Ferrimagnet powered memristors for neuromorphic computing

Neuromorphic computing (NC) represents a groundbreaking approach to computation, drawing inspiration from the human brain to enhance power and efficiency compared to traditional computing methods.

However, to achieve even greater efficiency, NC requires hardware that also mimics the behavior of neurons and synapses. Memristive synapses are pivotal components in emulating the synaptic plasticity observed in biological neural networks.

Nevertheless, the implementation of spintronic memristors compatible with state-of-the-art computing is challenging due to the large current density needed, the long switching times present in ferromagnets and the need of small applied magnetic fields.

The objective of this project is then to design, fabricate, and characterize field-free memristive synapses using Gd-based ferrimagnets.

We will use two approaches: all-optical magnetic switching and the use of multiferroics to switch the magnetization by the application of electric fields.

By taking advantage of the unique properties of ferrimagnets, ultrafast and energy efficient magnetic switching, we aim to develop highly efficient synapses that exhibit memristive behavior without the need for an external magnetic field.