4.16.4.1 Expected In-Reactor Performance

As implied by Figures 21 and 22, Tables 1 and 2 and

in the earlier sections, permeation barriers can be used to reduce the effective permeation in laboratory

testing.175-179,183,195,196198-203,205 PRFs from labora­tory experiments have been reported to be many

175,183,195,196,201,202

thousands in certain barrier systems.

However, while the data available in the open litera­ture are quite limited, there is significant evidence that the effectiveness of the permeation barriers decreases in radiation environments. There were three sets of experiments21 — performed in the high flux reactor (HFR) Petten reactor in the Neth­erlands. In the first of these experiments21 , reported in 1991 and 1992, tritium was produced by the liquid breeder material Pb—17Li. Permeation of tritium through a bare 316 stainless steel layer was compared with that through an identical layer cov­ered with a 146-p. m thick aluminide coating. Over the temperature range 540-760 K, the barrier was reported to decrease the permeation by a factor of 80 compared to the bare metal, that is, PRF = 80. In the LIBRETTO-3 experiments,220 three different permeation barrier concepts were tested with the tritium again produced by the liquid breeder mate­rial. One irradiation capsule for tritium breeding was coated on the outside with a 6-8-p. m thick CVD layer of TiC. A second capsule was coated on the inside with a 0.5-1.5-p. m thick layer of TiC followed by a 2-3-p. m thick layer of Al2O3. The third barrier was an aluminide coating produced by the cementation process. The aluminum-rich layer was ^120-p. m thick with about 5 mm of Al2O3 on the outside. The single TiC layer reduced the tritium permeation by a factor of only 3.2, the TiC and Al2O3 layer reduced the permeation by 3.4, and the pack cementation aluminide coating reduced the permeation by a factor of 14.7. These are surprisingly small reductions PRFs compared to laboratory experiments. In a third set of experi- ments,221 the tritium production was achieved with the solid ceramic breeder materials. Both double-wall tubes and single-wall tubes with a permeation bar­rier were tested. The double wall configuration had an inner layer of copper. The permeation barrier on the other system was an aluminide coating with a thickness of 7 mm. The aluminide coating was reported to be 70 times more effective than the double-wall configuration in suppressing permeation. Unfortunately, different breeder materials were used for the two different experiments, and the results could have been strongly affected by the amount of tritium released from the ceramic as well as the form of release (T2 vs. T2O). The bottom line on the irradiation testing of barriers is that barriers do not perform as well in a reactor environment as expected from laboratory experiments: a PRF > 1000 has not been achieved in reactor environments.