As the U.S. Dept. of Energy and its contractors rev up construction of permanent repositories for nuclear wastes now lingering in aging, corroded and heavily contaminated facilities at some of the agency’s former weapons-production sites, new technology is helping cleanup crews get at the nastiest and toughest-to-remove remnants.
At DOE’s Hanford site in eastern Washington, officials are using insect-like crawlers to prowl the radioactive floors of 177 enormous underground waste tanks (some holding up to 247,000 gallons), as well as massive retractable arms to methodically blast toxic sludge off the sides of those vessels. At least 53 million gallons of mixed nuclear and chemical wastes had been stored in 149 single-shelled tanks that date to the 1940s. Remaining liquids and other waste forms are being moved from the tanks into 28 newer and more secure double-shelled vessels on-site. From there, the waste will be vitrified—transformed into glass-like logs—for long-term site storage.
San Francisco-based Bechtel Group, which is building the Waste Treatment and Immobilization Plant at Hanford that will vitrify the waste, announced on Oct. 19 that it had reached the 50% milestone in construction of the innovative but controversial $13-billion megaproject that has faced tough scrutiny because of technical problems and cost and schedule overruns since work started in 2001. It is scheduled to operate in 2019 but could start some waste processing two years earlier.
Washington River Protection Solutions (WRPS), the URS Corp.-led DOE contractor at Hanford in charge of tank-waste management, has designed a retractable robotic arm that can reach 35 ft into a tank’s nooks and crannies to dislodge waste solids and sludges. It replaces a small, remote-controlled bulldozer that could be folded to fit through the 12-in. openings at the top of the tanks. The bulldozer became stuck in radioactive muck and lost a tread.
Armed with a tip that can be changed depending on the type of waste to be removed, the Mobile Arm Retrieval System (MARS) sprays liquid to break up wastes and pumps them out. Salt-cake wastes, toxic sludges with the consistency of peanut butter and a hard crust of radioactive chemicals, or the “heel,” at the tank bottom require different spray heads for MARS.
“For the first time, we have this ability using MARS to efficiently remove all the waste from the tanks,” says Scott Saunders, WRPS manager of retrieval technology development. “In the past, after we have done the bulk retrieval using a pump, we have been left with a heel. So we have had to spend money to remove that equipment and install new tools. The big advantage here is that it is a single tool to complete the job.”
MARS was built using off-the-shelf technology, essentially taking the crane arm off the back of a truck, says Saunders. WRPS spent about $10 million to modify and test the device so it could be inserted through a small hole at the top of a tank. Saunders estimates that once the arm is fully tested, more systems could be built for about $5 million to $6 million each.
Considering the intensely radioactive environment in which the robot arm works, the modifications are crucial. Saunders says engineers removed chemical lubricants from the commercial arm and replaced them with lubricating plates, to avoid unforeseen chemical reactions in the witch’s brew of chemicals found in the tanks. Protective sleeves were also added, so that when the arm is removed for reuse in another tank, it can easily be cleaned off and radioactive materials do not seep into the arm mechanism.
Another advantage of the arm is that it can pump liquid waste from a nearby tank, instead of water, to break up a tank’s solid-waste remnants before they are pumped into to a double-shelled tank. The process ultimately avoids introducing new water into the tanks, reducing the volume of liquid waste that eventually will have to be processed. Early challenges faced by the robot-arm developers have helped them improve the technology, says Saunders. Testing of new models is more rigorous and involves site operations staff to bolster their experience with the technology, he adds.
The arm and the associated technology used to move broken-up waste out of the tanks likely will be up and running and ready to be deployed in late 2010 or early 2011 at a tank in Hanford’s so-called C-farm, Saunders says. Design work is scheduled to conclude by the end of the year. One challenge is to make the system transportable between tanks and make use of existing water and electrical utilities at all locations.
At DOE’s Savannah River Site (SRS), another former nuclear-weapon manufacturing complex near Aiken, S.C., researchers are looking at a host of technologies to condense waste and package it safely for permanent storage. While the site’s waste is similar to that at Hanford, use of MARS or a similar robotic arm was not feasible because the old storage tanks have miles of cooling pipes to prevent radioactive waste from overheating, .
SRS officials developed a rod-like robot that can be lowered through the small opening at the top of a waste tank. Once at the tank floor, the so-called Sand Mantis scissors open, revealing metal wheels, a spray nozzle and a suction head. Small plates also fold out of the front, giving it an insect-like appearance. The device funnels waste broken up by the spray into a suction head as it drives around the tank bottom.
Robots modified with laser surveying equipment have been used at SRS to map the complex of tank pipes, and DOE is developing ways to create 3D maps for workers to use in the cleanup.
“If we can come up with a technology or an alternative that is going to somehow...accelerate the cleanup, it is going to ultimately cut years off the cleanup schedule, says Kurt Gerdes, who has been involved in DOE cleanup-technology development for more than a decade and now is acting director of DOE’s office of waste processing in Washington, D.C. “We are talking huge cost savings,” he says. DOE will save about $1 billion for every year it speeds up the cleanup schedule at the Hanford site, he adds.