Objectives

Computational design of phase-change heterostructures (PCHs) with low variability, high switching speed and retention, and tailored thermal and electrical conductivity.

Growth of promising material combinations by physical vapor deposition (PVD) and characterization of their properties by complementary spectroscopic and diffraction techniques.

Fabrication of heterostructure-based cells and characterization of their electrical, structural and thermal behaviour.

Testing of the ability of the PCH cells to emulate synaptic plasticity and implementation of basic learning rules and algorithms.

Demonstration of a PCH-based neural network (NN) for pattern recognition.