Bayesian Machine Learning for Efficient Minimization of Defects in ALD Passivation Layers

Abstract

Atomic layer deposition (ALD) is an enabling technology for encapsulating sensitive materials owing to its high-quality, conformal coating capability. Finding the optimum deposition parameters is vital to achieving defect-free layers; however, the high dimensionality of the parameter space makes a systematic study on the improvement of the protective properties of ALD films challenging. Machine-learning (ML) methods are gaining credibility in materials science applications by efficiently addressing these challenges and outperforming conventional techniques. Accordingly, this study reports the ML-based minimization of defects in an ALD-Al 2 O 3 passivation layer for the corrosion protection of metallic copper using Bayesian optimization (BO). In all experiments, BO consistently minimizes the layer defect density by finding the optimum deposition parameters in less than three trials. Electrochemical tests show that the optimized layers have virtually zero film porosity and achieve five orders of magnitude reduction in corrosion current as compared to control samples. Optimized parameters of surface pretreatment using Ar/H 2 plasma, the deposition temperature above 200 °C, and 60 ms pulse time quadruple the corrosion resistance. The significant optimization of ALD layers presented in this study demonstrates the effectiveness of BO and its potential outreach to a broader audience, focusing on different materials and processes in materials science applications.

Algorithm-Driven Paradigms for Freeform Optical Engineering

“…For materials growth, use of the data sciences to improve the quality of grown materials has become a field in itself . To cite a representative example, Bayesian optimization can be used to identify material growth parameters in just a few trials that yield optimal films with virtually no porosity (Figure a). Deep neural networks have become a standard tool to perform optical proximity correction (OPC) in the lithography process, in which lithography mask patterns are adjusted to improve the fidelity of the corresponding output patterns.…”

“…(c) Deep networks can learn and mitigate loading effects during etching processes. (a) Adapted with permission from ref . Copyright 2021 American Chemical Society.…”

Influence of indium tin oxide residues on the electrical performance of hydrogenated amorphous silicon thin-film transistors in the backplane of active-matrix displays

“…S1, ESI. † Orthogonal design, also named as Taguchi design or design of experiment (DOE), 22,23 is a unique method to reveal the interaction mechanisms among multiple parameters and factors in scientific experiments. Orthogonal design is applicable for complex multifactorial experiments.…”

Machine Learning‐Based Shear Optimal Adhesive Microstructures with Experimental Validation

“…Machine learning methods have been implemented in many different fields for optimizing the 3D design of structures in buildings, [ 27 ] ships, [ 28 ] aircraft, [ 29 ] antennae, [ 30 ] and materials. [ 31,32 ] Most of these optimizations used neural networks or genetic algorithms. However, these approaches have significant disadvantages in requiring an extensive training data set with more computational time.…”