In September 1990, industrial hygienists making a routine surveillance of individual laboratories at ORNL discovered a possibly explosive situation that had been building since the earliest days of the Laboratory's existence.

During the walk-through, Ann Shirley of ORNL's Office of Safety and Health Protection first observed the use of perchloric acid hotter than 100 degrees Celsius in a laboratory hood, a protective enclosure that provides ventilation of noxious fumes, dusts, and gases to enable the safe handling of chemical, biological or radioactive materials. The problem was that the hood was not approved for perchloric acid. Shirley had become aware of such a potential problem after discussions with Mark Haskew and other ORNL industrial hygienists. Her quick check confirmed that the acid, in fact, had been used more than once in the hood, which had a ventilation system that was unable to handle the chemical properly.

Shirley reported the problem to Charlie Phillips, an ORNL industrial hygiene group leader. He observed the same problem in a hood that was to be taken out of operation for another reason. He reported the episode to a committee that regarded it as a safety-and-health red flag for all of ORNL. For Phillips and the committee, it raised the question of how many other hoods not rated for perchloric acid had been used for such work at the Laboratory in the past five decades.

Perchloric acid is popular in chemical processing because it offers all the desirable properties of mineral acids without introducing ions such as chloride, nitrate, and sulfate, which often interfere with other chemical reactions. But perchloric acid has a dangerous downside: When it dries, it leaves behind a perchlorate salt that can be flammable and highly explosive when subjected to heat or impact or reaction with other specific chemicals. If hot perchloric acid is used in a vent system that does not have internal wash-down capabilities, the salts will show up over time in the hood, baffles, filters, fans, ducts, and exhaust stacks.

When additional research by ORNL industrial hygiene personnel indicated that, indeed, hot perchloric acid had been used in other unapproved fume hoods during ORNL's 50-year history, Laboratory management acted swiftly. Murray Rosenthal, then ORNL deputy director, placed an immediate moratorium on the use of hot perchloric acid, halted all maintenance on laboratory hoods, and convened a special committee to launch a survey of ORNL's 700-plus fume-hood systems.

The findings? Forty fume-hood vent systems were contaminated with the salts at levels deemed dangerous by the National Fire Protection Association (NFPA). "It doesn't take much of this stuff to be dangerous," says Phillips, who manages ORNL's perchloric acid project. "A single gram of perchlorate salt has been linked to disastrous explosions in vent systems."

No Accidents at ORNL

Although no such accidents have occurred at ORNL, the industrial hygienists were aware that in 1962 a maintenance worker on an Atomic Energy Commission-related project was killed and two others were seriously injured in an explosion touched off by routine use of a small ball peen hammer and 6-inch chisel. The workers were dismantling a perchloric acid fume vent system when the explosion--violent enough to be heard 4 miles away--occurred.

Several other accidents involving perchloric acid are documented in the literature, and rumors circulate about explosions and fires during numerous informal salvage operations around the country, Phillips said.

"With this in mind, we knew we had to act right away to ensure the safety and health of ORNL personnel," said Marwan Bader, an ORNL industrial hygienist and perchloric acid project site manager. The state Environmental Protection Agency, recognizing that quick action was needed, agreed to allow removal of the perchlorates as an emergency response action under the Resource Conservation and Recovery Act (RCRA).

"It was decided that a committee must be put in place quickly to oversee handling of the project and that this committee should represent all aspects of the work at ORNL, including industrial hygienists, chemists, industrial safety experts, health physicists, engineers, fire safety engineers, maintenance personnel, and environmental compliance advisors," Bader explained.

Action Plan

In addition to identifying hoods in which perchloric acid had been used over the years, the committee directed development of sampling and analysis protocols and techniques to analyze the amount of perchlorates present, set a threshold designating serious contamination, and generated a plan to decontaminate the hoods and return them to use.

The threshold for contamination was set conservatively, Bader said, with respect to national standards. "The NFPA code pertaining to perchlorate contamination says that, if a methylene blue test turns to violet, that's positive for the presence of perchlorates," he explained.

"We found that this color change occurs at a level of about 750 parts per million (ppm). We went a step further and set our own standard at 500 parts per million," he said. A methylene blue test is similar in principle to a litmus test. A few drops of the test solution in a small quantity of water washed from the area to be tested will produce a violet precipitate if perchlorates are present.

The committee picked 3 of the 40 contaminated hoods for a pilot decontamination study, and a temporary enclosure was built in which the hood baffles, fans, ducts and other parts were to be cleaned. The main superstructures of hoods were cleaned in place in the laboratory. The enclosure included four 96-cubic-foot tanks and facilities for decontaminating parts on which any radiological material was found.

Before any testing or hood dismantling began, health physics surveys and industrial hygiene evaluations were performed, workers were trained in all areas necessary, including RCRA, and any protective clothing or gear was issued. Maintenance workers who actually took apart the vent systems donned ballistic gear similar to that used by police bomb squads.

Keep Them Wet

The key to dismantling the hood systems, Phillips says, is to keep everything wet. "Perchlorates aren't a problem as long as they are kept wet," he said. "It's the anhydrous form that poses a potential for instantaneous fire or explosion." Team members wet the hoods and other vent system components simply by spraying them with water, then peered through fiber-optic scopes into confined areas, such as behind hood baffles, to ensure that even the hard-to-reach spots were wetted down.

Once dismantled and while still thoroughly wetted, the hood baffles, fans, exhaust stacks, and sections of duct were wrapped in plastic and carefully transported to the temporary enclosure for thorough decontamination.

The highest concentration of perchlorates found ranged from 140,000 ppm at an elbow in a duct to 800,000 ppm on the inlet side of a filter housing. "Those are very high concentrations," Bader said, "considering the small amount of perchlorates necessary to trigger a powerful explosion under the right circumstances."

But that's not the number of greatest concern that came out of the pilot project: Workers removed nearly three pounds of perchlorate salts from a single hood vent system. The perchlorate buildup gave inner walls of the duct the soft, chilly appearance of a freezer in need of defrosting.

"It's likely that many other federal labs, university labs and industrial facilities throughout the country have these same legacy buildups of perchlorate contamination in their systems and just aren't aware of it," Phillips said, "or aren't fully aware of the potential safety and health risks involved."

In fact, since the presentation of the ORNL problem and solution at April 1992 meeting of the American Chemical Society, the details of the copyrighted procedures for the ORNL-developed decontamination methods have been in great demand by universities and industry. ORNL presentations at the American Glove Box Society conference in Seattle in August have further piqued the interest of those who suspect they may have similar problems. An ORNL talk given at the October DOE Safety Conference in Chicago generated still more interest.

"The pilot study was fantastic, very well accomplished," Bader said. "The techniques developed for the three test hood systems proved completely effective." Furthermore, the ORNL methods cost one-fifth the amount of outside contract estimates to decontaminate and dismantle a typical hood.

A welcome surprise, Phillips said, was that the procedure deregulated low-level radioactive contamination in several of the treated vent systems. "This alone will save approximately $12,000 per hood each year because those hoods no longer require 'contamination area' posting and the related expenses," he said.

Not only were the protocols effective, efficient, and practical, they were convincing. So much so that the NFPA modified their testing methods for perchlorate salts based on an ORNL proposal to the organization.

A full-blown effort is now under way at ORNL to remedy the remaining 37 contaminated hood vent systems. One cleanup operation is up and going, and funding proposals are out for more startups.

Phillips is pleased. "We had to abate a possibly substantial fire and safety hazard, and we did it well. We did it within environmental safety, Occupational Safety and Health Administration, and radiological controls," he said. "Most importantly, the safety and health of ORNL workers is being significantly protected."

The author, Wayne Scarbrough, is a member of the Martin Marietta Energy Systems Public Affairs staff based at ORNL. A version of this article was published in the Winter 1994 issue of Safety Connection.

Search Magazine
Features Index Next Article Previous Article Comments Review Home

Web site provided by Oak Ridge National Laboratory's Communications and External Relations
ORNL is a multi-program research and development facility managed by UT-Battelle for the US Department of Energy
[ORNL Home] [CAER Home] [Privacy and Security Disclaimer]

Last Revised: Tuesday, February 5, 2008 2:38 PM