Radwastes may escape glass via new route.

Over the next several years, the Department of Energy (DOE) will begin vitrifying -- incorporating within glass -- highly radioactive liquids and sludgy nuclear wastes to prevent them from escaping into the environment. But new research indicates that exposure to the slow dripping of water can change the largely nonreactive borosilicate glass into a form that facilitates the flaking of microscopic mineralized shards. More important, the DOE-funded study shows, some of the water-borne flecks bea unexpectedly enriched concentrations of certain of the interred radionuclides.

This flaking process, or spalling, represents a previously unknown mechanism for directly generating colloids -- particles too tiny to settle out of water, says study leader John K. Bates of Argonne (III.) National Laboratory. As colloids, the chemically stable castoffs can hitch a ride downstream, carrying radioactive contamination far from its source (SN: 3/17/90, p.169).

For the past five years, Bates' team has dripped water slowly (about one drop per week) over glass suspended in a stainless steel vessel. Water passing over the inch-long, half-inch-diameter glassy cylinder -- containing the radionuclides neptunium, americium and plutonium-- collected in the enclosing vessel.

Engineers evaluating high-level-waste barrier systems have assumed that any plutonium or americium leached from vitrified waste would dissolve into passing water, Bastes says. However, his group found that only neptunium dissolved.

By filtering the water that collected in the vessel after one 39-week period, they discovered that all of the americium and plutonium remained suspended in water as colloids of clay containing the mineral brockite. Indeed, they report in the May 1 Science, the vitrified waste did not spall microscopic shards of the original borosilicate glass, but of clay and other glass-derived minerals formed under the water's influence.

Glass is only quasi-stable, explains Bates, a physical chemist. Water breaks bonds between its silicon and oxygen atoms. In the spongy, gel-like zone that results near the surface, the once evenly mixed elements become "free to diffuse around" and recongregate into more stable minerals, he says.

Bates and his co-workers say their findings suggest that engineers must account for spalled colloids when designing barriers to prevent the escape of radionuclides at repositories.

"I agree that the colloidal aspect should not be ignored in assessing the impact of [radioactive-waste] repositories," says Laura McDowell-Boyer of Oak Ridge National Laboratory's office in Grand Junction, Colo. But, she adds, "it would be premature to suggest that [colloid spalling] is necessarily a significant environmental problem."

The Energy Department agrees. According to DOE's Fred Lash, an internal assessment of Bates' paper states: "There are no data presented nor is there discussion that indicates that the alteration of glass to form colloids diminishes or compromises its performance, in [a waste repository]."

In fact, says Bates, as long as you can trap spalled colloids, "I think it's an advantage to form these," since their intrinsicaly greater stability should make their long-term behavior more predictable.