[Wien] Magnetocrystalline anisotropy energy

[Bin Shao]

Dear Pavel,

Thank you for your reply?

You are definitely right. In general cases, the expression of the total
energy should be modified. But in a uniaxial system, especially in cubic
case, the torque method is correct and we can get the MAE from the
expectation value of the derviative of the spin-orbit coupling Hamiltonian
at the angle of 45. The author has calculated the magnetostriction in Fe-Ga
alloy with variation of the c/c0 value in J. Phys.: Condens. Matter 15 S587.
However, it's difficult for me to implement the torque method in Wien2K,
since I am not familar with Fortran language. Would you like to tell me how
to do the implement? for example, which part of the code do you need to
change?

On the other hand, How many k-points do you need to get a precise using the
' force theorem' for fe monolayer?

Thank you!

Best regards,


On Tue, Sep 21, 2010 at 5:39 PM, Pavel Novak  wrote:

> Dear Bin Shao,
>
> we did implement the torque method in WIEN2k some years ago, but after
> gaining some experience we stopped using it. The problem is that the
> conception is not quite correct: when s-o coupling is strong and
> magnetization is in a general direction, spin is not along the quantization
> axis as we input it in .inso, exception being symmetrical directions, where
> Etot(M) has extrema - in case of (001) Fe film these are e.g. [001], [100],
> [110]. Then, however torque is zero by definition. It is thus more reliable
> to use the 'force theorem'
> (i) converge calculation without s-o
> (ii) using converged non s-o calculation run lapwso, lapw2up/dn for
>M along symmetry directions. Anisotropy is then given by differences
>   of [:SUMup + :SUMdn] for M along say [001] and [100].
> To get reliable results for Fe film is not too difficult, as the symmetry
> is axial. For bulk Fe it does not work - cubic anisotropy is too small.
>
> Regards
> Pavel
>
>
>
>  On Mon, 20 Sep 2010, Bin Shao wrote:
>
>  Dear all,
>>
>> I intend to calculate magetocrystalline anisotropy (MCA) energy of the bcc
>> Fe monolayer. Since the MCA energy usually has an order of magnitude about
>> 10^-6 eV, it's a tough work to get its calculated numerical value. The
>> reference PRB. *54*. 61 proposes a torque method and the MCA energy can be
>> easily evaluated through the expectation value of the angular derivative
>> of
>> the spin-orbit coupling Hamiltonian at an certain angle with this method.
>> The paper gives an example of the free monolayer Fe using FLAPW method and
>> mentioned that "one only needs the self-consistent scalar relativistic
>> charge-spin density or potential and then performs one second-variational
>> calculation with the SOC hamiltonian invoked."
>>
>> So, I want to know how to do this calculation, or to get the angular
>> derivative of the derivative of the spin-orbit coupling Hamiltonian  in
>> Wien2K.
>>
>> Please give me some comments, thank you in advanced!
>>
>> Best regards,
>>
>> --
>> Bin Shao, Ph.D. Candidate
>> College of Information Technical Science, Nankai University
>> 94 Weijin Rd. Nankai Dist. Tianjin 300071, China
>> Email: binshao1118 at gmail.com
>>
>>
> --
> ___
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>



-- 
Bin Shao, Ph.D. Candidate
College of Information Technical Science, Nankai University
94 Weijin Rd. Nankai Dist. Tianjin 300071, China
Email: binshao1118 at gmail.com
-- next part --
An HTML attachment was scrubbed...
URL: 

[Wien] Magnetocrystalline anisotropy energy

[Pavel Novak]

Dear Bin Shao,

we did implement the torque method in WIEN2k some years ago, but after 
gaining some experience we stopped using it. The problem is that the 
conception is not quite correct: when s-o coupling is strong and 
magnetization is in a general direction, spin is not along the 
quantization axis as we input it in .inso, exception being symmetrical 
directions, where Etot(M) has extrema - in case of (001) Fe film these are 
e.g. [001], [100], [110]. Then, however torque is zero by definition. 
It is thus more reliable to use the 'force theorem'
(i) converge calculation without s-o
(ii) using converged non s-o calculation run lapwso, lapw2up/dn for
 M along symmetry directions. Anisotropy is then given by differences
of [:SUMup + :SUMdn] for M along say [001] and [100].
To get reliable results for Fe film is not too difficult, as the symmetry 
is axial. For bulk Fe it does not work - cubic anisotropy is too small.

Regards
Pavel


  On Mon, 20 Sep 2010, Bin Shao wrote:

> Dear all,
>
> I intend to calculate magetocrystalline anisotropy (MCA) energy of the bcc
> Fe monolayer. Since the MCA energy usually has an order of magnitude about
> 10^-6 eV, it's a tough work to get its calculated numerical value. The
> reference PRB. *54*. 61 proposes a torque method and the MCA energy can be
> easily evaluated through the expectation value of the angular derivative of
> the spin-orbit coupling Hamiltonian at an certain angle with this method.
> The paper gives an example of the free monolayer Fe using FLAPW method and
> mentioned that "one only needs the self-consistent scalar relativistic
> charge-spin density or potential and then performs one second-variational
> calculation with the SOC hamiltonian invoked."
>
> So, I want to know how to do this calculation, or to get the angular
> derivative of the derivative of the spin-orbit coupling Hamiltonian  in
> Wien2K.
>
> Please give me some comments, thank you in advanced!
>
> Best regards,
>
> --
> Bin Shao, Ph.D. Candidate
> College of Information Technical Science, Nankai University
> 94 Weijin Rd. Nankai Dist. Tianjin 300071, China
> Email: binshao1118 at gmail.com
>

-- 

[Wien] Magnetocrystalline anisotropy energy

[Bin Shao]

Dear all,

I intend to calculate magetocrystalline anisotropy (MCA) energy of the bcc
Fe monolayer. Since the MCA energy usually has an order of magnitude about
10^-6 eV, it's a tough work to get its calculated numerical value. The
reference PRB. *54*. 61 proposes a torque method and the MCA energy can be
easily evaluated through the expectation value of the angular derivative of
the spin-orbit coupling Hamiltonian at an certain angle with this method.
The paper gives an example of the free monolayer Fe using FLAPW method and
mentioned that "one only needs the self-consistent scalar relativistic
charge-spin density or potential and then performs one second-variational
calculation with the SOC hamiltonian invoked."

So, I want to know how to do this calculation, or to get the angular
derivative of the derivative of the spin-orbit coupling Hamiltonian  in
Wien2K.

Please give me some comments, thank you in advanced!

Best regards,

-- 
Bin Shao, Ph.D. Candidate
College of Information Technical Science, Nankai University
94 Weijin Rd. Nankai Dist. Tianjin 300071, China
Email: binshao1118 at gmail.com
-- next part --
An HTML attachment was scrubbed...
URL: