Exploring the magnetic landscape of easily exfoliable novel 2D materials

Magnetic materials can display many self-consistent states in their electronic structure, associated with distinct minima of the energy functional. Finding the global minimum can become challenging, and standard heuristic approaches are not guaranteed to succeed. Here, we apply an automatic workflow to systematically explore the energy landscape and identify the ground-state magnetic order of about 250 magnetic monolayers from the Materials Cloud two-dimensional materials database (MC2D). To achieve this goal, we exploit the Robust Occupation Matrix Energy Optimiser (ROMEO) package, which allows us to effectively control and explore the orbital occupation matrices of magnetic ions, rapidly sampling the different local minima in the magnetic energy landscape. We demonstrate the emergence of a rich zoology of self-consistent collinear states, even more so when using Hubbard-corrected energy functionals (in passing, with the Hubbard U parameter calculated from first-principles from linear-response approach theory). We categorize the different monolayers based on their magnetic ordering and discuss the most promising ones for technological applications.