Nuclear waste heads into the virtual realm (April 2014)

Apr 16, 2014,

Illustration of the virtual underground laboratory at the Fraun
Testing ground: the virtual underground laboratory

A new computer-based tool designed to help find the best sites for nuclear-waste repositories and to win public confidence in them has been developed by researchers in Germany. The 3m VIRTUS virtual underground laboratory will allow scientists to explore the behaviour of highly radioactive materials inside specific rock formations, with the aim of making it cheaper to develop and build repositories. Critics, however, argue that the new software will do little to improve safety and might disrupt real laboratory studies of nuclear waste.

Underground disposal

Many scientists believe that the best way to dispose of spent nuclear fuel and other long-lived radioactive materials is to bury them hundreds of metres underground, with Sweden and Finland having both selected sites for national waste repositories next to existing nuclear power stations. France also plans to open its own facility in 2025, and, like Sweden, has built a major underground lab to test the geology and technologies to be used at the site.

However, there are severe technical and societal problems associated with repositories, not least that the waste they contain will remain harmful for hundreds of thousands of years. The development of a national repository in Germany, for example, has been mired in controversy. A formal site-selection process has still to be set up, even though exploratory work at the Gorleben salt mine in the north of the country began as far back as the 1970s. The nearby Asse mine, meanwhile, was set up in the 1960s as a research facility but was decommissioned in 1997 after a brine leak threatened to flood the complex and cause it to collapse.

Developed by the Fraunhofer Institute for Factory Operation and Automation (IFF) in Magdeburg, together with Germany’s nuclear-safety organization (GRS), the Federal Institute for Geosciences and Natural Resources and the waste-repository company DBE Technology, VIRTUS will attempt to partially address this issue. The software enables detailed models of specific rock formations or mine structures to be created and then fed into a simulation to calculate how a repository would evolve physically and chemically over time. The results of these calculations can then be visualized graphically, and it is planned that members of the public will in future be able to see those graphics inside a 360° projection system.

Modelling heat

Klaus Wieczorek from the GRS, who is head of VIRTUS, says that the software could, among other things, model the heat emitted by the radioactive decays taking place inside canisters, and the resulting temperature-induced stress that would build up in surrounding rocks. It could be used to better design laboratory experiments, he explains, and to simulate the performance of potential repositories ensuring that safety criteria, such as maximum-allowed temperatures, are met and that the position of tunnels can be optimized to minimize mining costs.

A prototype VIRTUS system was supposed to have been completed this spring, but unexpected difficulties associated with matching up the geological models with those simulating the behaviour of nuclear waste has now pushed that deadline back. "We are continually improving the prototype and we will present it to funding institutions in October this year," says Wieczorek. In fact, he admits that it might be "another two or three years" before it is ready for public use.

Arising uncertainities

Johan Swahn of Swedish nuclear-repository watchdog MKG believes that the new software has little or nothing to contribute to research on radioactive-waste disposal. He says that experiments carried out in underground laboratories continue to provide "a lot of surprises". For example, new uncertainties have emerged regarding how copper canisters designed to hold Sweden’s spent fuel behave in low-oxygen environments, and as a result the licence application for the proposed national repository may not be approved. "Creating a generic safety case with a nice visualization will in my opinion only enhance a dangerous belief in modelling, creating a false impression that we have understood more than we actually have, he says.

About the author

Edwin Cartlidge is a science writer based in Rome

Northwatch | Box 282 . North Bay . P1B 8H2 | Tel 705 497 0373 |

%d bloggers like this: