2  55.845    Fe.rel-pbe-spn-rrkjus_psl.0.2.1.UPF
>>  P1   30.9737   P.pbe-van_ak.UPF
>>  O1   15.999    O.pbe-van_ak.UPF
>>  O2   15.999    O.pbe-van_ak.UPF
>>  O3   15.999    O.pbe-van_ak.UPF
>>
>> Thanks & Regards,
>> Ajit,
>> IIT Madras, Chennai
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>
>
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://pwscf.org/mailman/listinfo/pw_forum
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150910/9ae83666/attachment.html>