There are three types of antiferromagnetism, namely A , C and G type. I want to perform DFT calculations with these types of antiferromagnetism using either Quantum Espresso or Wien2K codes in full inverse Heusler alloys with general formula X2YZ (X and Y being transition elements and Z being an sp element. The Xs are at (0,0,0) and (0.25,0.25,0.25), Y at (0.5,0.5,0.5) and Z at (0.75,0.750.75).
Dear Mahesh Ram,
I'm not quite sure what you're asking, but in general all you will need to do is to run the DFT calculation in "spin polarised" mode, and initialise the atomic spins to an appropriate AFM configuration. Depending on the code, there may be options to actually constrain the spins to a particular configuration; if not, you may be limited to studying stable and metastable AFM configurations (which might be fine, it depends what you want to study).
I should add that I don't use QE or Wien2K, I'm a CASTEP developer (http://www.castep.org), but I don't think there will be any fundamental differences in how the modelling is performed. Of course if you wanted to use CASTEP for the simulations I could give more specific advice.
All the best,
Phil Hasnip
Respected Philip James Hasnip sir,
Thanks for responding to my query and extending your help. But unfortunately, I don't use CASTEP.
I did use spin-polarized DFT for AFM configuration and accordingly gave initial atomic spins. And you can constrain the atomic spins in QE.
To add to my previous question, there are different elements say X and Y in a unit cell. And in antiferromagnetic calculations say for A type, the atomic spins in alternate planes are oriented in opposite directions. Now suppose if I give initial opposite spins to X atoms in alternate planes then what about Y?
Dear Mahesh,
You can do the antiferromagnetism calculation by the following general way. In Quantum ESPRESSO (QE), we should first mind the Atomic species.
1. If you would set the antiferromagnetism at X site, please write the input like this (ATOMIC_SPECIES card):
X1 {atomic mass) X.......UPF
X2 {atomic mass) X.......UPF
Y {atomic mass) Y.......UPF
Z {atomic mass) Z.......UPF
We can see that we have i = 1, 2, 3, 4 as the order of atoms in the systems.
2. Then, please set (in the SYSTEM card)
starting_magnetization(1) = 0.5 (I usually set this value for initial individual atomic magnetic moment)
starting_magnetization(2) = -0.5
We see that X1 and X2 have the opposite magnetic moment directions to each other, leading to the INITIAL antiferromagnetism configuration.
DO NOT FORGET to set nspin = 2.
3. In the ATOMIC_POSITIONS card, please use:
X1 0 0 0
X2 0.25 0.25 0.25
Y 0.5 0.5 0.5
Z 0.75 0.75 0.75
----------------
I have used this technique in an antiferromagnetic layered compound (not published yet) and it seems to work well (the nscf output shows the antiferromagnetism very well)
Note that I haven't consider the A-, C-, and G-type antiferromagnetism in this answer.
May this helps.
Best regards,
SM
Cite
5th Nov, 2019
Mahesh Ram
North Eastern Hill University
Respected Shibghatullah Muhammady
Thank you for your valuable suggestions. I am using the technique you described above for ferrimagnetic calculations. I shall try this way as well.
Thanks and regards.
Cite
5th Nov, 2019
Shibghatullah Muhammady
The University of Tokyo
Dear Mahesh Ram
You are welcome
SM
Cite
1st Apr, 2020
Deepika Shrivastava
Barkatullah University
Dear Shibghatullah Muhammady
I read your valuable suggestions and it helped me a lot, my question is, if we have ternary HH compounds like XYZ then how do we apply the antiferromagnetism ?
Cite
26th Feb, 2021
Saurav Lahiri
Indian Institute of Technology (ISM) Dhanbad
Respected Shibghatullah Muhammady , should we have to put tot_magnetization = 0 in case of AFM order?
Cite
27th Feb, 2021
Shibghatullah Muhammady
The University of Tokyo
Deepika Shrivastava, I do not know, sorry.
Saurav Lahiri, I do not think so. I did not set the tot_magnetization = 0
Cite
2nd Mar, 2021
Saurav Lahiri
Indian Institute of Technology (ISM) Dhanbad
OK. Thanks Shibghatullah Muhammady
Cite
Can you help by adding an answer?
Answer
- Badges
- Science topic
- Similar topics
- Names