Linear spin wave theory of large magnetic unit cells using the Kernel Polynomial Method

Linear spin wave theory (LSWT) has proved remarkably successful in describing the low energy dynamics of quantum magnets and is widely used to fit data from spectroscopic measurements. However, a key limitation is the numerical cost of performing LSWT calculations, especially in neutron scattering measurements where data are collected with fine momentum resolution over a large volume of reciprocal space.

In this talk we will discuss how the computational complexity of the LSWT modeling of dynamical correlations can be reduced from cubic to linear in matrix size. We will detail how this can be employed to describe systems with large magnetic unit cells such as skyrmion lattice and disordered systems and demonstrate that this approach dramatically reduces the calculation time, making inverse modeling a more tractable problem.