In the quest to reduce the environmental impact of cement, researchers at the Massachusetts Institute of Technology (MIT) have implemented an artificial intelligence system capable of identifying recycled materials that can be integrated into new mixes that create sustainable concrete.
MIT recovers Roman techniques with AI
To achieve this, the team led by Soroush Mahjoubi, together with the Olivetti Group and the Center for Concrete Sustainability (CSHub), has developed a framework of machine learning that analyzes physical and chemical properties of potential cement substitutes. This system scans thousands of pages of scientific literature and classifies more than one million mineral samples, allowing the isolation of materials with high reactive hydraulic or pozzolanic potential.
One of the most promising findings has been the use of ceramic fragments from tiles, bricks and demolished materials. These residues, when ground, present a high reactivity that can strengthen structures without the need for additional processes. The finding echoes techniques used in the concrete of ancient Rome, demonstrating that the intelligent recycling can recover millenary solutions.
Waste sorting paves the way for sustainable concrete
The framework developed by MIT classifies materials into 19 different categories, ranging from industrial by-products to biomass and mining waste. The advantage: many of these materials are globally available and allow for emissions savings without drastically altering current construction processes.
This methodology, in addition to identifying materials, leads to balanced mixtures based on their reactivity with water and their ability to strengthen concrete over time. These advances have been published in an open access article in the journal Nature Communications Materials. This opens the way for extending the use of green concrete in the construction industry on a global scale.
New direction in concrete
Going forward, the team plans to expand the database and conduct experimental tests with the best-rated candidates. The goal is to generate optimized mixes that reduce the use of conventional cement without compromising strength and durability.
As Professor Elsa Olivetti, lead author of the study, points out, this AI-based tool marks a change in the way materials are conceived. Instead of starting from the laboratory, design is now initiated with data, algorithms and sustainability as the fundamental pillars.
Follow us on social networks and don’t miss any of our publications!
YouTube LinkedIn Facebook Instagram X (Twitter) TikTok
Source and photo: MIT
