Air ingress benchmarking with computational fluid dynamics analysis

The air ingress accident is a complicated accident scenario that may limit the deployment of high-temperature gas reactors. The complexity of this accident scenario is compounded by multiple physical phenomena that are involved in the air ingress event. These include diffusion, natural circulation, and complex chemical reactions with graphite and oxygen. In an attempt to better understand the phenomenon, the FLUENT-6 computational fluid dynamics code was used to assess two air ingress experiments. The first was the Japanese series of tests performed in the early 1990s by Takeda and Hishida. These separate effects tests were conducted to understand and model a multi-component experiment in which all three processes were included with the introduction of air in a heated graphite column. MIT used the FLUENT code to benchmark these series of tests with quite good results. These tests are generically applicable to prismatic reactors and the lower reflector regions of pebble-bed reactors. The second series of tests were performed at the NACOK facility for pebble bed reactors as reported by Kuhlmann [Kuhlmann, M.B., 1999. Experiments to investigate flow transfer and graphite corrosion in case of air ingress accidents in a high-temperature reactor]. These tests were aimed at understanding natural circulation of pebble bed reactors by simulating hot and cold legs of these reactors. The FLUENT code was also successfully used to simulate these tests. The results of these benchmarks and the findings will be presented.