About antiferromagnet Neel temperature

[Jiacheng Shi]

Hi,

I see the video from youtube to simulate Neel temperature of antiferromagnets. But when I try to cut the system into a cylinder, the magnetisation value of one of the sublattices is always zero. In principle, both two sublattices should be very similar with each other. I also try to use jmol to visualize the system, and it shows cylinder without any problem. Does anyone have some advice on this problem? Thank you very much.

Best regards,

Jiacheng

[Jiacheng Shi]

I check the log file. And I find that at the very beginning, both sublattices share 50% of the atoms. But when I create the cylinders, one sublattice has 100% atoms, but another sublattice has 0% atoms. How would this change happen?

[Richard Evans]

Hi Jiacheng,

Yes - this is actually a feature, not a bug! You can attach different materials in the material file to different "materials" in the unit cell file. Without defining the structure, the default allocation is material 1 in the material file to material 1 in the unit cell file. However, for structures you need to define which material is tagged to which material site in the unit cell. To do this, you can add the following line in the material file:

material[1]:unit-cell-category=1

material[2]:unit-cell-category=2

etc

 
This is actually very powerful, as you can do core shell systems, crystals with different inversion, or define different kinds of alloys, eg NdFeB and DyFeB in the same system with only a single unit cell file.

All the best,

Richard

[Jiacheng Shi]

Thank you very much! That works pretty well.

Best regards,

Jiacheng

[Busheng Wang]

Hi Jiacheng,

May I ask about the consumed time for simulating Neel temperature within Vampire? For example, an AFM system contains 12 atoms totally. 

Even a general idea would be appreciated.

Best,

Busheng