[Pw_forum] Total magnetization and local magnetic moment

Gabriele Sclauzero sclauzer at sissa.it
Mon Jul 5 12:43:52 CEST 2010


On 07/05/2010 12:05 PM, Padmaja Patnaik wrote:
> I have proceeded with the supercell which is giving correct DOS. Here 
> is the value of total magnetizatin from the scf.out file without 
> relaxation.
>
> !    total energy              = -1494.09037638 Ry
>      Harris-Foulkes estimate   = -1494.09037639 Ry
>      estimated scf accuracy <        6.7E-09 Ry
>
>      The total energy is the sum of the following terms:
>
>      one-electron contribution =   257.19918198 Ry
>      hartree contribution      =   202.08158140 Ry
>      xc contribution           =  -548.49776710 Ry
>      ewald contribution        = -1404.85946098 Ry
>      smearing contrib. (-TS)   =    -0.01391169 Ry
>
>      total magnetization       =     1.02 Bohr mag/cell
>      absolute magnetization    =     1.12 Bohr mag/cell
>
>      convergence has been achieved in  17 iterations
> ------------------------------------------------------------------------
> Here is the portion of the file *.pdos.out  without relaxation
>
>  Atom #   1: total charge =  14.3653, s =  2.4622, p =  6.8403, d =  
> 5.0628,
>                  spin up      =   7.7466, s =  1.2349, p =  3.4229, d 
> =  3.0889,
>                  spin down    =   6.6187, s =  1.2274, p =  3.4174, d 
> =  1.9739,
>                  polarization =   1.1279, s =  0.0075, p =  0.0054, d 
> =  1.1150,
>      Atom #   2: total charge =   2.9244, s =  0.8587, p =  2.0656, d 
> =  0.0000,
>                  spin up      =   1.4798, s =  0.4351, p =  1.0446, d 
> =  0.0000,
>                  spin down    =   1.4446, s =  0.4236, p =  1.0210, d 
> =  0.0000,
>                  polarization =   0.0352, s =  0.0115, p =  0.0236, d 
> =  0.0000,
>      Atom #   3: total charge =   2.9244, s =  0.8587, p =  2.0656, d 
> =  0.0000,
>                  spin up      =   1.4798, s =  0.4351, p =  1.0446, d 
> =  0.0000,
>                  spin down    =   1.4446, s =  0.4236, p =  1.0210, d 
> =  0.0000,
>                  polarization =   0.0352, s =  0.0115, p =  0.0236, d 
> =  0.0000,
> -----------------------------------------------------------------------------------------------------------------
> As you can see atom number 1(this is the impurity) is showing a local 
> magnetic moment. Now, below i add portions of the same two files from 
> with relaxation calculations.
> !    total energy              = -1494.15610974 Ry
>      Harris-Foulkes estimate   = -1494.15610974 Ry
>      estimated scf accuracy <        1.9E-10 Ry
>
>      The total energy is the sum of the following terms:
>
>      one-electron contribution =   270.52901167 Ry
>      hartree contribution      =   195.52169853 Ry
>      xc contribution           =  -548.43270406 Ry
>      ewald contribution        = -1411.76028316 Ry
>      smearing contrib. (-TS)   =    -0.01383273 Ry
>
>      total magnetization       =     1.13 Bohr mag/cell
>      absolute magnetization    =     1.26 Bohr mag/cell
>
>      convergence has been achieved in   9 iterations
>
>


It looks like your impurity (Ge, I guess) acquires a lot of charge upon 
relaxing the structure, so that the d electrons get filled and you lose 
the local moment. This may be a real description or some spurious effect 
due to orbitals overlap (though it's unlikely if these are very 
localized 3d electrons).

About the total magnetization, well I don't know. It may come from 
neighboring atoms if they have donated some charge to the impurity.

HTH

GS

> ------------------------------------------------------------------------------------------------------------
> Atom #   1: total charge =  19.3947, s =  2.5253, p =  7.6414, d =  
> 9.2280,
>                  spin up      =   9.7222, s =  1.2688, p =  3.8289, d 
> =  4.6245,
>                  spin down    =   9.6725, s =  1.2566, p =  3.8125, d 
> =  4.6035,
>                  polarization =   0.0496, s =  0.0122, p =  0.0164, d 
> =  0.0210,
>      Atom #   2: total charge =   3.0599, s =  0.9045, p =  2.1555, d 
> =  0.0000,
>                  spin up      =   1.5315, s =  0.4527, p =  1.0787, d 
> =  0.0000,
>                  spin down    =   1.5285, s =  0.4517, p =  1.0767, d 
> =  0.0000,
>                  polarization =   0.0030, s =  0.0010, p =  0.0020, d 
> =  0.0000,
>      Atom #   3: total charge =   3.0599, s =  0.9045, p =  2.1555, d 
> =  0.0000,
>                  spin up      =   1.5315, s =  0.4527, p =  1.0787, d 
> =  0.0000,
>                  spin down    =   1.5285, s =  0.4517, p =  1.0767, d 
> =  0.0000,
>                  polarization =   0.0030, s =  0.0010, p =  0.0020, d 
> =  0.0000,
> ----------------------------------------------------------------------------------------------------------------------
> Here as you can see the total magnetization increases but the atom 
> number 1 (impurity) is showing zero local magnetic moment.
> I hope i made the problem more clear this time.
>
> Regards
>


-- 

Gabriele Sclauzero, EPFL SB ITP CSEA
PH H2 462, Station 3, CH-1015 Lausanne

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.democritos.it/pipermail/pw_forum/attachments/20100705/f9d2e088/attachment-0001.htm 


More information about the Pw_forum mailing list