Business is booming.

AI breakthrough paves way for durable carbon-neutral concrete production

 

By Abbas Nazil

Researchers at the University of Southern California have developed a groundbreaking artificial intelligence model capable of simulating billions of atoms simultaneously, potentially transforming the construction industry by enabling the creation of carbon-neutral concrete.

The AI model, Allegro-FM, can simulate the behavior of more than four billion atoms at once with 97.5 percent efficiency, a computational leap nearly 1,000 times beyond conventional approaches.

Published in The Journal of Physical Chemistry Letters and featured on the cover, this innovation offers the ability to test concrete chemistries virtually before proceeding with expensive real-world experiments.

Concrete production currently accounts for approximately 8 percent of global carbon dioxide emissions.

However, the new model demonstrates the potential to recapture CO₂ emitted during concrete production and reintegrate it into the final product, making the material carbon-neutral.

This CO₂ sequestration process, long considered a challenge, is now theoretically feasible thanks to this high-powered simulation model.

The research was led by Aiichiro Nakano, professor of computer science, physics, and computational biology, and Ken-Ichi Nomura, professor of chemical engineering and materials science.

They were joined by colleagues Priya Vashishta and Rajiv Kalia, all of USC Viterbi School of Engineering.

Their goal is to design more durable, sustainable materials in response to the increasing environmental toll of climate change.

The inspiration came in the aftermath of the January wildfires in Los Angeles, prompting the team to explore ways of making concrete not only more resilient to disasters but environmentally beneficial.

In their simulations, Allegro-FM revealed that incorporating CO₂ into concrete could enhance its robustness, with the potential to exceed the typical 100-year lifespan of modern concrete, possibly achieving the longevity seen in ancient Roman structures.

Traditionally, simulating complex materials like concrete was hindered by the need to model quantum mechanics for each atom type separately.

With Allegro-FM, the AI leverages machine learning to train on atomic interactions, reducing computational load while achieving quantum-level accuracy.

The model supports simulations involving 89 chemical elements, covering a vast array of material possibilities, from construction to energy storage.

Its scalability allows researchers to simulate large-scale materials like cement while also exploring smaller molecular systems.

Nakano emphasized that this method bypasses the time-consuming mathematical derivations required by older models.

Nomura highlighted the efficiency gains, noting that what once demanded massive supercomputing resources can now be handled with a fraction of the power thanks to AI.

The team envisions continuing their research with more complex concrete geometries and surface interactions, moving closer to real-world applications.

Allegro-FM marks a major advancement in materials science, positioning AI as a key tool in achieving carbon neutrality in critical infrastructure.

below content

Quality journalism costs money. Today, we’re asking that you support us to do more. Support our work by sending in your donations.

The donation can be made directly into NatureNews Account below

Guaranty Trust Bank, Nigeria

0609085876

NatureNews Online

This website uses cookies to improve your experience. We'll assume you're ok with this, but you can opt-out if you wish. Accept Read More