Materials Modelling: A mathematical study of nuclear materials

When a radioactive element’s half-life is measured in hundreds of thousands of years, studying its behaviour in real-time is out of the question.

But when that element is found in radioactive waste, it’s essential that we understand exactly how it will behave over long periods of time following encapsulation in a material for long-term storage; will it remain locked and safe within the storage material for as long as is needed?

To answer that question within something approaching a human lifespan, scientists at the Australian Nuclear Science and Technology Organisation (ANSTO) are build- ing virtual versions of the materials found in radioactive waste.

But these aren’t just digital versions of the ball-and-stick models from high-school chemistry; these are mathematical calculations of energy and structural changes at the level of electrons, and they are so complex they require the supercomputing power of MASSIVE to model just a few hundred atoms.

“It’s got to calculate the energy of the atomic structure, and then it will move the atoms around a bit and see how that energy can change,” says Dr Eugenia Kuo, a physicist at ANSTO.

The idea is to calculate the lowest energy state of a material—because that’s the state that any material will head to—then calculate what that structure might look like.

Snapshots in time (L to R) from a molecular dynamics simulation of radiation damage in uranium dioxide (UO2), a nuclear fuel. The black atoms are uranium, the red ones oxygen. The snapshots include about 10,000 atoms from a total of about half a million. L: large damaged area at the time of radiation; R: structural changes after radiation.

Snapshots in time (L to R) from a molecular dynamics simulation of radiation damage in uranium dioxide (UO2), a nuclear fuel. The black atoms are uranium, the red ones oxygen. The snapshots include about 10,000 atoms from a total of about half a million. L: large damaged area at the time of radiation; R: structural changes after radiation.

For example, technetium-99 is one of the most challenging components of spent nuclear fuel rods. Over time, technetium-99 decays into ruthenium. The question is then how that decay impacts the chemistry of the

material technetium-99 is encapsulated in: how stable is the process, and how stable is the end product?

“We’re trying to understand the local interactions and how that affects the stability of the structure that we’re interested in,” Kuo says.

Other materials being investigated with the use of MASSIVE supercomputers are those used inside the reactor itself.

The inside of a nuclear reactor is one of the most complex engineering environments that exist. Materials used in reactors, such as the ceramic fuel cells, have to be able to withstand intense radiation without being compromised or changed in such a way that might risk their integrity.

But studying that in situ is essentially impossible, so researchers are turning to computer-assisted simulations.

“We use MASSIVE’s computer cluster to do computer simulations of a material’s proper- ties so we can get a better understanding of how these structures change at the atomic level when a reactor is operational, and how the change of structures is going to change the properties,” says Dr Meng Jun Qin, a physicist at ANSTO.

These simulations model the molecular dynamics of millions of atoms, with the goal of finding or developing materials that will be as radiation-tolerant as possible.