magnetization calculation for LaFeO3

Posted on : October 09, 2021 (Sat) 16:39:51

by Hong Jian Zhao

Dear all,
I am testing the calculation for Pbnm LaFeO3. I am expecting to have a ferromagnetic solution with magnetization of about 20 uB [In this compound, Fe3+ have 5 d electrons, leading to local magnetic moment of ~ 5 uB; so in total we should have 20 uB in the cell containing 4 Fe ions]. However, after the calculation, I get null magnetization and the Fe is not spin polarized. Could you please help me solve this problem? Thanks in advance.


c----------------------LaFeO3----------------------------------
go data/lafeo3
c------------------------------------------------------------
c brvtyp a c/a b/a alpha beta gamma
so 10.582 1.41881 1.0107 90 90 90
c------------------------------------------------------------
c edelt ewidth reltyp sdftyp magtyp record
0.001 1.6 sra mjw mag init
c------------------------------------------------------------
c outtyp bzqlty maxitr pmix
update 1 150 0.025
c------------------------------------------------------------
c ntyp
4
c------------------------------------------------------------
c type ncmp rmt field mxl anclr conc
La 1 0 0.0 2 57 100
Fe1 1 0 0.0 2 26 100
Fe2 1 0 0.0 2 26 100
O 1 0 0.0 1 8 100
c------------------------------------------------------------
c natm
20
c------------------------------------------------------------
c atmicx atmtyp
0.990619004 0.038259000 0.250000000 La
0.009380996 0.961740971 0.750000000 La
0.490619004 0.461741000 0.750000000 La
0.509380996 0.538259029 0.250000000 La
0.500000000 0.000000000 0.000000000 Fe1
0.000000000 0.500000000 0.000000000 Fe2
0.500000000 0.000000000 0.500000000 Fe2
0.000000000 0.500000000 0.500000000 Fe1
0.212594002 0.212025002 0.546275020 O
0.787405968 0.787975013 0.453724980 O
0.712594032 0.287975013 0.453724980 O
0.287405998 0.712024987 0.546275020 O
0.787405968 0.787975013 0.046275020 O
0.212594002 0.212025002 0.953724980 O
0.287405998 0.712024987 0.953724980 O
0.712594032 0.287975013 0.046275020 O
0.586120009 0.021734999 0.750000000 O
0.413879991 0.978264987 0.250000000 O
0.086120009 0.478264987 0.250000000 O
0.913879991 0.521735013 0.750000000 O
c------------------------------------------------------------

[Re:03] magnetization calculation for LaFeO3

Posted on : October 11, 2021 (Mon) 20:01:36

by Hong Jian Zhao

Dear Dr. Hitoshi GOMI,
Many thanks. The problem has been solved. Indeed, I should use "> 0.990619004a 0.038259000b 0.250000000c La" as the input for atomic position. Thanks again.

[Re:02] magnetization calculation for LaFeO3

Posted on : October 11, 2021 (Mon) 17:22:25

by Hong Jian Zhao

Dear Dr. Hitoshi GOMI,
Thanks a lot for your suggestions. Using the fractional coordinates, it seems that the calculation is improved. Now I have got the magnetic moment of ~ 16.2374 uB (still smaller than 20 uB, but much better than 0).
Now I am confused by two aspects. First, after giving coordinates as
0.990619004a 0.038259000b 0.250000000c La
0.009380996a 0.961740971b 0.750000000c La
0.490619004a 0.461741000b 0.750000000c La
0.509380996a 0.538259029b 0.250000000c La
0.500000000a 0.000000000b 0.000000000c Fe
0.000000000a 0.500000000b 0.000000000c Fe
0.500000000a 0.000000000b 0.500000000c Fe
0.000000000a 0.500000000b 0.500000000c Fe
0.212594002a 0.212025002b 0.546275020c O
0.787405968a 0.787975013b 0.453724980c O
0.712594032a 0.287975013b 0.453724980c O
0.287405998a 0.712024987b 0.546275020c O
0.787405968a 0.787975013b 0.046275020c O
0.212594002a 0.212025002b 0.953724980c O
0.287405998a 0.712024987b 0.953724980c O
0.712594032a 0.287975013b 0.046275020c O
0.586120009a 0.021734999b 0.750000000c O
0.413879991a 0.978264987b 0.250000000c O
0.086120009a 0.478264987b 0.250000000c O
0.913879991a 0.521735013b 0.750000000c O

I got the following output:
atoms in the unit cell
position= 0.99061900 0.03866837 0.35470250 type=La
position= 0.00938100 0.97203160 1.06410750 type=La
position= 0.49061900 0.46668163 1.06410750 type=La
position= 0.50938100 0.54401840 0.35470250 type=La
position= 0.50000000 0.00000000 0.00000000 type=Fe
position= 0.00000000 0.50535000 0.00000000 type=Fe
position= 0.50000000 0.00000000 0.70940500 type=Fe
position= 0.00000000 0.50535000 0.70940500 type=Fe
position= 0.21259400 0.21429367 0.77506046 type=O
position= 0.78740597 0.79640635 0.64374954 type=O
position= 0.71259403 0.29105635 0.64374954 type=O
position= 0.28740600 0.71964365 0.77506046 type=O
position= 0.78740597 0.79640635 0.06565546 type=O
position= 0.21259400 0.21429367 1.35315454 type=O
position= 0.28740600 0.71964365 1.35315454 type=O
position= 0.71259403 0.29105635 0.06565546 type=O
position= 0.58612001 0.02196756 1.06410750 type=O
position= 0.41387999 0.98873242 0.35470250 type=O
position= 0.08612001 0.48338242 0.35470250 type=O
position= 0.91387999 0.52731758 1.06410750 type=O
It still seems that the position is not correct. For example, the lattice parameter a is 10.582, while the coordinates in the input and output do not coincide with each other.

Second, what should the input "ewidth" be?
For example, I have a system with the density of state as:

__core state (e1 to e2)__valence state (e3 to e4)__fermilevel e5___conduction state (e6 to e7)___
Should I input the ewidte as e5-e3, or e7-e3?
Many thanks for your help.

[Re:01] magnetization calculation for LaFeO3

Posted on : October 11, 2021 (Mon) 13:02:19

by Hitoshi GOMI

Hello Hong Jian Zhao,

I am not familiar with LaFeO3, but I think you are specifying the atomic positions incorrectly.
In your input file, the atomic positions are given in the Cartesian coordinates.
> 0.990619004 0.038259000 0.250000000 La

However, I think they should be given in the fractional coordinates.
> 0.990619004a 0.038259000b 0.250000000c La

Hitoshi GOMI