the O-rings were subsequently deposited into the stainless steel sealing
surfaces above and below the trapped O-rings, which led directly to the
introduction of chloride-induced stress-crack corrosion in the stainless steel.
The operational conditions that set up the introduction of the stress-crack
corrosion were moderately elevated temperatures (i.e., less than 100 EC), low
pressures (i.e., less than 3 atmospheres), and exposure times of 3-5 years.
Fortunately, the damage was discovered before any failures occurred, the
neoprene O-rings were removed, and the seal design was changed to a non-
O-ring type of seal.
In a second case, six failures out of six tests occurred when high-quality Type
316 stainless steel was exposed to tritium gas in the presence of TeflonTM
shavings and 500 ppm moisture. All of the failures were catastrophic, and all
were the result of massively induced stress-crack corrosion. The conditions
that set up the introduction of the massively induced stress-crack corrosion in
this case were moderately elevated temperatures (i.e., 104 EC), high pressures
(i.e., 10,000 to 20,000 psi), and exposure times that ranged from 11 to 36
hours. Since the time to failure for all the tests was directly proportional to the
pressure (i.e., the higher pressure tests failed more quickly than the lower
pressure tests), since identical control tests with deuterium produced no
failures, and since comparable testing without the TeflonTM shavings indicated
no failures after 3,200 hours, it was concluded that fluorides were being
leached out of the TeflonTM and deposited directly into the bodies of the
stainless steel test vessels. An interesting sideline to this test is that, after the
tests, the TeflonTM shavings showed no obvious signs of radiation damage
(i.e., no apparent discoloration or other change from the original condition.