Session Z25: Particles-Laden Compressible Flows

Investigation of Mach number and relative position effects on aerodynamic interference between two particles by direct numerical simulation

In the present study, the effects of Mach number and relative position on aerodynamic interference between two particles were investigated by direct numerical simulations of compressible Navier-Stokes equations. The particle Reynolds number was set to be 150, and the particle Mach number was set to be between 0.3 and 1.2. The apparent attractive and repulsive directions acting between the two particles were calculated from the aerodynamic coefficients of those particles. The apparent force in the repulsive direction is generated only in the side-by-side arrangement, and the force in the attractive direction is generated in the in-line and diagonal arrangements. In addition, attractive and repulsive forces are larger at high Mach numbers than that at low Mach numbers. This result suggests that the influence of interparticle interaction on the particle distribution in compressible particle-laden flows might be stronger than that in incompressible cases. However, the aerodynamic interference is significantly reduced at the Mach number of 1.2 with a certain condition. This is because the influence of particles on the flow field is spatially limited due to the formation of the bow shock. The critical distance is related to the angle of the detached shock wave.   

Millimeter wave interferometry in optically thick particle-laden flows: an application to plume-surface interactions

Millimeter wave radar interferometry provides a unique capability for measuring the volume fraction of particles in dispersed multiphase mixtures. It allows the measurement of particle concentrations in optically thick flows independently of the size distribution of the dispersed phase, in contrast with state-of-the-art methods based on optical attenuation and scattering. In this work, a proof-of-concept based on a COTS mmWave radar measures the concentration of particles in ejecta clouds generated by the impingement of a Mach 5 jet on a granular surface. This method is extended from a single path-integrated measurement to a multi-path tomographic measurement.

The instrument is calibrated under vacuum against a slot funnel producing gravity-driven particle sheets with known particle concentrations. The dielectric permittivity of the particle material is measured via the waveguide method in a separate experiment. Interferometric measurements are performed on a plume-surface impingement experiment using a 6.5 N cold gas thruster replicating the ambient pressures, Mach numbers and pressure ratios of lunar and martian landings.   

Unsteady Compressible Pairwise Interaction Extended Point-Particle Model

An efficient pairwise interaction framework is developed using acoustic wave solution and compressible Maxey-Riley-Gatignol (MRG) equation to predict unsteady forces on a particle subjected to an incoming planar shock, in the presence of other neighboring particles. The time-dependent perturbing effect of each neighbor is evaluated with the acoustic approximation and the total perturbing influence of all the neighbors is calculated with a summation of all the neighbors. Comparing the force signal obtained with the model with that experienced by particle-resolved (PR) simulations, it is established that (i) the compressible MRG model is capable of capturing the unsteady nature of force as the primary shock and its reflections off the neighbors pass over the particle, and (ii) superposition of perturbations by each neighbor is a reasonable approximation. Furthermore, even though in computing the perturbation flow the acoustic equation was used, the framework is extendable to higher shock Mach number cases, and if needed PR simulations can be used.   

Validation of a Shell-Shedding Model for Reacting, Liquid Ejecta Particles

Results from ejecta experiments performed at Los Alamos National Laboratory show that liquid metal particles ejected from a shocked surface show vastly different behavior depending on whether their accepting medium is inert or reactive. Those ejected into an inert medium demonstrate an expected deceleration behavior with velocities which are monotonically decreasing at a near constant rate. On the other hand, those ejected into a reactive, hydrogen-based medium exhibit a staged deceleration phenomenon in recorded LDV data. A hypothesis is that the ejecta particles travelling in the reactive medium form a hydride shell and the presence of the shell leads to physical processes which cause the non-monotonic deceleration. The specific physical processes which control this phenomenon are mostly unknown and are of interest for accurate models for simulating the trajectories of the metal particles after ejection. This work describes the development of point-particle based models for the hydride shell meant to simulate shedding of small particles from the hydride shell surface along with any phase changes simultaneously occurring in the shell. A framework for coupling the mass and energy transfer of the shed particles to the surrounding fluid flow is also described before discussing ongoing model validation efforts against the liquid, reactive ejecta experimental data.

    

Modeling pseudo-turbulence for shock-induced flow through particle suspensions

Shock waves interacting with a suspension of particles in the compressible flow regime give rise to complex microscale phenomena – primarily, drag and particle-induced velocity fluctuations termed pseudo-turbulent kinetic energy (PTKE). Particle-resolved simulations of a planar shock interacting with a dense suspension of particles are performed to quantify the budget of PTKE. We propose a two-equation model compatible with coarse-grained simulation techniques such as Euler-Lagrange and Euler-Euler methods in which PTKE and its dissipation are transported. Comparisons are made against existing algebraic models from the literature.   

Author not Attending

Previously published results from the Horizontal Shock Tube (HST) facility at Los Alamos National Laboratory showed unsteady drag an order of magnitude above the standard drag curve. However, forensic investigation of these results showed evidence that a faulty assumption of the particle size during processing of the data led to this result rather than new physics. To confirm the results of the forensic investigation, new experimental data is needed. In this presentation, the previous experiments are re-created but with a tightly controlled diameter. The experiment involves solid Nylon particles nominally 4 μm in diameter in dilute suspensions subject to shocks of Mach 1.2, 1.3, and 1.4. An eight-pulse particle tracking diagnostic measures individual particle positions, and a shadowgraph system measures shock location, with pressure transducers providing shock speed at the test section. These diagnostics give us detailed measurements of particle positions versus time. From the particle positions, empirical fits are performed to determine the unsteady particle drag. The newly obtained data with the tightly constrained particle diameters will be compared with the previously published data.