Question about magnetic force theorem

[Khaled Badawy]

Hello 

Thank you for providing the valuable TB2J tool.

I have a question regarding its application. 

I’m trying to calculate the curie temperature for a Cr-based monolayer (CrSe2). The process of calculating the exchange went smooth without software technical errors. 

The magnetic moment on Cr atom is 2.8 and the exchange interaction is AFM for 1st NN then FM for 2nd NN. I note there calculated these from a FM ground state. The FM state is only 10 meV/atom higher than the AFM state and I’m interested in its curie temp. 

When I run multibinit I get zero moments for all atoms at any temp higher than zero. No AFM or FM order. Nevertheless o do observe a peak at heat capacity but magnetization starts from zero and grows to residual values without ordering. 

My question is about the applicability of TB2J for magnetic material that are metallic, in my sample, both spin channels are metallic, would that cause the Heisenberg Hamiltonian and magnetic force theorem to be not suitable ? Also I was expecting the AFM TO FM transition but I’m merely seeing any magnetic order at all. 

Note that the exchange interactions are 5-10 meV and anisotropic interactions are ~0.05 meV. I’m concerned that there is no magnetic order at any temp even if 1K. 

Your help is appreciated 

Regards 

[Xu He]

Hello, 

Could you share the related files and the TB2J command, and output files? 

The magnetic force theorem is not limited to insulators. 

From your description it seems that there is the possibility that the system is frustrated due to the competition of the AFM and FM interactions.

Best regards, 

HeXu