ISMSE 2024 Lunar dust simulation - plume regolith interaction

A combined Eulerian-Lagrangian approach to simulation of plume-regolith interaction during the descent and ascent phase of the lunar lander

1. A combined Eulerian-Lagrangian approach to simulation of plume-regolith interaction during the descent and ascent phase of the lunar lander Armen Jaworski*, Bruno Bras**, Jeroen van den Eynde**, Tomasz Gajek*, Aleksander Jurecki*, Przemysław Kosewski*, Łukasz Lech* and Riccardo Rampini** *CIM-mes Projekt Sp. z o.o ., Warsaw, Poland, [email protected] ** European Space Agency , Noordwijk, The Netherlands, [email protected] ISMSE 16th edition of International Symposium on Materials in the Space Environment and the ICPMSE 14th International Conference on Protection of Materials and Structures in the Space Environment Saint-Raphaël, 7-1 1 October 2023

2. Motivation Lunar dust problem 2 Lunar dust particles are a severe threat to different sensitive surfaces Technical challenge - include major factors impacting the contamination transport: plume-regolith interaction regolith erosion, craters electrostatics particle-particle interactions Estimate surface contamination (on the lander and surroundings) Estimate risk of surface damage due to kinetic impingement Photo: courtesy of ESA

3. MoonDUST – ESA project Modelling dust impact in lunar conditions Prediction of particulate contamination after landing and ascent at the Moon and after docking to the orbital station Simulation of the engine plume and dust interaction Geometry + BC Simulation of the descent phase Simulation of the ascent phase Particulate contamination distribution Lunar vehicles dust associated risks Simulation of the docking phase Orbiting station dust associated risks Image source: https://doi.org/10.3390/aerospace9070358 3

4. No Problem Approach Done 1 Large number of particles – at the order of 7x10 11 Eulerian / Lagrangian, in future, also stochastic approach 2 Large differences in gas density – combined molecular and continuous flow FVM-DSMC flow solution / only FVM 3 Impact of the electrostatic forces Modelling the electrostatic forces , initial conditions taken from SPIS software, each particle impact on the electrostatics forces 4 Strong interaction between particles (collisions) Modelling collisions in Lagrangian /Stochastic 5 Complex regolith erosion due to plume impingement Using Robert’s model 6 Crater formation Predefined constant lunar surface geometry 7 Particle surface interactions Custom deposition models 8 Irregular particle shapes May be included in future (stochastic approach) Modelling challenges Problems to solve - Approach 4 Courtesy of NASA

5. Particle-Surface interaction Deposition models No Deposition model Key features 1 Wetted wall Particle cannot be resuspended after impact 2 Escape model Particle removed from the simulation after impact (for outlets) 3 Energy-based by Dahneke (1971) Energy balance before and after the impact with wall energy barrier 4 Momentum-based model by Kosinski and Hoffmann (2009) Account for change of momentum after rebound 5 Momentum-based model by Breuer and Almohammed (2015) Substrate compression is also considered More accurate rebound velo c ity Few orders of magnitude difference for smallest particles rebound attached Impact of electrostatics

6. Resuspension models Flow, electrostatics, vibrations and shocks No Model 1 Vainshtein, Ziskind, Fichman, and Gutfinger (VZFG) 2 Rock and Roll model 3 NRG3 model 4 User subroutine Flow-driven resuspension can be simulated using built-in models and user-subroutine The particle is detached from a surface when the accumulated energy exceeds the potential energy The vibrationally driven resuspension is simulated using a separate model Adhesive forces includes : van der Waals and contact mechanics, capillary, electrostatic, inertia forces 6

7. SPIS potential field Imported field Electrostatics External field & inter particle forces 1. External electric field SPIS data Force: , - particle charge, - electric field 2. Inter particle electrostatic interaction Particle in cel method (PIC) Efficient for large number of particles   2D PIC algorithm example -> implemented in 3D Source: https://boltzplatz.eu/intro-particle-in-cell/ Splitting to nodes Solve electric field Interpolation back to particle Calculate electrostatic force

8. Regolith erosion model Roberts’ model 8 Designed for Apollo landings , still commonly used in the literature It provides eroded mass using the surface shear stress (τ) and the threshold soil shear strength (τ*) More accurate for larger altitudes     Mass distribution JSC-1A R Flow data Courtesy of NASA

9. Large number of interacting particles Particle modelling approach 9 Particle Concentration Euler Lagrange It is impossible to simulate all dust particle interactions – 10 11 particles The computed number of particles can be reduced using stochastic or/and Eulerian approach

10. Euler approach 10 A. Simple scalar transport   B. Mode complex 5 equation model Particle concentration as a scalar field Particles are introduced as a new flow phase with a certain volume fraction ( α ) Flow data as input [1] Ray, M. et al., An Euler-Euler model for mono-dispersed gas-particle flows incorporating electrostatic charging due to particle-wall and particle-particle collisions , Chemical Engineering Science 197, 2019, DOI: 10.1016/j.ces.2018.12.028 [2] Kolehmainen, J. et al., Eulerian modelling of gas–solid flows with triboelectric charging , J. Fluid Mech. (2018), vol. 848, pp. 340–369 , doi:10.1017/jfm.2018.361 exchange of momentum with collisions, transport of momentum electrostatic force drag force gravitational forces interfacial force s between phases complex solver

11. Argonaut test-case: flow simulation FVM flow solution 11 High thrust Low thrust Plume flow during descent/ascent Strong flow interaction for multi-engine configuration Strong plume deflection from the middle engine

12. Argonaut test-case: particle collisions Impact of collisions (Lagrange) 12 No collisions Collisions Implemented forces: Drag, Lift, Brownian, Turbulence, Gravity Collision model:

13. Argonaut test-case : thrust impact Lagrangian model Particles > 250 um, JSC-1a simulant size distribution Dust release according to Roberts’ law Height = 1 m 8 x higher thrust 13

14. Euler + Lagrange Estimate contamination PPM rate calculated from combined Euler + Lagrange Overall contamination estimated using the descent/ascent profile PPM rate Euler Lagrange + = ascent / descent profile Estimate surface contamination (on the lander and surroundings)

15. Conclusions Framework for lunar dust modelling The simulation models for plume-dust interactions developed Contamination levels can be estimated after descent/ascent A software tool is being developed to be used by engineers and researchers Activity funded by the European Space Agency Disclaimer: the view expressed in this presentation can in no way be taken to reflect the official opinion of the European Space Agency Please contact me if you want to use the tool or collaborate on this topic Armen Jaworski [phone] a.jaworski @cim-mes.com.pl