Materials that matter

Open this photo in gallery:

New materials are critical to human progress, from the stone-age tools that advanced civilization to the semiconductors that underpin today’s digital devices, but discovering them can be slow and expensive.

Now the University of Toronto’s Acceleration Consortium is working to speed up that process through self-driving labs that draw on expertise in artificial intelligence (AI), automation, chemistry, advanced computing, materials science and economics.

“Advanced materials have the potential to transform our lives and our world for the better, and we can make them faster and cheaper,” says Alán Aspuru-Guzik, a professor of chemistry and computer science at U of T and director of the consortium. The global coalition of academia, industry and government was formed last year to harness advances in materials science, robotics and AI to address urgent challenges related to areas such as climate change, health, urbanization and economic prosperity.

The consortium is one of more than 20 Institutional Strategic Initiatives – interdisciplinary research networks bringing experts together from across U of T and beyond to collaborate on complex global issues.

“We are trying to do scientific discovery differently,” says Dr. Aspuru-Guzik. The Canada 150 Research Chair in Theoretical and Quantum Chemistry feels it’s possible to accelerate the time and reduce the cost to discover new materials and molecules by a factor of 10. That means where the discovery of an interesting material once might have required 30 years and cost tens of millions of dollars, it could now take just a couple of years and cost a million dollars.

“We do that by using a combination of artificial intelligence, automation, chemicals and materials methods,” explains Dr. Aspuru-Guzik, noting that in these self-driving labs, “AI is controlling the next experiment you’re going to be making the same way AI is steering the wheel of a self-driving car.”

He says the consortium has so far built a roster of international scientists, created a post-doctoral program that’s brought candidates from around the world to do research at U of T and launched a call for proposals for research sponsored by industry aimed at understanding questions related to accelerated discovery.

“Exciting research is now being carried out,” he says, noting that the consortium has recruited international experts to U of T, is building four floors of new lab space and recently held its first annual Accelerate Conference there. The event brought together more than 200 people, featuring talks and panels with some 60 experts shaping the emerging field of accelerated science.

U of T is the globally facing consortium’s home base and primary sponsor “and is actively supporting it at the highest level,” Dr. Aspuru-Guzik says. “U of T has a very special place in the field of artificial intelligence because deep learning was essentially invented here. U of T recognizes that role. And therefore, as we think about applications of machine learning for the future, U of T is deeply invested in using machine learning in new areas.”

Open this photo in gallery:

For discoveries to happen faster, it’s critical to have an AI ecosystem with “investors, startups, large corporations and industry all working together,” he comments. “The consortium aims to bring together all those pieces and make Toronto a leader of AI for materials discovery.”

“Over the coming years, advanced materials are expected to yield massive breakthroughs in medicine, energy, transportation, construction and electronics and blossom into a $1-trillion industry,” says Leah Cowen, vice-president, Research and Innovation, and Strategic Initiatives. “U of T is thrilled to be leading the way.”

Dr. Aspuru-Guzik points out that in the past, the time lag between making a discovery and a product reaching the market has been about 30 years. For example, music CDs came out decades after lasers were discovered in the 1960s. “That’s universal for all materials, because all the processes are disconnected.”

Once materials are developed through accelerated science, it will still be necessary to think about scaling up and commercialization. Examples of products Dr. Aspuru-Guzik expects to see come out of self-driving labs include greener and cheaper batteries to store the world’s energy supply, resistance-evasive antibiotics and better coatings for preventing rust in metals.

There is wide recognition about the positive impact that can be achieved by accelerating the application of these advanced materials, he says, and this informs efforts to commercialize them “at a faster rate, because if discoveries are happening quickly, then a lot of investment is going to be poured into them.”

Advertising feature produced by Randall Anthony Communications. The Globe’s editorial department was not involved.