Fifth Interim Status Report: Model 9975 PCV O-Ring Fixture Long-Term Leak Performance

A series of experiments to monitor the aging performance of Vitonreg. GLT O-rings used in the Model 9975 package has been ongoing for six years at the Savannah River National Laboratory. Sixty-seven mock-ups of 9975 Primary Containment Vessels (PCVs) were assembled and heated to temperatures ranging from 200 to 450 F. They were leak-tested initially and have been tested at nominal six month intervals to determine if they meet the criterion of leaktightness defined in ANSI standard N14.5-97. Fourteen additional tests were initiated in 2008 with GLT-S O-rings heated to temperatures ranging from 200 to 400 F. High temperature aging continues for 36 GLT O-ring fixtures at 200--350 F. Room temperature leak test failures have been experienced in 5 of the GLT O-ring fixtures aging at 300 and 350 F, and in all 3 of the GLT O-ring fixtures aging at higher temperatures. No failures have yet been observed in GLT O-ring fixtures aging at 200 F for 30--48 months, which is still bounding to O-ring temperatures during storage in KAMS. High temperature aging continues for 6 GLT-S O-ring fixtures at 200--300 F. Room temperature leak test failures have been experienced in all 8 of the GLT-S O-ring fixtures aging at 350 and 400 F. No failures have yet been observed in GLT-S O-ring fixtures aging at 200 or 300 F for 19 months. For O-ring fixtures that have failed the room temperature leak test and been disassembled, the O-rings displayed a compression set ranging from 51--95%. This is significantly greater than seen to date for packages inspected during KAMS field surveillance (23% average). For GLT O-rings, service life based on the room temperature leak rate criterion is comparable to that predicted by compression stress relaxation (CSR) data at higher temperatures (350--400 F). While there are no comparable failure data yet at aging temperatures below 300 F, extrapolations of the data for GLT O-rings suggests that CSR model predictions provide a conservative prediction of service life relative to the leak rate criterion. Failure data at lower temperatures is needed to verify this apparent trend. Insufficient failure data exist currently to perform a similar comparison for GLT-S O-rings. Aging and periodic leak testing will continue for the remaining fixtures.