On July 27, 1994, six CAM alarms actuated when personnel performing tests on an iridium capsule overpressurized a glovebox at the Los Alamos National Laboratory (LANL) TA-55 plutonium processing facility. Contamination spread throughout the room and two of nine individuals present in the room were contaminated. One individual was contaminated on the hand and a nasal smear taken from another individual was positive. There was no release of contamination outside the room. (ORPS Report ALO-LA-LANL-TA55-1994-0024)
Facility personnel were testing the venting capability of the capsule, which was used to encapsulate plutonium-238 pellets used for Radioisotope Thermoelectric Generators. There was no plutonium in the capsule. The test involved placing a surgeons glove on one end of the capsule, applying 60-psi air to the other, and attempting to pass air through the vent. The vent consisted of a small hole drilled in one half of the capsule and packed with crushed iridium metal to permit the venting of helium gas. The air system consisted of a 100-120 psi supply line and two regulators which reduced the pressure to 60 psi.
Shortly after opening the air valve, technicians realized that the glovebox was overpressurized. Almost immediately the first of six continuous air monitors (CAMs) alarmed. The technicians reduced air line pressure to 2 psi using the regulators, opened a door to an adjacent drop box, and evacuated to the adjacent corridor. Two Radiological Control Technicians (RCTs) entered the room with full anti-contamination clothing and respiratory protection, observed several CAMs alarming, and left. Three RCTs returned later with self-contained breathing apparatuses to change CAM filters and turn off the compressed air to the glovebox.
Facility personnel determined that the glovebox became overpressurized because the glovebox exhaust configuration was insufficient to vent the air volume introduced by the air supply. They discovered that the glovebox exhaust damper was closed, isolating the glovebox exhaust system and preventing the release of air. Site personnel identified a number of other factors that contributed to the event.
The test was not part of an experimental plan or any other approved work document. Such documentation would have required a safety review. The activity was not reviewed by the Area Work Supervisor, as is required to assure that all evolutions are properly coordinated and performed by trained personnel and with adequate safety review. The glovebox configuration was not routinely verified prior to commencing experimental activities.
Personnel reviewing the event reported that it was not clear whether the glovebox magnahelic gauge or the exhaust butterfly damper position were checked. However, they did indicate that the magnahelic gauge could have indicated a negative pressure if (1) the glovebox was under residual negative pressure from another glovebox that the material was transferred from, or (2) leakage past the damper was sufficient, with no air being introduced into the glovebox, to maintain a slight negative pressure.
All nine employees in the room at the time of the event submitted nasal smears and were issued bioassay kits. One employee had a positive nasal smear (91/0 dpm) and will undergo an in-vivo analysis. Facility managers held a critique and initiated an event investigation. A verbal announcement was made over the facility PA system that external compressed gases in gloveboxes were not to be used in isolated boxes until the event investigation was completed.
The facility manager suspended use of gloveboxes using compressed gases pending walkdowns, review of compliance with design criteria, and verification of adequate exhaust ventilation configuration. Site personnel issued a Safety Alert on July 27, suspending all PF-4 activities pending completion of a number of activities including an evaluation of compressed gas service for each glovebox.
A review of ORPS reveals a number of glovebox overpressurization events at DOE facilities.
In January 1994, a glovebox overpressurized and caused CAM alarms and contamination in another room at LANL TA-55. The room was not occupied at the time, so there were no adverse consequences to facility personnel. A technician had increased air flow to a cascade dissolver to resuspend solids in the dissolver. However, the increased air flow was inadvertently left on for an extended period of time, resulting in the generation of above-normal levels of acid aerosol. The increased levels of acid aerosol saturated the exhaust HEPA filter material, causing it to swell and effectively block the filtered exhaust from the glovebox. The lack of exhaust flow, coupled with the above-normal air flow and a closed guillotine door, caused the glovebox to lose its negative pressure and become slightly positive. (ORPS Report ALO-LA-LANL-TA55-1994-0002)
In June 1992, a glovebox window at Argonne National Laboratory-East failed when the glovebox became overpressurized while a scientist was returning a cryogenic purification system to normal operation after a weekend regeneration. An error in determining the pressure of the unit resulted in a release beyond the pressure-relieving capabilities of the glovebox. There were no radioactive materials involved and chemicals stored in the glovebox were not affected. (ORPS Report CH-AA-ANLE- ANLEER-1992-0005)
In March 1992, a glovebox overpressurization occurred at the Oak Ridge Y-12 Site during testing of fire suppression system equipment, resulting in the rupture of the left side viewing glass. There were no personnel injuries or contaminations, and no release of contaminated materials. The overpressurization was caused by failure of a fire suppression system exhaust valve to open and allow argon to be vented to atmosphere. Facility personnel issued a Yellow Alert advising site personnel of the potential for overpressurization of gloveboxes during testing. (ORPS Report ORO-- MMES-Y12DEFPGM-1992-9007)
These events are significant because overpressurization of gloveboxes can result in a release of significant quantities of radioactive material and the potential for significant personnel exposure. A number of lessons can be learned from reviewing glovebox overpressurization events.
All potential sources of air or gases that could cause a glovebox overpressure condition should be identified, analyzed, and controlled. Typical sources include compressed gases servicing gloveboxes, cryogenic materials that could significantly increase in volume with a temperature increase, and air or gases that could be introduced during testing and operation of glovebox support systems. The use of fixed restrictors, such as orifices or pipe size reductions, should be considered to limit air or gas volume introduction to rates below the glovebox design basis for maintaining negative pressure in the event of a pipe break or regulator failure.
Facility personnel should not assume that the glovebox safety envelope is complete based solely on indication of negative pressure on a magnahelic gauge. Examination of the total configuration, including the position of the exhaust damper, is necessary to assure safe glovebox operation.