Deep-Learning-Assisted Focused Ion Beam Nanofabrication.

Content Provider	Europe PMC
Author	Buchnev, Oleksandr Grant-Jacob, James A. Eason, Robert W. Zheludev, Nikolay I. Mills, Ben MacDonald, Kevin F.
Copyright Year	2022
Abstract	Focusedion beam (FIB) milling is an important rapid prototypingtool for micro- and nanofabrication and device and materials characterization.It allows for the manufacturing of arbitrary structures in a widevariety of materials, but establishing the process parameters fora given task is a multidimensional optimization challenge, usuallyaddressed through time-consuming, iterative trial-and-error. Here,we show that deep learning from prior experience of manufacturingcan predict the postfabrication appearance of structures manufacturedby focused ion beam (FIB) milling with >96% accuracy over a rangeof ion beam parameters, taking account of instrument- and target-specificartifacts. With predictions taking only a few milliseconds, the methodologymay be deployed in near real time to expedite optimization and improvereproducibility in FIB processing.
ISSN	15306984
Journal	Nano Letters
Volume Number	22
PubMed Central reference number	PMC9097578
Issue Number	7
PubMed reference number	35324209
e-ISSN	15306992
DOI	10.1021/acs.nanolett.1c04604
Language	English
Publisher	American Chemical Society
Publisher Date	2022-03-24
Access Restriction	Open
Rights License	Permits the broadest form of re-use including for commercial purposes, provided that author attribution and integrity are maintained (https://creativecommons.org/licenses/by/4.0/). © 2022 American Chemical Society
Subject Keyword	nanofabrication deep learning focused ionbeam milling
Content Type	Text
Resource Type	Article
Subject	Nanoscience and Nanotechnology Chemistry Condensed Matter Physics Bioengineering Materials Science Mechanical Engineering