Mohamad Moosavi: Accelerating the search for climate solutions with AI

2026-05-25 · Source: Vector Institute for Artificial Intelligence · Field: Science & Research — Engineering & Applied Sciences, Environmental Science & Earth Systems, Mathematics & Computational Sciences · Depth: Advanced, short

Summary

Mohamad Moosavi, an Assistant Professor at the University of Toronto and a Vector Institute Faculty Member, is accelerating climate solution discovery by applying AI to materials science. His research focuses on transforming the design of materials like metal-organic frameworks (MOFs), which were recognized with the 2025 Nobel Prize in Chemistry for their potential in clean energy, water, and air. Historically, synthesizing over 120,000 MOF variations took 20 years to yield only one carbon capture material under specific conditions. Moosavi's team uses deep learning to treat molecules as mathematically differentiable, continuous variables, a breakthrough that allows powerful engineering optimization techniques to be applied to chemical systems. This enables computational navigation of vast material spaces, identifying promising candidates rapidly and potentially reducing development timelines from decades to months. He joined Vector in 2023 as a Faculty Affiliate, becoming a Faculty Member in 2025, contributing to Toronto's interdisciplinary AI for science ecosystem.

Key takeaway

For AI Scientists and Research Scientists focused on sustainable technologies, applying deep learning to materials science offers a paradigm shift. You should explore methods that treat molecular structures as continuous variables, enabling advanced optimization techniques previously impossible in chemistry. This approach can significantly reduce decades-long timelines for developing new materials. It accelerates your project's impact, potentially leading to breakthrough innovations in months for carbon capture or energy storage. Consider collaborating within interdisciplinary ecosystems like Toronto's to maximize these opportunities.

Key insights

Deep learning enables treating molecules as differentiable, continuous variables, revolutionizing materials discovery for climate solutions.

Principles

• AI can dramatically accelerate materials discovery timelines.
• Interdisciplinary collaboration is crucial for innovation.
• Treating molecules as continuous variables unlocks new optimization.

Method

Topological deep learning models learn the "grammar" and "syntax" of molecular structures, encoding material language for computational navigation and rapid identification of promising candidates.

In practice

• Apply deep learning to model molecular structures.
• Utilize computational navigation for material design.
• Integrate AI with chemical engineering workflows.

Topics

• AI for Science
• Materials Discovery
• Deep Learning
• Metal-Organic Frameworks
• Carbon Capture
• Sustainable Technologies
• Molecular Differentiability

Best for: AI Scientist, Research Scientist, Director of AI/ML

Related on AIssential

• 🔬Searching the Space of All Possible Materials — Prof. Max Welling, CuspAI — AI for science is an exploding field, combining deep scientific problems with high-impact technological solutions.
• Predicting Scale-Up of Metal-Organic Framework Syntheses with Large Language Models — LLMs fine-tuned with PU learning can predict MOF synthesis scalability from small-scale literature data.
• 🔬Why There Is No "AlphaFold for Materials" — AI for Materials Discovery with Heather Kulik — Deep integration of domain expertise with AI and critical assessment of model outputs are crucial for success in AI for science.
• Closed-loop AI framework could shorten clean energy materials development by years — A closed-loop AI framework can significantly accelerate clean energy materials development, reducing time and cost.
• The evolution of AI from image interpretation toward scientific inference in nanoparticle electron microscopy — AI is evolving electron microscopy from descriptive imaging to data-driven scientific inference and autonomous materials discovery.
• Accelerating science with AI and simulations — AI is at a second inflection point, integrating language and multi-modal reasoning to accelerate scientific discovery.

See Counsel's argued verdicts on the open AI decisions leaders are weighing →

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Vector Institute for Artificial Intelligence.