Modeling Spin Fluctuations and Magnetic Excitations from Time-Dependent Density Functional Theory

Harnessing spin fluctuations and magnetic excitations in materials is key in many technologies, spanning from memory devices to information transfer and processing, to name but a few. A proper understanding of the interplay between collective and single-particle spin excitations is still lacking, and it is expected that ab initio simulations based on TDDFT will shed light on this interplay, as well as on the role of such important effects as spin-orbit coupling and related magnetic anisotropies. All the numerical approaches proposed so far to tackle this problem are based on the computationally demanding solution of the Sternheimer equations for the response orbitals or the even more demanding solution of coupled Dyson equations for the spin and charge susceptibilities. In the case of charge excitations, the Liouville-Lanczos approach to TDDFT has already been demonstrated to be a valuable alternative, the most striking of its features being the avoidance of sums over unoccupied single-particle states and the frequency-independence of the main numerical bottlenecks. In this talk we present an extension of this methodology to magnetic excitations and its implementation in Quantum ESPRESSO, as well as a few preliminary results on the magnon dispersions in bulk Fe and Ni.