Today’s nuclear reactors rely on fission, the splitting of heavy atoms like uranium, to reliably and securely generate about 20% of America’s electricity. Scientists are now looking to the future: a new generation of even safer and more efficient and economical nuclear reactors designed to meet growing energy needs. At the same time, scientists are pursuing the potential use of fusion reactors. Instead of splitting atoms, fusion joins them together, releasing vast amounts of energy and mimicking on Earth the reactions that power the sun and stars.

Helping move these goals closer to reality is an award-winning software package called OpenMC, led by the U.S. Department of Energy’s (DOE) Argonne National Laboratory and the Massachusetts Institute of Technology (MIT). Because OpenMC is open source, scientists across universities, companies and countries can both use and improve the code.

OpenMC, which recently received an R&D 100 Award, predicts how subatomic particles — chiefly neutrons and photons — move through complex systems and interact with materials. It is based on the Monte Carlo method, which uses repeated random sampling to model complex systems. By enabling ​“virtual experiments,” it helps accelerate innovation across nuclear fission and fusion devices before costly prototypes are built.

Paul Romano began developing OpenMC more than a decade ago as an MIT graduate student. He is now a computational scientist at Argonne, and he and others at Argonne and MIT have grown OpenMC into a widely used tool for advancing both fission and fusion research. ​“It can predict, for example, how quickly nuclear fuel will be consumed or how much damage radiation will cause to reactor materials,” Romano said.

Although today’s commercial nuclear fleet is mature, a wave of advanced fission designs aims to improve safety, economics and performance. Because physical prototyping is expensive and highly regulated, developers rely on predictive modeling. ​“OpenMC lets teams de-risk designs computationally,” Romano said. ​“By the time they build hardware, they already understand key behaviors and can demonstrate the advantages of new designs and systems.”

OpenMC has rapidly become a go-to tool in new fusion concepts, with users ranging from national laboratories and universities to a growing roster of private companies. ​“We’re seeing strong adoption across the U.S. and Europe,” Romano said. ​“DOE’s Fusion Energy Sciences support has helped, and private investment is driving additional momentum.”

To strengthen this community, Argonne hosted the OpenMC Application to Tokamak Neutronics Analysis Meeting in May, drawing about 90 international participants from other national laboratories, industry, universities and research organizations. The gathering highlighted the diversity of OpenMC applications and marked a renewed focus on fusion at Argonne.

In the past, Romano explained, such modeling required specialized codes, large teams and huge amounts of computing power. OpenMC offers a new approach: state-of-the-art algorithms combined with ease of use and free access for the global research community. OpenMC also features an interface that makes it quick to build and adjust models. It can run on almost any computer, from laptops to the world’s fastest supercomputers, giving researchers unmatched flexibility.

One of its greatest strengths is its ability to run on high performance computing machines. As part of DOE’s Exascale Computing Project, OpenMC has been optimized for supercomputers capable of a quintillion, or a billion billion, calculations per second (“exascale”). This power is vital for solving the massive, data-heavy problems involved in nuclear and fusion design.

OpenMC has already proven its performance on supercomputers at DOE’s Office of Science user facilities, including two exascale systems: Aurora at the Argonne Leadership Computing Facility and Frontier at DOE’s Oak Ridge Leadership Computing Facility. By combining advanced physics models with exascale computing, researchers can now simulate entire fission reactors and fusion devices at a level of detail not possible before.

From small modular reactors to massive international fusion experiments, OpenMC is helping researchers chart the future of nuclear energy. The software also supports work on used nuclear fuel and radiation protection — whether for power plants, medical facilities or even spacecraft. With its open-source foundation, strong research community and DOE support, OpenMC is poised to remain a cornerstone in the global effort to design the energy systems of the future.

This work is supported by DOE’s Fusion Energy Sciences and Advanced Scientific Computing Research programs. In addition to Romano, Argonne contributors include Patrick Shriwise, John Tramm, Amanda Lund and Joffrey Dorville. MIT contributors include Benoit Forget and Ethan Peterson.