AI Stirs Up Concrete Recipe in MIT Study, Promising Greener Construction
Cambridge, MA – In a groundbreaking study, researchers at MIT’s Olivetti Group and the Concrete Sustainability Hub (CSHub) have demonstrated how artificial intelligence can revolutionize concrete production, making it more sustainable and cost-effective. The team’s innovative approach addresses the urgent need to reduce cement usage in concrete, a major contributor to global carbon emissions.
The research, published in Nature’s Communications Materials, details a machine-learning framework that sifts through vast amounts of data on potential materials to identify suitable cement replacements. “We realized that AI was the key to moving forward,” explains postdoc Soroush Mahjoubi, who led the team. “There is so much data out there on potential materials — hundreds of thousands of pages of scientific literature. Sorting through them would have taken many lifetimes of work.”
The AI framework evaluates materials based on key properties like hydraulic reactivity (the ability to harden when exposed to water) and pozzolanicity (the ability to react with calcium hydroxide to strengthen concrete over time). By analyzing over 1 million rock samples and countless scientific papers, the AI identified 19 types of promising materials, ranging from biomass to mining byproducts and demolished construction materials.
One of the most exciting findings is that many of these materials, such as ceramics from old tiles and bricks, can be incorporated into concrete mixes with minimal processing. This allows for significant emissions and cost savings.
“Some of the most interesting materials that could replace a portion of cement are ceramics,” notes Mahjoubi. “Old tiles, bricks, pottery — all these materials may have high reactivity. That’s something we’ve observed in ancient Roman concrete, where ceramics were added to help waterproof structures.”
This approach promotes a circular economy by repurposing waste materials that would otherwise end up in landfills. The team plans to further refine the AI framework and experimentally validate the most promising material candidates.
“Concrete is the backbone of the built environment,” says Randolph Kirchain, co-author and CSHub director. “By applying data science and AI tools to material design, we hope to support industry efforts to build more sustainably, without compromising on strength, safety, or durability.”
Professor Elsa Olivetti, senior author of the study, emphasizes the potential of AI to accelerate materials research: “AI tools have gotten this research far in a short time, and we are excited to see how the latest developments in large language models enable the next steps.”
