Issue Date: March 1, 2004
RIDDING THE NATION OF CHEMICAL WEAPONS
The Army's 18-year-old program to rid the nation of its 31,000-ton arsenal of chemical weapons has been bedeviled by technical glitches, equipment malfunctions, and cost overruns. At its Anniston, Ala., incinerator facility, the Army is again facing one of its thorniest problems: aging rockets filled with the nerve agent GB.
A chemical stabilizer was added to some, but not all, of the GB agent that the Army produced during the Cold War era. Over time, the stabilizer--which the Army calls "di-c-di" for diisopropylcarbodiimide--has reacted with GB to form crystals or a gel, Anniston facility spokesman Mike Abrams explains. The Army estimates that up to 30% of the nearly 43,000 GB rockets in the Anniston inventory have undergone this chemical reaction.
So far, Anniston has only found--and incinerated--crystal-containing GB rockets. "The crystals appear to look like the rock salt used to make ice cream," Abrams says.
Army labs are now analyzing the crystals to determine their chemical composition, which military chemists suspect is most likely diisopropylurea. The labs are also determining whether crystalline GB can be destroyed by neutralization, the process the Army plans to use to destroy problematic GB rockets at its Blue Grass, Ky., storage site.
Most of the U.S.'s chemical weapons, however, will be destroyed by incineration. But the Army's incineration process is primarily designed to handle liquid agents, so the service has had to improvise ways around the gel and crystal problem.
At its incineration facility in Tooele, Utah, the Army chopped up problematic GB rockets and fed the metal casings and explosives, fiberglass firing tubes, and gelled GB into one furnace, the deactivation furnace. In contrast, during the processing of normal GB rockets, the liquid agent would have been drained and sent to a separate liquid incinerator for destruction. Using the one-furnace process, the Utah facility was able to process only 1.6 rockets per hour.
The Anniston facility, which is operated by contractor Anniston-Westinghouse, will demonstrate the one-furnace process this month during trial burns. Donavan Mager, spokesman for the contractor, says that, "employing lessons learned from Tooele," Anniston plant managers "believe we can destroy from five to nine rockets per hour."
But with possibly 12,000-plus gelled or crystalline GB rockets waiting to be destroyed, Anniston sought a way to safely increase the destruction throughput rate. The solution that engineers came up with was to insert a socklike filter to trap crystals in the process line between the agent holding tank and the incinerator. This sock has a smaller micrometer mesh than filters used in Utah.
With the sock filter straining out the crystals, crystal-free liquid GB can be destroyed in the liquid incinerator, while the rocket's metal parts and firing tube and the clogged sock filter can be burned in the deactivation furnace. This two-furnace procedure may increase to perhaps 14 the number of rockets destroyed per hour.
THE PROBLEM IS that the sock filter clogs up "after processing about 75 to 100 rockets," Mager says. This means that workers have to go more often into "hot" areas of the incinerator facility where chemical agents are present to replace the socks. But, Mager says, the two-furnace process "decreases the risk to the surrounding community because we are getting rid of the rockets more quickly."
Craig Williams, who directs the Berea, Ky.-based Chemical Weapons Working Group opposing incineration, contends the Army at Anniston "is experimenting with the disposal of nonliquid-agent rockets" and placing workers and the surrounding community at greater risk.
Williams claims that the Army has never used the sock filter--not at its now-shuttered disposal plant on Johnston Atoll in the Pacific Ocean and not at its operating facility in Utah. But Gregory Mahall, spokesman for the Army's Chemical Materials Agency, says filters have always been inserted in processing lines to strain out impurities. The smaller mesh sock filter "might be a new procedure or a variation on the project," Mahall concedes, "but it is not experimental in the sense that it's something new."
In a letter to Michael W. Wynne, a Pentagon undersecretary of defense for acquisition, technology, and logistics, Williams and 19 others representing community and environmental groups ask that operations at Anniston be halted until the identified problems are corrected. Williams tells C&EN that "we are concerned about technical problems and system malfunctions, worker exposure, and lack of information or intentional misinformation that is coming out of the facility."
"Obviously lessons at previous plants have not been learned or implemented."
Williams says Anniston plant managers have "inappropriately" stressed "schedule considerations over safety" and thereby increased risks to plant workers. He cites a Feb. 4 incident in which he says two workers were "exposed to GB nerve agent."
The workers, wearing protective clothing, went into a "hot" room in the main destruction building that houses GB holding tanks to change a clogged sock filter. During the course of what the Army says was routine maintenance, but which Williams believes was related to rocket processing, their protective suits became contaminated with low levels of the deadly nerve agent.
The workers then went through a decontamination process, which unfortunately resulted in contamination of their cotton undergarments. In the meantime, alarms sounded to alert workers to the presence of nerve agent. One worrisome alarm went off in a corridor in which negative air flow dictates that no agent should be present. The Army is still trying to determine why this alarm sounded and what changes need to be instituted to correct the problem.
Because their undergarments had become contaminated with low levels of nerve gas, the workers were sent to the facility's medical clinic, which is outside the main destruction building. Alarms sounded in the clinic, indicating the presence of GB. But blood tests revealed no change in the workers' baseline levels of cholinesterase, an enzyme indicator of nerve agent exposure, and the workers have shown no health effects.
In his letter, Williams cites facility press releases about the incident that he says were deceptive and false. "The language used by [facility] management in describing events at the incinerator are inconsistent and misleading and significantly downplay the serious nature of the incidents," he writes.
The incident made for good headlines in the local press. And Williams welcomes those media efforts. In his letter, he writes "that more accurate information only surfaces after extensive investigative efforts by the media or citizens and/or is brought to light by anonymous sources within the facility."
Mager tells C&EN that "we believe the entire incident was blown out of proportion." He says: "All safety procedures and precautions were followed to the letter. Worker safety is and always will be our top priority."
Still, Williams counters, Anniston's short--six month--operational history foreshadows "serious consequences for workers and the community" if what he believes are "serious technical problems" are not addressed.
The numerous agent alarms, unanticipated shutdowns, system malfunctions, and other technical problems are not unique to Anniston, Williams says. They were part of the operational history of the Pacific Ocean facility and are part of the ongoing operational problems at Utah.
"The Army continues to represent the Alabama facility as having captured and implemented the lessons learned at those two plants in order to avoid similar problems in Alabama," Williams says. But, he adds, "based on operational experience in Alabama to date, the same problems are occurring. So obviously lessons at previous plants have not been learned or implemented."
The Pentagon's Wynne has not had time to respond to Williams' letter.
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society