Magnetism in crystalline materials is traditionally described based on the magnetic moments carried by magnetic atoms, with the formation of magnetic structures understood as self-organization arising from pairwise interactions between these moments. In this talk, I will discuss two examples that transcend this traditional framework, discovered through neutron scattering experiments. Firstly, in cobalt oxides with a layered hexagonal honeycomb lattice [1], we observed novel magnetic structures known as multi-q order [2], characterized by clusters of magnetic moments that are separated from each other by less spinful sites [3], in analogy to molecular solids (such as iodine) featuring weak chemical bonds. Theoretical analysis [4] suggests that multi-moment interactions are crucial for the emergence of such magnetic orders. Secondly, in the near-ferromagnet MnSi, we find that the fundamental magnetic units are not individual Mn atoms, but are instead multi-atom electron clouds dubbed magnetic molecular orbitals [5]. We propose that the understanding of our observations can benefit from a reexamination of the single-particle physics with topological band theory. The associated correlated electron problems may represent a new frontier in magnetic quantum materials.

References:
[1] Yao and Li, PRB 101, 085120 (2020); Li et al., PRX 12, 041024 (2022).
[2] Chen et al., PRB 103, L180404 (2021); Yao et al., PR Research 5, L022045 (2023); Gu et al., arXiv:2306.07175 (to appear as a Letter in PRB).
[3] Yao et al., PRL 129, 147202 (2022).
[4] Krüger et al., PRL 131, 146702 (2023).
[5] Jin et al., Sci. Adv. 9, eadd5239 (2023).