Every year, thousands of new materials are created, yet many never reach their full potential because their applications aren't immediately obvious - a challenge University of Toronto researchers aim to address using artificial intelligence.
In a study published in Nature Communications , a team led by Faculty of Applied Science & Engineering researcher Seyed Mohamad Moosavi introduced an AI tool that can predict how well a new material might perform in real-world scenarios - right from the moment it's synthesized. The system focuses on a class of porous materials known as metal-organic frameworks (MOFs), which have tunable properties and a wide range of potential applications.
Moosavi notes that materials scientists created more than 5,000 different types of MOFs last year alone, underscoring the scale of the challenge.
"In materials discovery, the typical question is, 'What is the best material for this application?'" says Moosavi, an assistant professor of chemical engineering and applied chemistry. "We flipped the question and asked, 'What's the best application for this new material?' With so many materials made every day, we want to shift the focus from 'What material do we make next?' to 'What evaluation should we do next?'"
MOFs can be used, for example, to separate CO2 from other gases in waste streams, preventing the carbon from reaching the atmosphere and contributing to climate change. They can also be used to deliver drugs to specific areas of the body, or to enhance the functionality of electronic devices.
Often, an MOF created for one purpose turns out to have ideal properties for a completely different application. Moosavi cites a previous study in which a material originally synthesized for photocatalysis was later found to be highly effective for carbon capture - but only seven years after its creation.
The new AI-powered approach aims to reduce this time lag between discovery and deployment.
To achieve this, PhD student Sartaaj Khan developed a multimodal machine learning system trained on various types of data typically available immediately after synthesis - specifically, the precursor chemicals used to make the material and its powder X-ray diffraction (PXRD) pattern.
"Multimodality matters," says Khan. "Just as humans use different senses - such as vision and language - to understand the world, combining different types of material data gives our model a more complete picture."
The AI system uses a multimodal pretraining strategy to gain insights into a material's geometry and chemical environment, enabling it to make accurate property predictions without requiring post-synthesis structural characterization. This can accelerate the discovery process and help researchers identify promising materials before they're overlooked or shelved.
To test the model, the team conducted a "time-travel" experiment: they trained the AI on material data available before 2017 and asked it to evaluate materials synthesized afterward. The system successfully flagged several materials - originally developed for other purposes - as strong candidates for carbon capture. Some of those are now undergoing experimental validation in collaboration with the National Research Council of Canada .
Looking ahead, Moosavi plans to integrate the AI into the self-driving laboratories (SDLs) at U of T's Acceleration Consortium , a global hub for automated materials discovery and one of several U of T institutional strategic initiatives .
"SDLs automate the process of designing, synthesizing and testing new materials," he says.
"When one lab creates a new material, our system could evaluate it - and potentially reroute it to another lab better equipped to assess its full potential. That kind of seamless inter-lab co-ordination could accelerate materials discovery."
