Publikationsrepositorium - Helmholtz-Zentrum Dresden-Rossendorf

Radioactive carbonaceous aerosol particles deposit within the primary circuit of a high temperature reactor (HTR). These particles can pose a considerable threat if resuspended by a shock wave in the case of an unlikely loss-of-coolant accident and partially released into the environment (Kissane, 2009).
Stempniewicz et al. (2010) carried out simulations of the spatial and temporal development of carbonaceous dust deposits in the primary circuit of a HTR. Maximum dust layer thickness after full operation time of 60 years was well above 10 mm. However, the underlying multilayer models require additional experimental verification.
Therefore, particle deposition and resuspension experiments of multilayer graphite deposits have been conducted in a small-scale test facility. The results for particle layer thickness have been correlated to corresponding flow data as measured by means of a 2C2D Particle Image Velocimetry (PIV) system.
The experimental facility consists of a ventilated square duct (d = 100 mm), which is operated with air at ambient conditions. The average air velocity can be regulated over the range u = 1..7 m/s, corresponding to Reynolds numbers between Red = 8.9k..42k. As flow obstacles, periodic steps have been chosen (Fig. 1) allowing the comparison of results with previous studies on monolayer deposition (e.g. Lai et al., 1999).
Graphite aerosol particles (Thielmann Graphite, 23061) have been injected into the duct 15 diameters upstream of the test section by means of a dust disperser TOPAS - SAG 410. Setting the mass flux to 80 g/h an average mass concentration of airborne particles of up to 2 g/m³ is established. The aerodynamic particle size distribution of the graphite particles was determined by means of a TSI – APS 3321confirming that the particles cover well the range of interest for HTR conditions (quantiles:
dP,10 = 1.83 µm, dP,50 = 4.40 µm and dP,90 = 10.64 µm).
The deposition experiments have been conducted under different flow conditions and the development of graphite particle deposits in the vicinity of the steps have been observed by hourly laser triangulation measurements. Subsequently, the generated multilayer has been resuspended at a higher flow velocity and the thickness of the remaining structure has been measured using the same techniques.
Results show that for a relatively low Reynolds number of Red = 8.9k (Fig.2), deposition is mainly governed by gravitational settling and the deposition rate is almost independent of position and time. On the other hand, subsequent resuspension at higher flow velocities (Red =27.2k) reveals a strong relation between flow field and remaining layer (Fig.3), which is further being investigated.