[Wien] Results of ferromagnetic state is same as ferrimagnetic state?

[Stefaan Cottenier]

>  As you mentioned, I obtained twice the ferrimagnetic solution and this frequently happens. 
>In fact, I used this "flip spin configution" method to get energy difference between ferromagnetic state and ferrimagnetic state
>in the past, and it seems working well. Now I wonder in what situation can this happen? 
>
It happens when DFT (or nature) "does not like at all" the spin 
configuration you have started from. And following the reply of Gerhard 
Fecher, it is clear why this is the case for your compound: being a 4d 
element, Mo is unlikely to develop a big magnetic moment of its own. If 
it gets a moment, it is induced by the Fe moment. And therefore Fe 
dictates completely what the magnitude and orientation of this moment 
should be. If you try to flip it, it will refuse. If, on the other hand, 
you would t ry the same compound with e.g. Co instead of Mo, you will 
probably get solutions both for the ferromagnetic and ferrimagnetic cases.

>And as you mentioned, I can try a 
>FSM calculation to get the virtual ferromagnetic state, the magnetic moment in the unit cell I should specify is 0.00605+3.77188+0.26145+0.00538*2+0.08241*6 = 4.53855 mu_B ? 
>  
>
That is what I meant, yes, but it will be clear now that for Mo this 
doesn't make much sense.

Stefaan


Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm