Abstract
The core of the German HTR-500 with 1250 MW(th) contains several hundred tons of graphite in the form of fuel elements and reflectors. In the case of core heatup accidents, when fission products are released from the coated fuel particles, this large graphite mass carries a high potential for binding the metallic fission products cesium and strontium by chemisorption and prevents their release into the environment. Based on available laboratory data, iodine shows a much weaker sorption on graphite than cesium and strontium. The computer code FRESCO was developed for examining the fission product behaviour in the pebble bed HTR under accident conditions. Sorption isotherms and effective diffusion coefficients are used in this code for the calculation of fission product transport and retention in graphite. Two methods are used in KFA for determining the sorption isotherms. The first is high temperature mass spectrometry with a cylindrical molybdenum Knudsen cell. It allows the measurement of partial pressures down to 10-7 Pa. The second method is the isopiestic measurement where the sorption of a certain element is determined at the same time for several samples, each being put into a crucible and placed in a molybdenum vessel. Sorption isotherms have been measured so far on A3-3 fuel element matrix for cesium (<= 1500 K) and strontium (<= 2000 K) . The sorption on graphite strongly depends on the kind of fission products, their chemical form, their concentrations and the different graphitic materials. Calculations with the FRESCO-code for the metallic fission products show the extremely high dependence of the release on the sorption isotherms. By the use of the sorption isotherms of this work, those computations yielded a drastically reduced calculated release of cesium and strontium compared to former analyses. For instance, the Cs core release from the depressurized HTR-500 during a core heatup accident will be reduced from a 0.22 fraction, when no sorption on matrix is assumed, to a 0.001 fraction, when sorption at all temperatures is assumed. It is assumed that the lower limit of the cesium release from the core (0.5%, in the above case) is given by the iodine release due to the formation of CsI, which probably shows no significant sorption effect on graphite. Our computations show an urgent demand for the experimental study to determine the sorption isotherms of Cs on A3 matrix at high temperatures (<= 2000 K) and of Cs and Sr on the German type of reflector graphite, which has not yet been investigated so far.
view the full text of this article (14 pages, format: PDF, size= 540kB)