[QE-users] Magnetic frustration, MAE, and smearing value for 'relax' of supercell

[후신 부지드]

Dear QE users, I would like to get your opinion about few points which are not clear to me: - If I would like to calculate the Magnetic Anisotropy Energy (MAE) using the total energy difference (Non-collinear + SOC), should I include the Hubbard U correction during the calculation or it is not needed since it is an energy difference calculation ? - If I suspect that my system is frustrated (it is Mn doped DMS) - because Mn itself has this non-collinear magnetic property - and I run a calculation of non-collinear scf calculation in-plane (IP) and out-of-plane (OOP) without SOC. Then I do the same with including SOC this time. What information we can get from the difference between {E(IP)-E(OOP)}without SOC and {E(IP)-E(OOP)}with SOC ? Is it for example that there are another causes of the non-collinearity besides the SOC, like a geometrical frustration ? - Last question is regarding the relaxation of a large supercell: I used 0.05 gaussian smearing. I didn't check the convergence w.r.t. to the smearing term, because of the expensive calculation. Do you have any idea if this smearing value is okay or may be it is very large ? (just for relaxing the system) some details from the input file: &control     calculation ='relax'     restart_mode='restart'     prefix = 'mn451_relax'     outdir = '/home/houcine/outdir'     pseudo_dir = '/home/houcine/pseudo'     nstep=200     etot_conv_thr = 1e-6     forc_conv_thr = 1e-4     verbosity = 'high' &system     ibrav = 0,     nat   = 60,     ntyp  = 3,     ecutwfc=50     ecutrho=400     vdw_corr = 'dft-d3'     occupations='smearing',smearing='gaussian',degauss=0.05     nspin = 2     starting_magnetization(3)=0.5 / &electrons     electron_maxstep=120     conv_thr=1e-6     mixing_beta = 0.2 ATOMIC_SPECIES Sn    118.71000  Sn.pbe-dn-kjpaw_psl.1.0.0.UPF S      32.06600  S.pbe-n-kjpaw_psl.1.0.0.UPF Mn     54.93804  Mn.pbe-spn-kjpaw_psl.0.3.1.UPF K_POINTS (automatic) 5 4 12 0 0 0 it is Mn-doped SnS2, semiconductor, in this calculation among the 20 Sn atoms only one was replaced by Mn Thank you Best Regards Houcine BOUZID Sungkyunkwan University, Korea

_______________________________________________
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users