Speaker
Description
Based on recent proposals, small arrays containing a handful of magnetostatically-coupled single-domain nanomagnets, encoding binary information in their two stable magnetic configurations, could be employed for low-power information processing. Whereas thermally-activated moment reversal, which is at the heart of the computation cycle, complicates the design of reliable Boolean gates, it does open the possibility to implement nanomagnetic circuits for novel probabilistic computation schemes.
Up to now, thermal relaxation from an initial field-set to a final low-energy state of a magnetic configuration has been facilitated by global, and slow, heating provided via thermal contacts to an external heat bath. Here, light-controlled thermoplasmonic heating schemes offer a novel approach to local and fast temperature changes, using a hybrid system which combines an efficient plasmonic absorber with a nanomagnetic element. In addition to temperature control by power and focus of the light source, suitable sample design and choice of light polarisation allow for selective heating of sub-lattices. This enables implementation of annealing schemes that have been previously impossible.
Using these new degrees of control, we explore how the relaxation paths of a nanomagnetic circuit can be controlled by different illumination conditions, and test for strategies to steer the system towards specific target states.
