Incidence of collisions on the resuspension of microparticles by turbulent air flows (ref RES22-5)

To assess the safety of nuclear installations and the relevance of the associated radiation protection measures, an important step consists in determining the terms sources of contamination in normal operation and for different accident scenarios. Assuming contamination in particulate form, these source terms are calculated using resuspension coefficients which relate the quantity of particles emitted to the initial quantity involved, depending on the scenario considered. For this, IRSN has been conducting experimental and numerical research for several years in order to develop phenomenological models capable of describing the resuspension of contaminants for different industrial situations [1, 2].

In fact, and although relatively well documented in the literature, the description of the resuspension of particles by air flows remains incomplete and current scientific obstacles relate to improving the characterization of adhesion and aerodynamic forces, as well as the conditions of particle detachment. In particular, the theoretical models of particle resuspension have been, for the most part, developed by simply considering the case of isolated particles deposited in a monolayer on a flat surface [3, 4]. The detachment condition considered by these models is based on three mechanisms that can lead to the resuspension of a particle: direct detachment, sliding and rolling of the particle on the surface before it is resuspended. The modeling of these mechanisms involves a precise description of the distributions of the aerodynamic and adhesion forces which are exerted on the particles and which depend on many parameters such as the properties of the flow, the size, the shape and the nature of the particles, as well as the nature and roughness of the substrates [5, 6, 7].

As the surface concentration of particle deposits increases, several studies have shown that a fourth mechanism, identified as an inter-particle collision regime, can significantly participate in the resuspension. For example, Ibrahim et al. [8] have shown that large particles (70 μm) can move along the surface after their detachment and then collide with other particles present on the surface. Very recently, Banari et al. [9] analyzed this phenomenon of collision thanks to a high frequency optical microscopy technique and were able to highlight collision cascades between particles of identical size (40 μm). On the other hand, the effect of the poly-dispersion of the particles in the above scenarios could make the predictions even more complex. Indeed, a recent study carried out in our laboratory on monolayer deposits composed of poly-dispersed particles indicates that the setting in motion of the largest particles leads to an abnormally large resuspension of the smallest particles [10]. This result again suggests the preponderant role of the collision mechanism in the resuspension of particles constituting more realistic deposits regarding industrial situations studied at IRSN (high surface concentration and poly-dispersion in particle size).

The objective of the thesis work is to carry out experiments allowing the validation of hypotheses on the role of the collision mechanism in the resuspension of particles. Depending on the results obtained, an adaptation of the analytical and numerical models describing this phenomenon could be proposed, in particular by the Monte Carlo method. This work should lead to the inclusion of this mechanism in the suspension model recently implemented in ANSYS CFX and used by IRSN [11].

The student should, first of all, become familiar with the literature on the subject and with the experimental facilities already available in the laboratory. This bibliography phase should involve a critical analysis of the resuspension models currently used and the identification of the parameters to be varied experimentally in order to study the collision mechanism. The next step will be to design and size an adequate high-frequency optical measurement method that can be easily adapted to the BISE facility located in Saclay.

The second phase will consist of implementing an experimental campaign to vary the parameters of interest previously identified. A major effort is expected in the manufacture of calibrated particles, reproducible deposits and substrates with controlled roughness. For this, the candidate will have access to the metrology equipment park (aerosol generators, granulometers, optical and electronic microscopes, etc.) available in the laboratory.

The third phase of the work will be divided between the analysis of the results of the previous phase, their interpretation and the improvement of existing models to take into account these collisions

between particles. To do this, the candidate will rely on the sensitivity analysis tools used by the CaliSto team to identify the main factors influencing the results. When relevant, these analyzes will also be used to refine new measurements on certain parameters whose influence may have been initially neglected. These analyzes will also serve as a basis for establishing new models that take these mechanisms into account. In particular, the candidate will participate in the improvement of the model accounting for the dynamics of particles on rough walls developed by C. Henry in the CaliSto team. In connection with the issues identified in [9], the objective of these new models will be to include a richer scenario accounting for the complexity of the interactions between particles (the transfer of energy during the collision which can give rise to the setting the two particles in motion, or stopping them, etc.). Finally, the candidate will also participate in the improvement of a resuspension model developed by the team of Ana Maria Vidalès (CONICET, Argentina) with whom the LPMA has been collaborating since 2016. Thus, the candidate will spend several months at the University of San Luis (Argentina) to train in the use of the numerical model based on a Monte Carlo method. Finally, the last phase will be devoted to writing the thesis.

[1] Peillon, S., Roynette, A., Grisolia, C., and Gensdarmes, F. (2014). Resuspension of carbon dust collected in Tore Supra and exposed to turbulent airflow: Controlled experiments and comparison with model. Fusion Engineering and Design, 89(11), 2789–2796.

[2] Rondeau, A. (2015). Etude de la mise en suspension aéraulique appliquée à la problématique des poussières dans le futur tokamak ITER. PhD thesis, Université Paris-Saclay.

[3] Ziskind, G. (2006). Particle resuspension from surfaces - Revisited and re-evaluated. Reviews in Chemical Engineering, 22,1-2.

[4] Henry, C. and Minier, J.-P. (2014). Progress in particle resuspension from rough surfaces by turbulent flows. Progress in Energy and Combustion Science, 45, 1–53.

[5] Zhang, F., Reeks, M., and Kissane, M. (2013). Particle resuspension in turbulent boundary layers and the influence of non-Gaussian removal forces. Journal of Aerosol Science, 58, 103–128.

[6] Henry, C. and Minier, J.-P. (2018). Colloidal particle resuspension: On the need for refined characterisation of surface roughness. Journal of Aerosol Science, 118, 1–13.

[7] Peillon, S., Autricque, A., Redolfi, M., Stancu, C., Gensdarmes, F., Grisolia, C., and Pluchery, O. (2019). Adhesion of tungsten particles on rough tungsten surfaces using Atomic Force Microscopy. Journal of Aerosol Science, 137, 105431.

[8] Ibrahim, A.H., Brach, R.M. and Dunn, P.F. (2004). Microparticle detachment from surfaces exposed to turbulent air flow: microparticle motion after detachment. Journal of Aerosol Science, 35, 1189-1204.

[9] Banari, A., Henry, C., Fank Eidt, R.H., Lorenz, P., Zimmer, K., Hampel, U. and Lécrivain, G. (2021). Evidence of collision-induced resuspension of microscopic particles from a monolayer deposit. Physical Review Fluids, 6, L082301.

[10]Rondeau A., Peillon S., Vidales A. M., Benito J., Uñac R., Sabroux J.-C. and Gensdarmes F. (2021) Evidence of inter-particle collision effect in airflow resuspension of poly-dispersed non-spherical tungsten particles in monolayer deposits. Journal of Aerosol Science, 154, 105735.

[11]Gelain, T., Gensdarmes, F., Peillon, S., and Ricciardi, L. (2020). CFD modelling of particle resuspension in a toroidal geometry resulting from airflows during a loss of vacuum accident (LOVA). Fusion Engineering and Design, 151, 111386.

[12]Benito, J., Uñac, R., Vidales, A., and Ippolito, I. (2016). Validation of the Monte Carlo model for resuspension phenomena. Journal of Aerosol Science, 100, 26–37

Funding category: Contrat doctoral

PHD Country: France