AI Breakthrough: Simulating Billions of Atoms to Create Carbon-Neutral, Long-Lasting Concrete

Concrete that lasts centuries and captures carbon? AI just made it possible

Now, researchers at the USC Viterbi School of Engineering have developed a revolutionary AI model that can simulate the behavior of billions of atoms simultaneously, opening new possibilities for materials design and discovery at unprecedented scales. The current state of the world's climate is a dire one. Brutal droughts, evaporating glaciers, and more disastrous hurricanes, rainstorms and wildfires devastate us each year. A major contributor to global warming is the constant emission of carbon dioxide into the atmosphere. Aiichiro Nakano, a USC Viterbi professor of computer science, physics and astronomy, and quantitative and computational biology, was contemplating these issues after the January wildfires in Los Angeles. So, he reached out to longtime partner Ken-Ichi Nomura, a USC Viterbi professor of chemical engineering and materials science practice, with whom he's collaborated for over 20 years. Discussing these issues together helped spark their new project: Allegro-FM, an artificial intelligence-driven simulation model. Allegro-FM has made a startling theoretical discovery: it is possible to recapture carbon dioxide emitted in the process of making concrete and place it back into the concrete that it helped produce. "You can just put the CO2 inside the concrete, and then that makes a carbon-neutral concrete," Nakano said. Nakano and Nomura, along with Priya Vashishta, a USC Viterbi professor of chemical engineering and materials science, and Rajiv Kalia, a USC professor of physics and astronomy, have been doing research on what they call "CO2 sequestration," or the process of recapturing carbon dioxide and storing it, a challenging process. By simulating billions of atoms simultaneously, Allegro-FM can test different concrete chemistries virtually before expensive real-world experiments. This could accelerate the development of concrete that acts as a carbon sink rather than just a carbon source -- concrete production currently accounts for about 8% of global CO2 emissions. The breakthrough lies in the model's scalability. While existing molecular simulation methods are limited to systems with thousands or millions of atoms, Allegro-FM demonstrated 97.5% efficiency when simulating over four billion atoms on the Aurora supercomputer at Argonne National Laboratory. This represents computational capabilities roughly 1,000 times larger than conventional approaches. The model also covers 89 chemical elements and can predict molecular behavior for applications ranging from cement chemistry to carbon storage. "Concrete is also a very complex material. It consists of many elements and different phases and interfaces. So, traditionally, we didn't have a way to simulate phenomena involving concrete material. But now we can use this Allegro-FM to simulate mechanical properties [and] structural properties," Nomura said. Concrete is a fire-resistant material, making it an ideal building choice in the wake of the January wildfires. But concrete production is also a huge emitter of carbon dioxide, a particularly concerning environmental problem in a city like Los Angeles. In their simulations, Allegro-FM has been shown to be carbon neutral, making it a better choice than other concrete. This breakthrough doesn't only solve one problem. Modern concrete only lasts about 100 years on average, whereas ancient Roman concrete has lasted for over 2,000 years. But the recapture of CO2 can help this as well. "If you put in the CO2, the so-called 'carbonate layer,' it becomes more robust," Nakano said. In other words, Allegro-FM can simulate a carbon-neutral concrete that could also last much longer than the 100 years concrete typically lasts nowadays. Now it's just a matter of building it. Behind the scenes The professors led the development of Allegro-FM with an appreciation for how AI has been an accelerator of their complex work. Normally, to simulate the behavior of atoms, the professors would need a precise series of mathematical formulas -- or, as Nomura called them, "profound, deep quantum mechanics phenomena." But the last two years have changed the way the two research. "Now, because of this machine-learning AI breakthrough, instead of deriving all these quantum mechanics from scratch, researchers are taking [the] approach of generating a training set and then letting the machine learning model run," Nomura said. This makes the professors' process much faster as well as more efficient in its technology use. Allegro-FM can accurately predict "interaction functions" between atoms -- in other words, how atoms react and interact with each other. Normally, these interaction functions would require lots of individual simulations. But this new model changes that. Originally, there were different equations for individual elements within the periodic table, with several unique functions for these elements. With the help of AI and machine-learning, though, we can now potentially simulate these interaction functions with nearly the entire periodic table at the same time, without the requirement for separate formulas. "The traditional approach is to simulate a certain set of materials. So, you can simulate, let's say, silica glass, but you cannot simulate [that] with, let's say, a drug molecule," Nomura said. This new system is also a lot more efficient on the technology side, with AI models making lots of precise calculations that used to be done by a large supercomputer, simplifying tasks and freeing up that supercomputer's resources for more advanced research. "[The AI can] achieve quantum mechanical accuracy with much, much smaller computing resources," Nakano said. Nomura and Nakano say their work is far from over. "We will certainly continue this concrete study research, making more complex geometries and surfaces," Nomura said. This research was published recently in The Journal of Physical Chemistry Letters and was featured as the journal's cover image.

New AI simulates 4 billion atoms to build carbon-neutral concrete

Called Allegro-FM, the tool could help create a new generation of smart concrete: stronger, longer-lasting, and capable of capturing carbon dioxide instead of emitting it. Concrete is the most widely used man-made material on Earth, but it comes with a high environmental price. Cement production alone is responsible for roughly 8 percent of global CO₂ emissions. Allegro-FM changes the equation by allowing scientists to test different molecular compositions virtually, accelerating the discovery of carbon-trapping, self-reinforcing formulas without expensive trial-and-error lab work. Nakano explained that the model's simulations suggest it's possible to trap CO₂ released during cement production back into the concrete itself, effectively making it carbon-neutral. "You can just put the CO₂ inside the concrete," he said, "and that makes a carbon-neutral concrete." Developed by professors Aiichiro Nakano and Ken-Ichi Nomura, the AI system simulated over 4 billion atoms with 97.5 percent efficiency using the Aurora supercomputer at Argonne National Laboratory -- a scale nearly 1,000 times larger than conventional models.

AI just designed concrete that could reverse climate damage - Earth.com

Concrete is everywhere - holding up highways, homes, office towers, and schools. It's tough, fire-resistant, and essential to modern infrastructure. But it's also a major problem. Concrete production is responsible for about 8 percent of global carbon dioxide emissions. On top of that, most modern concrete starts to crack and crumble after about 100 years. Now imagine this: concrete that can trap CO₂, make itself stronger, and possibly last thousands of years. That's the idea behind a new AI-powered model created by scientists at the USC Viterbi School of Engineering. This tool simulates billions of atoms at once, offering a new way to design cleaner, longer-lasting materials - starting with concrete. The climate crisis isn't slowing down. Droughts, heat waves, and wildfires continue to get worse. After the January wildfires in Los Angeles, a group of USC researchers started thinking differently about concrete and carbon. What if the very material used to rebuild fire-damaged buildings could also help pull carbon out of the air? That's when their 20-year collaboration turned into a new project: Allegro-FM. "You can just put the CO₂ inside the concrete, and then that makes a carbon-neutral concrete," said Aiichiro Nakano, a USC Viterbi professor of computer science, physics and astronomy, and quantitative and computational biology. Their research focuses on "CO₂ sequestration," the process of capturing carbon dioxide and storing it - ideally within the concrete itself. This isn't easy, but that's where the AI model comes in. Normally, testing new materials means expensive, time-consuming lab work. Allegro-FM changes that. It runs digital experiments with billions of atoms, all simulated in virtual environments. This allows researchers to test different chemical recipes for concrete - searching for mixes that don't just reduce carbon emissions but absorb CO₂ during production. Better yet, the model runs fast and big. On the Aurora supercomputer at Argonne National Laboratory, Allegro-FM simulated over four billion atoms at 97.5 percent efficiency. That's about 1,000 times larger than older models could handle. It's also flexible. Allegro-FM covers 89 different chemical elements and can be used for everything from cement chemistry to long-term carbon storage. "Concrete is also a very complex material. It consists of many elements and different phases and interfaces," said Ken-Ichi Nomura, a USC Viterbi professor of chemical engineering and materials science practice. "Traditionally, we didn't have a way to simulate phenomena involving concrete material. But now we can use this Allegro-FM to simulate mechanical properties [and] structural properties." In a fire-prone city like Los Angeles, where cutting emissions is just as urgent as fire resistance, that combination matters. Simulations show that Allegro-FM can model concrete that does both - stand up to extreme heat and offset its own carbon impact. Beyond the climate benefits, there's another bonus to storing CO₂ in concrete - it may actually make it stronger. "If you put in the CO₂, the so-called 'carbonate layer,' it becomes more robust," Nakano said. That means concrete could potentially last far longer than the current 100-year standard. In fact, the team is thinking about concrete that could rival the durability of ancient Roman structures, some of which have stood for over 2,000 years. Traditionally, simulating materials at the atomic level required heavy mathematics based on quantum mechanics. The process was slow and extremely technical. But machine learning is changing that. "Now, because of this machine-learning AI breakthrough, instead of deriving all these quantum mechanics from scratch, researchers are taking [the] approach of generating a training set and then letting the machine learning model run," Nomura said. That shift lets the model handle more data and more complexity while using fewer resources. Allegro-FM can now predict how atoms interact with one another - a job that used to require countless hours of calculation. The result? Faster simulations, broader material options, and huge efficiency gains. According to Nakano, the AI can achieve quantum mechanical accuracy with much, much smaller computing resources. "We will certainly continue this concrete study research, making more complex geometries and surfaces," Nomura said. The big picture is this: AI isn't just helping us understand the world at the atomic level - it's helping us rethink the materials we rely on every day. And in this case, that could mean stronger buildings, cleaner air, and a future where concrete becomes part of the climate solution instead of the problem. The full study was published in the journal The Journal of Physical Chemistry Letters. -- - Like what you read? Subscribe to our newsletter for engaging articles, exclusive content, and the latest updates.