Automation of Microscopy Experiments to Explore Functional Properties of Materials at Nanoscale.

Automated microscopy explores functional properties at the nanoscale

Scientists at the Department of Energy’s Oak Ridge National Laboratory (ORNL) are programming microscopes to stimulate discoveries using an intuitive algorithm, formulated at ORNL’s Center for Nanophase Materials Sciences (CNMS), that could facilitate advances in new materials for sensing, energy technologies and information technology.

A smart approach to microscopy and imaging developed at Oak Ridge National Laboratory could guide material discoveries for future technologies. Image credit: Adam Malin / ORNL, United States Department of Energy.

There are so many potential materials, some of which we cannot study with conventional tools at all, that they need more efficient and systematic approaches to design and synthesize. We can use intelligent automation to access unexplored materials and create a shareable and reproducible path to discoveries that were not previously possible.

Maxim Ziatdinov, Division of Computational Sciences and Engineering and CNMS, ORNL

The method, reported in Intelligence of the machine of natureintegrates machine learning and physics to automate microscopy-based experiments designed to explore the functional properties of materials at the nanoscale.

Functional materials are receptive to stimuli such as electricity or heat and are designed to support ordinary and future technologies, ranging from solar cells and computers to shape memory materials and artificial muscles.

Their unique properties are related to atomic structures and microstructures that can be perceived with advanced microscopy. However, the challenge has been to create efficient techniques for finding regions of interest from which these properties can be examined.

Scanning probe microscopy is a fundamental tool for studying the structure-property relationships of functional materials. The instruments scan the surface of a material with an atomically sharp probe to trace the structure at the nanoscale, the length of one billionth of a meter.

They can also detect responses to a variety of stimuli, offering insights into the fundamental mechanisms of polarization switching, plastic deformation, electrochemical reactivity, or quantum phenomena.

Microscopes today can perform a dot-by-dot scan of a square grid (nanometer size), but the process can be laboriously slow, with measurements collected in days for just one material.

“Interesting physical phenomena often occur only in a small number of spatial places and are linked to specific but unknown structural elements. While we generally have an idea of ​​what the characteristics of the physical phenomena we aim to discover will be, efficiently identifying these regions of interest is a major bottleneck “, said former ORNL CNMS researcher and lead author Sergei Kalinin, now at the University of Tennessee, Knoxville.

“Our goal is to teach microscopes to search for regions with interesting physics actively and in a much more efficient way than doing a grid search,” Sergei Kalinin added.

Researchers have turned to machine learning and artificial intelligence (AI) to overcome this challenge, but traditional algorithms require huge human-coded datasets and may not save time.

For a smarter approach to automation, the ORNL workflow integrates human-based physical reasoning into machine learning approaches and uses miniature datasets – images taken from less than 1% of the sample – as a preliminary point. . The algorithm chooses points of interest based on what it learns within the experiment and knowledge beyond the experiment.

As a proof of concept, a workflow using scanning probe microscopy was shown and applied to well-explored ferroelectric materials. Ferroelectrics are a type of functional material that possesses a reorientable surface charge that can be exploited for actuation, computation and sensing applications.

Researchers are eager to understand the connection between the amount of energy or data these materials can store and the local domain structure that controls this property. The automated experiment detected the precise topological defects for which these factors increased.

The point is that the workflow was applied to material systems familiar to the research community and made a fundamental discovery, something that was not previously known, very quickly, in this case, within hours.

Maxim Ziatdinov, Division of Computational Sciences and Engineering and CNMS, ORNL

Results have been faster, by orders of magnitude, than traditional workflows and represent a new path in intelligent automation.

We wanted to move away from training computers solely on data from previous experiments and instead teach computers how to think like researchers and learn on the fly. Our approach is inspired by human intuition and recognizes that many material discoveries have been made through the trial and error of researchers who rely on their expertise and experience to guess where to look.

Maxim Ziatdinov, researcher, division of computational sciences and engineering and CNMS, ORNL

Yongtao Liu of ORNL was tasked with the technical challenge of running the algorithm on an operating microscope at the CNMS.

This is not a standard feature and a lot of work is required to connect the hardware and software. We have focused on scanning probe microscopy, but the setup can be applied to other experimental imaging and spectroscopy approaches accessible to the broader user community.

Yongtao Liu, researcher, CNMS, ORNL

The research received support from the CNMS, which is a user facility of the DOE Office of Science, and the Center for 3D Ferroelectric Microelectronics, which is an Energy Frontier Research Center led by Pennsylvania State University and aided by the DOE Office of Science.

Journal reference:

Liu, Y., et al. (2022) Experimental discovery of structure-property relationships in ferroelectric materials through active learning. Intelligence of the machine of nature. doi.org/10.1038/s42256-022-00460-0.

Source: https://www.ornl.gov

Leave a Comment

Your email address will not be published.