Chalcogenide PC materials have two stable phases solid phases with different electric conductivity or optical reflectivity, and the material can be switched repeatedly between the phases. PC materials are used in data storage for computers, but traditionally they have been used in the alloy form, where one phase is ordered crystalline and another is disordered amorhpous, and the transition between these requires collective movement of all atoms.
Chalcogenide heterostructures are built by stacking different 2D-layers of chalcogenide materials into a 3D material. These show the same kind of electrical or optical contrast as the alloy PC materials but here the switch is constrained in a small subset of atoms, and thus is usually faster and requires less energy. The properties of a heterostructure can be tuned by changing the layer order, thickness of materials.
This project uses machine learning and material modelling to study how the changed in layering affects the properties of the final material. The aim is to identify 'building blocks' of a few layers that could be used to desing materials with bespoke properties, and to optimize the heterostructures for data storage applications.
Funding source
Academy of Finland
Coordinating organisation
Tampere University