The HTR primary-system environment comprises nuclear graphites, alloys, dust (primarily carbonaceous) and high-purity helium. The amount of carbonaceous dust produced in a pebble-bed system would be considerably greater than one using a prismatic core with a significant contribution arising from the partially-graphitized binder of the pebbles. The dust is very fine, <10μm in size. Experience with HTRs shows the primary system to be contaminated by the isotopes 134Cs, 137Cs, 90Sr, 110mAg, 131I, 135Xe, 85Kr and tritium at a level representing an occupational-health issue rather than a safety issue. However, strong sorption of caesium, strontium, iodine and tritium onto carbonaceous dust has been observed. Hence, the extent to which deposited dust can be resuspended during a depressurization accident is a safety issue since the dust comprises the main vector for release of radioactivity into the confinement. For fine dust on a surface, the principal force keeping it in place arises from inter-molecular (van der Waals) forces while aerodynamic forces, mainly drag, act to remove it. The reference model chosen here for improving resuspension predictions is the so-called Rock’n’Roll model. This model is based on a statistical approach leading to a resuspension rate for the escape of particles from a potential well via the action of the fluctuating aerodynamic force caused by turbulence. The as-published Rock’n’Roll model assumes that the fluctuations of the aerodynamic force obey a Gaussian distribution. Here, we introduce calculated statistics for the fluctuations taken from a large-eddy simulation of turbulent channel flow (work is in progress on generating these statistics using direct numerical simulation of turbulence). The overall influence of more-realistic (non-Gaussian) forces on the resuspension rate is found to be an increase in short-term resuspension. Given this and the fact that the adhesive force is based on an empirical correlation, work has started on developing specific modelling for multi-layer deposits.
Read full abstract