Friday, Aug. 15, 2008

Leak ruled out in probe of Hanford's underground tank waste

By Annette Cary, Herald staff writer

A leak has been ruled out as the cause for a drop in the level of liquid inside one of Hanford's huge underground waste tanks holding high-level radioactive waste.

That's been the good news at the Hanford tank farms this summer, as an equipment malfunction has suspended work at the only tank where radioactive waste was being retrieved. Waste from 142 older, leak-prone, single-shell tanks is being transferred to 28 newer double-shell tanks.

The Department of Energy and its contractor CH2M Hill Hanford Group had feared that a leak was responsible for a sudden drop in the liquid level earlier this year in a single-shell tank believed to have leaked radioactive waste into the soil beneath it in the past.

In May, Hanford workers saw a drop of about 1.7 inches in Tank SX-104.

But a leak was the worst-case scenario, with several other possibilities also considered for the fluctuation of the liquid level in the tank.

"Every 10 years (the tank) seems to sort of shove itself into the spotlight," said Steve Pfaff, the DOE federal project director for tank retrieval.

In 1988, the tank was declared an "assumed leaker" when its waste level dropped, although a leak never was detected. In 1998 level fluctuations again were detected, and that time responses to changes in barometric pressure were blamed.

But by 2008 all the pumpable liquid had been removed, leaving just liquid oozing among solids too deep in the waste to likely respond to barometric pressure.

Instead, the fluctuating level of liquid has been linked to the installation of a vertical pipe inside the tank in December. It's being used as a new observation well to measure the level of liquid among the tank's solids.

A water lance sprayed about 200 gallons of water into the tank to dissolve or wash away solids in the tank and drill a hole down to its bottom to install the well.

The new liquid earlier appeared to have settled among the solids in the tank to bring the liquid level to about 91 inches, although it's difficult to measure in the enclosed tank.

Now engineers believe that the liquid had remained high above difficult-to-penetrate solids in one area of the tank and the liquid in the rest of the tank was at about 73 inches. It appears that the drop in the liquid level observed this year was the water slowly leveling out as liquid finally dispersed through the solids.

"The true level does not appear to have changed," Pfaff said.

New contamination was not detected with dry wells that surround the exterior of the tank. However, any tank waste that leaks is likely to gel once it cools outside the tank rather than seeping through the soil to the monitoring wells.

The Washington Department of Ecology has agreed with DOE's assessment of the tank.

But results of the investigation still must be presented to representatives of the Defense Nuclear Facilities Safety Board, DOE said at a Tuesday committee meeting of the Hanford Advisory Board.

The board also heard an update on progress on emptying single shell tanks after retrieval of waste was stopped for 10 months after a spill of waste at Tank S-102.

The tank picked for the restart of retrieval was Tank C-109, but CH2M Hill Hanford Group ran into a mechanical problem before retrieval began. One of the two tracks that roll the robot across the waste came off a couple of days after the robot began working to position the waste to be pumped out of the tank.

The robot, which cost about $500,000, was not intended to be removed from the tank after it was contaminated with radioactive waste, limiting options for repairs.

CH2M Hill opted to continue running the robot with a single track. But the extra stress on one side of the operating equipment apparently caused a hydraulic oil leak of about 15 to 20 gallons inside the tank.

Work was stopped because of the mechanical problems and problems removing the waste from the tank.

The robot was breaking up solids with high-pressure liquid sprays and using its blade to push the waste toward the pump at the center of the tank.

But the waste particles still were so large that they could not be pumped out, said Herb Berman, a CH2M Hill vice president and chief engineer.

DOE considers the robot still to be a potentially viable tool, said Ron Frink, the DOE facility representative for the tank farms. With more operating time, it might have performed better, he said.

"We're really disappointed it has failed prematurely," he said.

What to do next will be up to the incoming tank farm contractor, Washington River Protection Solutions, which takes over Oct. 1.

It may decide to try pulling the contaminated robot out of the tank to repair it or it could try a new technology.

This is the second technology used to empty the tank after modified sluicing failed to empty more than 88 percent of the waste in the tank. Modified sluicing uses liquid sprayed from the top of the tank to move waste toward a pump.

CH2M Hill is preparing to begin retrieving waste from Tank C-110, expecting modified sluicing to work on its waste. Pumping should begin before Oct. 1, and then the new contractor will take over.