Modelling Ultrafast THz-Induced Antiferromagnetic Magnetisation Dynamics for Next Generation Data Processing

Globally, we are seeing a transition towards cloud-based storage and computing. To meet this ever increasing demand for data, endeavours should be made to find ways of storing digital information at greater densities and improved efficiencies. At present, cloud-based storage works by storing digital information in large data centres filled with Hard Disk Drives (HDDs). These drives contain platters of ferromagnetic grains, which correspond to 0 and 1 bit states. The most efficient means of reversal is via the resonance mode in ferromagnets, which sits in the GHz range meaning reversal is generally limited to ns timescales. Antiferromagnets are a class of magnetic material where neighbouring atomic magnetic moments are aligned antiparallel, resulting in no net magnetisation. For digital storage applications, antiferromagnets offer high stability and insensitivity to external magnetic fields due to their inherently strong exchange field that arises from the coupling between the two sublattices. This opens the possibility for reduced bit sizes because of the lack of stray fields between grains. The exchange field also gives rise to inherently fast THz magnetisation dynamics, orders of magnitudes faster than the ferromagnets currently found in HDD technology opening the possibility for switching on picosecond, or ultrafast, timescales. To better understand the applicability of antiferromagnets for future storage and memory applications, in this thesis we use computational models to study properties and switching dynamics of the antiferromagnet Mn2Au and toy models of layered materials with antiferromagnetic and ferromagnetic order. A multiscale model of Mn2Au is presented and verified against previous analytical and theoretical work. The feasibility of switching using THz frequency fields is then investigated using atomistic and micromagnetic models across a range of temperatures. We then present an atomistic model of Mn2Au coupled ferromagnetically to Permalloy and perform further switching simulations and show that there is a significant speedup in the switching compared to a pure ferromagnetic system while still being able to access the information via conventional readout methods of the ferromagnetic Permalloy layer. Finally, we study toy multilayer and thin film systems and investigate how standing spinwave modes can be used to reduce the minimum field strengths for switching.