Machine Learning for Ink Optimization for Printed Electronics: A Perspective

Abstract

Printed electronics, embracing the ability to achieve electronic technologies on surfaces and geometries of varying complexities, have captured massive attention over the past decades. At the heart of the printed electronics is the ink that is used. This perspective article sheds light on the use of machine learning (ML) in optimizing the formulation and the processability (e.g., deposition rate, sintering, etc.) of such inks with the aim of achieving printed traces and components with desired conductivity and structure (e.g., structures with reduced voids, cracks, and coffee stain effects).