MIT has been chosen by the U.S. Department of Energy's National Nuclear Security Administration to launch a new research center aimed at simulating some of the harshest physical environments ever studied.
The project is called CHEFSI -- short for the Center for the Exascale Simulation of Coupled High-Enthalpy Fluid-Solid Interactions. It will bring together researchers working at the edges of computing, materials, and applied science to model extreme scenarios that are difficult, and sometimes impossible, to recreate in physical testing.
The center is funded through the DOE's Predictive Science Academic Alliance Program IV. One of its main goals is to improve how scientific data gets turned into usable, predictive insight. That includes developing new tools that blend AI with exascale computing, while also building strong ties with national labs to share data and verify results. Much of the work will connect directly to systems used in national security, aerospace, and defense.
The research effort cuts across departments. Teams from mechanical and aerospace engineering, materials science, computing, and applied math will all be involved. The situations they're studying involve more than just heat or speed -- they require simulating rapid, layered changes in materials under very high stress. This is the kind of work where physics, chemistry, and computation all overlap, and no single area can cover it alone.
One of the key challenges will be figuring out how materials behave when they are pushed far beyond their normal limits. Spacecraft reentry, for example, isn't just about staying intact. It's about how heat moves through layers, how surfaces erode, and how all of that unfolds in real time. The team at CHEFSI will be working to build models that can make sense of these conditions and help others design systems that hold up under pressure -- literally and figuratively.
"CHEFSI will capitalize on MIT's deep strengths in predictive modeling, high-performance computing, and STEM education to help ensure the United States remains at the forefront of scientific and technological innovation," says Ian A. Waitz, MIT's vice president for research. "The center's particular relevance to national security and advanced technologies exemplifies MIT's commitment to advancing research with broad societal benefit."
CHEFSI is one of five new Predictive Simulation Centers funded through PSAAP-IV, joining other university-led efforts focused on modeling extreme events like combustion instability and dynamic material failure. Each center contributes to a shared goal: building more accurate and reliable simulations for high-stakes national security challenges.
Much of the real work at CHEFSI will start with data. Without the right kind of inputs, even the best simulations won't tell you much. The materials, the heat conditions, the fluid dynamics -- all of it has to be grounded in information pulled from experiments, past studies, and specialized testing setups. That data needs to be cleaned, structured, and sorted before it ever gets used to train a model or run a simulation.
A big part of this will come from national lab partnerships. Teams at Lawrence Livermore, Los Alamos, and Sandia have been collecting data on extreme environments for years, and CHEFSI will work closely with them to make use of it. The goal isn't just to run simulations -- it's to compare those results against something real and keep adjusting as new information comes in. That kind of back and forth will help the models get better over time.
AI tools will play a role too. Some of the models CHEFSI builds will use AI to fill in gaps or simplify specific parts of a problem. These aren't full replacements for traditional simulations, but they make it easier to test things quickly. Still, that only works if the training data is solid. One bad set can throw everything off, so part of the job is making sure the data is trustworthy from the start.
Students and early-career researchers will also get hands-on experience with this. They'll learn how to work with large datasets, make sense of inconsistencies, and trace how small choices in data handling affect big outcomes. That kind of training matters just as much as the code itself.
"By integrating high-fidelity physics models with artificial intelligence-based surrogate models, experimental validation, and state-of-the-art exascale computational tools, CHEFSI will help us understand and predict how thermal protection systems perform under some of the harshest conditions encountered in engineering systems," says Raúl Radovitzky, the Jerome C. Hunsaker Professor of Aeronautics and Astronautics, associate director of the ISN, and director of CHEFSI. "This knowledge will help in the design of resilient systems for applications ranging from reusable spacecraft to hypersonic vehicles."
With its mix of data-driven modeling, next-generation computing, and real-world validation, CHEFSI is positioned to shape how the next decade of materials and aerospace research gets done -- not just at MIT, but across the entire field.