Design Engineering

U of T Machine learning platform accelerates materials discovery

By DE Staff   

Automation General

AI platform aims to minimize resources required in the development of new materials for use in industrial processes.

A computer model of a nanoporous material autonomously designed by an artificial intelligence system developed by U of T and Northwestern University. (Photo credit: University of Toronto)

Researchers at the University of Toronto and Northwestern University announced the development of a machine learning platform that can quickly hone in on the optimal molecular building blocks from which to craft materials for use in targeted industrial applications.

The platform, developed with collaborators from Harvard and the University of Ottawa, is designed to identify optimal web-like frameworks. These frameworks self-assemble from molecular building blocks to form crystalline porous materials that function as molecular “sponges” and may be instrumental in addressing certain challenges.

The problem, however, is that developing these materials, which can facilitate CO2 separation, greenhouse gas reduction and vaccine development, often requires extensive trial-and-error since the molecular building blocks can be in an infinite number of ways.

“Designing reticular materials is particularly challenging, as they bring the hard aspects of modeling crystals together with those of modeling molecules in a single problem,” says senior co-author Alán Aspuru-Guzik, Canada 150 Research Chair in Theoretical Chemistry in the Departments of Chemistry and Computer Science at U of T and Canada CIFAR AI Chair at the Vector Institute. “By using an AI model that can ‘dream’ or suggest novel materials, we can go beyond the traditional library-based screening approach.”


To test their AI platform, the researchers focused on developing metal-organic frameworks (MOFs), which are considered the ideal material for absorbing and removing CO2 from flue gas and other combustion processes.

“We began with the construction of a large number of MOF structures on the computer, simulated their performance using molecular-level modeling and built a training pool applicable to the chosen application of CO2 separation,” said study co-author Randall Snurr, the John G. Searle Professor and chair of the Department of Chemical & Biological Engineering in the McCormick School of Engineering at Northwestern University.

“In the past, we would have screened through the pool of candidates computationally and reported the top candidates,” he adds. “What’s new here is that the automated materials discovery platform developed in this collaborative effort is more efficient than such a “brute force” screening of every material in a database. Perhaps more importantly, the approach uses machine learning algorithms to learn from the data as it explores the space of materials and actually suggests new materials that were not originally imagined.”

According to the research team, whose work was recently published the journal, Nature Machine Intelligence. their AI system shows significant prediction and optimization capability and is fully customizable to address many technology challenges.


Stories continue below

Print this page

Related Stories