Development of a computer code for prediction of crud transport in PWRs

Abstract

Mathematical modeling and computer code development for corrosion product transport are essential to the Operational Radiation Exposure (ORE) reduction and evaluations of various control methods in primary circuits of PWRs. However currently available mathematical modelings and computer codes are very limited and are most often proprietorial. For example, CRUDMIT code has limitations such as using only two nodes in describing primary circuits, assuming single layer of crud deposits and considering only two radionuclides($Co^{58},\; Co^{60}$) out of major five radionuclides($Co^{58},\; Co^{60},\; Fe^{59},\; Cr^{51}$ and $Mn^{54}$). In this work, we have developed a computer code called CRUDTRAK, which improves these limitations by assuming multiple nodes, double layer deposits, considering five radionuclides and adding crud exchange diagrams used for PACTOJ code. PCCL operation at MIT was simulated for the verification of CRUDTRAK code. The simulated results were compared with experimental data of PCCL operation and sensitivity analyses were performed with some control methodes and input data. For the longer time simulations(about 600 days), $Co^{60}$ becomes the dominant radionuclide, however at the short runs(about 50 days), $Co^{58}$ becomes the dominant one and the others are negligible. High pH(7.4) operation is found to be a good strategy for ORE reduction and the increment of purification flow has negligible effect. Results of present work are turned out to be quite consistent with measured experimental data, and the simulation by CRUDTRAK code showed much precise results than that by CRUDMIT code.