[Pw_forum] Regarding Constraint Magnetization

[paresh rout]

Dear all,
I have done magnetic calculation of a compound having two transition metal
 on various magnetic spin order environment (like
G-type,C-type,A-type,ferro etc) using GGA+U method. I have got C-type as my
ground state for GGA+U method and G-type simply for GGA. Here U value
alters my magnetic order and the calculated magnetic moment per site for
transition metal (highlighted portion) are different for C-type and G-type
ordering respectively as follows

Magnetic moment per site: (C-type)
     atom:    1    charge:   13.0085    magn:    3.8352    constr:    0.0000
     atom:    2    charge:   10.9829    magn:   -2.5480    constr:    0.0000
     atom:    3    charge:    5.4002    magn:    0.1295    constr:    0.0000
     atom:    4    charge:    5.4190    magn:    0.1111    constr:    0.0000
     atom:    5    charge:    5.4002    magn:    0.1295    constr:    0.0000
     atom:    6    charge:    5.4190    magn:    0.1111    constr:    0.0000
     atom:    7    charge:   10.6989    magn:    0.0021    constr:    0.0000
     atom:    8    charge:   10.6989    magn:   -0.0021    constr:    0.0000
     atom:    9    charge:    5.4529    magn:   -0.0769    constr:    0.0000
     atom:   10    charge:    5.4529    magn:    0.0769    constr:    0.0000
     atom:   11    charge:   10.9829    magn:    2.5480    constr:    0.0000
     atom:   12    charge:   13.0085    magn:   -3.8352    constr:
 0.0000       (Atom 1 and Atom 12 are of same kind and Atom 2 and Atom 11
are same.)
     atom:   13    charge:    5.4002    magn:   -0.1295    constr:    0.0000
     atom:   14    charge:    5.4190    magn:   -0.1111    constr:    0.0000
     atom:   15    charge:    5.4190    magn:   -0.1111    constr:    0.0000
     atom:   16    charge:    5.4002    magn:   -0.1295    constr:    0.0000
     atom:   17    charge:   10.6989    magn:    0.0021    constr:    0.0000
     atom:   18    charge:   10.6989    magn:   -0.0021    constr:    0.0000
     atom:   19    charge:    5.4529    magn:   -0.0769    constr:    0.0000
     atom:   20    charge:    5.4529    magn:    0.0769    constr:    0.0000

    Magnetic moment per site:    (G-type)
     atom:    1    charge:   13.0381    magn:    3.8388    constr:    0.0000
     atom:    2    charge:   10.9450    magn:   -2.4769    constr:    0.0000
     atom:    3    charge:    5.4272    magn:    0.1417    constr:    0.0000
     atom:    4    charge:    5.4462    magn:    0.1162    constr:    0.0000
     atom:    5    charge:    5.4272    magn:    0.1417    constr:    0.0000
     atom:    6    charge:    5.4462    magn:    0.1162    constr:    0.0000
     atom:    7    charge:   10.6886    magn:    0.0004    constr:    0.0000
     atom:    8    charge:   10.6886    magn:    0.0004    constr:    0.0000
     atom:    9    charge:    5.4781    magn:    0.1434    constr:    0.0000
     atom:   10    charge:    5.4781    magn:    0.1434    constr:    0.0000
     atom:   11    charge:   10.9450    magn:   -2.4769    constr:    0.0000
     atom:   12    charge:   13.0381    magn:    3.8388    constr:    0.0000
     atom:   13    charge:    5.4272    magn:    0.1417    constr:    0.0000
     atom:   14    charge:    5.4462    magn:    0.1162    constr:    0.0000
     atom:   15    charge:    5.4462    magn:    0.1162    constr:    0.0000
     atom:   16    charge:    5.4272    magn:    0.1417    constr:    0.0000
     atom:   17    charge:   10.6886    magn:    0.0004    constr:    0.0000
     atom:   18    charge:   10.6886    magn:    0.0004    constr:    0.0000
     atom:   19    charge:    5.4781    magn:    0.1434    constr:    0.0000
     atom:   20    charge:    5.4781    magn:    0.1434    constr:    0.0000

*I am just wondering whether the magnetic moment per site should be same
for all type of magnetic ordering or it is just fine , because in some
literature people have mentioned that difference  in magnetic moment can
change the total energy of the system*.
If I constrain the magnetic moment for *atom 2 *of C-type to it's value in
the G-type is it meaningful to compare the energy with the unconstrained
G-type calculation?


Kind Regards
Paresh Chandra Rout
Ph.D Student
Indian Institute of Science Education and Research , Bhopal
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150324/a0723da3/attachment.html>