Home » Big Science, Stories

Your Mileage May Vary

November 3, 2011

How old is too old for America’s nuclear reactors?

By Whitney Heins

Not long ago, Brian Wirth, Governor’s Chair for computational nuclear engineering, was driving his wife’s car when something catastrophic happened under the hood: One of the fan blades shattered, spewing polycarbonate shrapnel all around the engine compartment.

Naturally, Wirth began wondering about the fan blade’s operating lifetime—natural because examining the operating lifetime of materials in nuclear reactors is what he does on a daily basis.

His work is of critical importance now more than ever, as many reactors have aged past their initially regulated operating lifetime of forty years. In the past decade or so, sixty-six of America’s 104 nuclear reactors have been issued license renewals, according to the Associated Press. Adding fuel to the fire (or reactor), the predicted lifetime of forty years is now being questioned as nuclear industry officials contend the number originated from when construction loans were to be amortized or from an arbitrary Congressional compromise over operations of a plant.

“In my mind, forty years was really an arbitrary number,” says Wirth. “There are very complicated, multi-variable, synergistic interactions that influence the judgment call of how long we can safely operate these reactors, and I will probably spend the bulk of my remaining career continuing to understand and predict this. But I know it will not be a blanket answer.”

Much like automobiles, nuclear reactors have different components with different expiration dates. Some are easily and frequently replaced, such as spark plugs and fan belts for cars—or nuclear fuel assemblies and piping for nuclear reactors.

Much like automobiles, nuclear reactors have different components with different expiration dates. Some are easily and frequently replaced, such as spark plugs and fan belts for cars—or nuclear fuel assemblies and piping for nuclear reactors.

To understand why it will not be a blanket answer, we must get back in the car with Wirth and the exploding fan blade. Automobiles and nuclear reactors have many different components, with different expiration dates. Some are easily and frequently replaced, such as spark plugs and fan belts for cars or nuclear fuel assemblies and above-ground piping for nuclear reactors. But what is going to determine whether you replace the car—or the nuclear reactor—are the more serious failures such as a crack in the engine block or, for the reactor, a crack in the reactor pressure vessel.

For a car, variables such as driving habits and weather can cause variation in the lifetime of its components, just as history dependence and elements in a pressure vessel’s steel do in reactors. In both cases, operators weigh the costs of repairs versus replacements when it comes to failures. Engines and vessels are costly repairs. Wirth estimates vessels to cost upwards of $500 million—although one has never been replaced.

The reactor pressure vessel is a 100-ton steel container, eight to ten inches thick, twenty to forty feet high, and thirty to sixty feet in diameter. It consists of many steel plates welded together and sits inside a concrete dome. This is the heart of the nuclear reactor, containing the coolant and reactor core, where the nuclear reaction occurs.

According to Wirth, it is one of only four components in the nuclear reactor for which scientists have not been able to determine an operating lifespan—and of those four, it is arguably the most critical.

“It is the number-one, safety-critical component in the power plant, and it is not easy to replace because it is embedded in a big concrete dome, has insulation on it, and has inlet and outlet piping attached to it,” Wirth explains. “So this is the component that will ultimately, in my mind, determine what the safe operating lifetime is.”

Brian Wirth’s discoveries are driving research for reactor performance improvements and the development of new reactor materials.

Brian Wirth’s discoveries are driving research for reactor performance improvements and the development of new reactor materials.

Understandably, Wirth is dedicating the bulk of his research trying to determine the operating lifespan of these vessels.

It’s no easy feat. Wirth is tasked with trying to follow the evolution of defects created from neutron–atom collisions that proceed incredibly fast—within ten trillionths of a second—but then evolve over decades. The big question he is trying to answer: What causes properties in materials to change and ultimately fail?

To help illustrate his challenge, Wirth takes a paper clip out of his desk and starts to bend it.

“When you bend a paper clip back and forth, it initially makes the paper clip harder to bend as this movement continues, but the paper clip then becomes brittle and breaks,” he says. “This weakening or degradation is a significant concern. We are trying to better understand this process in order to make material adjustments, to prolong the life of materials, and to prevent disasters and failures.”

In reality, the problem is when neutrons released during uranium fissions escape the nuclear reactor vessel and penetrate the fuel and cladding. Eventually they pass into the structural materials and cause degradation of their mechanical performance.

“Some of the neutrons that are used to split the uranium atoms actually leak out of the vessel’s core and pass through the vessel. When they do that, they induce violent collisions with atoms in the solid that cause defects that, over decades, cause changes in the internal structure, much like those during the bending of the paper clip,” said Wirth. “But we can’t tell this until it breaks, so we are trying to see the defects caused by the neutrons, how they evolve, and how they change properties.”

Conducting research for the Department of Energy (DOE), Wirth works with innovative tools to determine how long a pressure vessel can safely operate with neutrons passing through it.

Wirth and his team are leveraging the power of UT’s Newton Computer Complex with the Kraken and Jaguar supercomputers at ORNL.

Wirth and his team are leveraging the power of UT’s Newton Computer Complex with the Kraken and Jaguar supercomputers at ORNL.

Wirth and his team are leveraging the power of UT’s Newton Computer Complex with the Kraken and Jaguar supercomputers located at ORNL to integrate computational multi-scale materials modeling with experimental processing, environmental exposure, and property characterization to reveal the dynamics of materials.

They also are using very low-energy neutrons to characterize precipitates in materials by neutron-scattering techniques at the ORNL High Flux Isotope Reactor and plan to use the lab’s Spallation Neutron Source, as well.

The next goal is to develop models to understand all the dynamic processes that cause the interior of the material to change at the micron scale, how that impacts the strength of the material, and whether it is going to fail under accident loading conditions.

Wirth’s models must accelerate an already quick process to predict what would happen if reactors are used for sixty or eighty years. The ultimate goal is to develop predictive performance models to inform the Nuclear Regulatory Commission (NRC), utilities, and the DOE on what the lifetime should be.

In the meantime, Wirth is using his discoveries about the processes and degradation mechanisms to drive research for reactor performance improvements and the development of new reactor materials. He currently serves as deputy focus area lead for materials performance and optimization within one of three DOE hubs—launched by Energy Secretary Steven Chu—called the Consortium for Advanced Simulation of Light Water Reactors (CASL).

CASL is a large-scale, $25 million-per-year effort to team up across national laboratories, universities, and industry in an attempt to model the performance of nuclear reactors. Wirth uses the high-performance computing capability of Jaguar to simulate nuclear fuel performance and the performance of nuclear reactors within a virtual reactor to better understand operating safety limits with the goal of optimizing power usage.

While there is still a lot to learn about a nuclear reactor’s lifespan, Wirth does not believe America’s reactors are anywhere near suffering the fate of his wife’s car’s fan blade.

Like cars, some reactors may last longer than others, but there is no predictive model for determining that yet. Wirth does say he feels pretty confident that the nation’s current reactors can last sixty years and possibly longer, but he has a lot of work to do before he can arrive at a precise answer.

“I believe the NRC does a good job at independently assessing reactors,” Wirth says. “Of course, we can always do better. As an engineer, I want an evaluation process to be done rigorously but see no reason reactors can’t operate safely after a thorough investigation.”

Tags: