MODELING ATMOSPHERIC RELEASES OF TRITIUM FROM NUCLEAR INSTALLATIONS

Tritium source term analysis and the subsequent dispersion and consequence analyses supporting the safety documentation of Department of Energy nuclear facilities are especially sensitive to the applied software analysis methodology, input data and user assumptions. Three sequential areas in tritium accident analysis are examined in this study to illustrate where the analyst should exercise caution. Included are: (1) the development of a tritium oxide source term; (2) use of a full tritium dispersion model based on site-specific information to determine an appropriate deposition scaling factor for use in more simplified, broader modeling, and (3) derivation of a special tritium compound (STC) dose conversion factor for consequence analysis, consistent with the nature of the originating source material. It is recommended that unless supporting, defensible evidence is available to the contrary, the tritium release analyses should assume tritium oxide as the species released (or chemically transformed under accident's environment). Important exceptions include STC situations and laboratory-scale releases of hydrogen gas. In the modeling of the environmental transport, a full phenomenology model suggests that a deposition velocity of 0.5 cm/s is an appropriate value for environmental features of the Savannah River Site. This value is bounding for certain situations but non-conservative compared to the full model in others. Care should be exercised in choosing other factors such as the exposure time and the resuspension factor