Science-based, data-driven developments in plasma processing for material synthesis and device-integration technologies

TLDR

Low‑temperature plasma‑processing technologies are essential for material synthesis and device fabrication, and their development requires understanding multiscale plasma behaviors—including generation, transport, and surface reactions—which are pivotal for new applications and control methods. The review aims to examine the requirements and challenges of research and development in material science and plasma processing. It outlines how science‑based, data‑driven approaches—plasma informatics—can predict plasma parameters and discover processing recipes.

Abstract

Abstract Low-temperature plasma-processing technologies are essential for material synthesis and device fabrication. Not only the utilization but also the development of plasma-related products and services requires an understanding of the multiscale hierarchies of complex behaviors of plasma-related phenomena, including plasma generation in physics and chemistry, transport of energy and mass through the sheath region, and morphology- and geometry-dependent surface reactions. Low-temperature plasma science and technology play a pivotal role in the exploration of new applications and in the development and control of plasma-processing methods. Presently, science-based and data-driven approaches to control systems are progressing with the state-of-the-art deep learning, machine learning, and artificial intelligence. In this review, researchers in material science and plasma processing, review and discuss the requirements and challenges of research and development in these fields. In particular, the prediction of plasma parameters and the discovery of processing recipes are asserted by outlining the emerging science-based, data-driven approaches, which are called plasma informatics.