Automag: An automatic workflow software for calculating the ground magnetic state of a given structure and estimating its critical temperature
We developed a Python package capable of finding the lowest-energy magnetic state of a given structure and to estimate its critical temperature from a Monte Carlo simulation of its effective Hamiltonian. In this paper, we introduce the code and present the results of tests performed on known materials: α-Fe2O3 (hematite), Ca3MnCoO6 and Ni3TeO6. After checking the calculation parameters for convergence, we computed the linear response value of U for DFT+U and then the single-point energies of a number of collinear magnetic configurations. The magnetic ground state has been correctly predicted for α-Fe2O3 and Ni3TeO6, while for Ca3MnCoO6 the DFT calculations did not reproduce the experimental low-spin states on Co atoms. For α-Fe2O3 and Ni3TeO6 we were able to estimate the Néel temperature and the computed values of 911 K and 31 K are both in good agreement with experiment (955 K and 52 K).