Pushing down the lateral dimension of single and coupled magnetic dots to the nanometric scale: characteristics and evolution of the spin-wave eigenmodes

Abstract

Planar magnetic nanoelements, either single- or multi-layered, are exploited in a variety of current or forthcoming spintronic and/or ICT devices, such as read heads, magnetic memory cells, spin-torque nano-oscillators, nanomagnetic logic circuits, magnonic crystals and artificial spin-ices. The lateral dimensions of the elemental magnetic components have been squeezed down during the last decade to a few tens of nanometers, but they are still an order of magnitude larger that the exchange correlation length of the constituent materials. This means that the spectrum of spin-wave eigenmodes, occurring in the GHz range, is relatively complex and cannot be described within a simple macrospin approximation. On the other hand, a detailed knowledge of the dynamical spectrum is needed to understand or to predict crucial characteristics of the devices. With this focused review we aim at the analysis and the rationalization of the characteristics of the eigenmodes spectrum of magnetic nanodots, paying special attention to the following key points: (i) Consider and compare the case of in-plane and out of-plane orientation of the magnetization, as well as of single- and multi-layered dots, putting in evidence similarities and diversities, and proposing a unifying nomenclature and labelling scheme; (ii) Underline the evolution of the spectrum when the lateral size of magnetic dots is squeezed down from hundreds to tens of nanometers, as in current devices, with emphasis given to the occurrence of soft modes and to the change of spatial localization of the fundamental mode for in-plane magnetized dots; (iii) Extend the analysis from isolated elements to twins of dots, as well as to dense arrays of dipolarly interacting dots, showing how the discretized eigenmodes distinctive of the single element transform in finite-width frequency bands of spin waves propagating through the array.