Dear all, May I ask: is it correct to assume that the WIEN2k code will Optimizes the spin structure of a stucture/cell in ground state if, I choose spin-polarized calculation for the structure/cell from the beginning?
What I mean is: for a compound, I only suspecting that it is magnetic and I have to include spin-polarization in the calculation, but I don't know wheather it is ferromagnetic, perfect-antiferromagentic or canted antiferromagnetic or even a mixed state of the magnetic status, so I would just do a spin-polarized calculation (selecting all the atoms in the cell to be spin-up from the beginning), then the resulting converged spin state would be the loweset energy state of the compound? - if that is a yes then it i s basically the case that, Wien2k sort it out itself and practically it was an Optimized spin structure for the compound in the ground state? Thank you for your time. Kind regards, Qiwen
If you start from a ferromagnetic situation, you will converge towards a ferromagnetic state (I never encounter other situations). However, it does not mean that you will find the global minimum in terms of magnetic structure. It simply means that you have converged the electronic density related to the ferromagnetic configuration of your system (It will be a local minimum - a metastable situation).
To know which magnetic structure is the more stable, you should try different magnetic ordering, or at least one ferromagnetic situation and few antiferromagnetic situations (if possible). Then by comparing the total energy of the different situations you will see which one is the more stable. In the other hand, another way to estimate the magnetic interactions in a system is to estimate the magnetic exchange couplings, J(ij), using a mapping analysis (based on energy differences between different magnetic orders). It will give you the nature of the magnetic interaction in all directions and will help you to figure out what is the fundamental magnetic state of your system. Best Regards Xavier
Actually.... I have come across several cases where I started with a ferromagnetic state and it converged to non-magnetic or even antiferromagnetic, or the opposite. It does take a long time for it to change, so as Xavier says it is probably better to start from different minima.
Note: the atomic positions can couple in to the spin-state, so you may need to do a full minimization not just a single set of atomic positions.
Reference/ https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg06341.html
My name is Dr Abderrahmane Reggad. I'm a 54 year old and I am a researcher in materials' sciences. 
0 Comments