Powered by Indico
v3.2.3
In this talk, I will first explain how exchange interactions can be derived from DFT calculations using the mapping method. This approach involves computing the total energies of different magnetic configurations and fitting them to a Heisenberg Hamiltonian to extract the exchange parameters. I will then introduce a method we developed for identifying optimal supercells for calculating exchange interactions. By systematically generating supercells and analyzing the null space of the mapping equations, we can select supercells that maximize the number of independent exchange parameters while minimizing computational costs. Following this, I will explore the accuracy of exchange interactions calculated using different DFT methods, such as DFT+$U$ and meta-GGA functionals. Their predictive capabilities are evaluated by comparing the calculated transition temperatures with experimental values, highlighting the limitations of traditional GGA+$U$ approaches and the significant improvements offered by newer meta-GGA functionals.