On the left is a platinum nanoparticle, imaged via electron microscopy. The data have sufficient spatial resolution to display individual atoms. However, these images are heavily corrupted by noise due to the high-temporal resolution, which is nonetheless necessary to visualise fundamental dynamic behaviour on the nanoparticle surface associated with its functionality. At right is the output of an AI system, which is able to effectively remove the noise and reveal the atomic structure of the nanoparticle. Credit: Courtesy of Arizona State’s Peter Crozier and Joshua Vincent and NYU’s Carlos Fernandez-Granda and Sreyas Mohan.
AI allows new visualisation of how nanoparticles, the building blocks of pharmaceuticals, electronics, and industrial materials, behave.
A team of scientists has developed a method to illuminate the dynamic behaviour of nanoparticles, which are foundational components in the creation of pharmaceuticals, electronics, and industrial and energy-conversion materials.
The advance combines artificial intelligence with electron microscopy to render visuals of how these tiny bits of matter respond to stimuli.
Carlos Fernandez-Granda, director of NYU’s Center for Data Science and a professor of mathematics and data science, and one of the paper’s authors, said: “Nanoparticle-based catalytic systems have a tremendous impact on society
“It is estimated that 90% of all manufactured products involve catalytic processes somewhere in their production chain.
“We have developed an artificial-intelligence method that opens a new window for the exploration of atomic-level structural dynamics in materials.”
Blend
The work, which also included researchers from Arizona State University, Cornell University, and the University of Iowa, blends electron microscopy with AI to enable scientists to see the structures and movements of molecules that are one-billionth of a metre in size at an unprecedented time resolution.
Peter A Crozier, a professor of materials science and engineering at Arizona State University and one of the paper’s authors, said: “Electron microscopy can capture images at a high spatial resolution, but because of the velocity at which the atomic structure of nanoparticles changes during chemical reactions, we need to gather data at a very high speed to understand their functionality.
“This results in extremely noisy measurements.
“We have developed an artificial-intelligence method that learns how to remove this noise – automatically – enabling the visualisation of key atomic-level dynamics.”
Observing the movement of atoms on a nanoparticle is crucial to understand functionality in industrial applications.
The problem is that the atoms are barely visible in the data, so scientists cannot be sure how they are behaving – the equivalent of tracking objects in a video taken at night with an old camera.
To address this challenge, the paper’s authors trained a deep neural network, AI’s computational engine, that is able to ‘light up’ the electron-microscope images, revealing the underlying atoms and their dynamic behaviour.
New statistic
David S Matteson, a professor and associate chair of Cornell University’s Department of Statistics and Data Science, director of the National Institute of Statistical Sciences, and one of the paper’s authors, said: “The nature of changes in the particle is exceptionally diverse, including fluxional periods, manifesting as rapid changes in atomic structure, particle shape, and orientation; understanding these dynamics requires new statistical tools.
“This study introduces a new statistic that utilises topological data analysis to both quantify fluxionality and to track the stability of particles as they transition between ordered and disordered states.”
The study is reported in the journal Science.
