The effects of secondary electron emission on plasma sheath characteristics and electron transport in an ${\bf{E}}\times {\bf{B}}$ discharge via kinetic simulations

Abstract

Hall-effect thrusters, which are electrostatic devices based on an E B plasma discharge, have successfully been used as satellite propulsion systems for the last few decades. However, the presence of anomalous electron cross-field transport is still poorly understood, and involves complex and strongly coupled mechanisms such as azimuthal electron drift instabilities and intense secondary electron emission (SEE) from the thruster walls. The present work focuses on the relative importance of these two phenomena. We use a 2D particle-in-cell/Monte Carlo collision model configured to simulate the radial-azimuthal directions near the thruster exit plane. A constant radial magnetic field and axial electric field are imposed, and electron drift instabilities are observed in the azimuthal (E B) direction. A simplified SEE model is implemented and an extensive parametric study is performed to directly determine the effect on electron transport. It is found that, for the operating conditions used in our simulations, SEE enhances the near-wall electron mobility by a factor 2, while reducing the bulk plasma mobility by about 20% (due to electron cooling). However, the dominant contribution to anomalous electron transport is still observed to be caused by electron drift instabilities driven by the E B discharge configuration. SEE modifies the electron mobility profile, but the spatially-averaged value remains relatively constant. Three different operating regimes are identified depending on the SEE rate value: two that are stable, and a third which shows an oscillatory behaviour. In addition to electron transport, the kinetic simulations give new insight into the plasma sheath formation at the radial walls, and comparison with typical analytical sheath models demonstrate important differences.

2D axial-azimuthal particle-in-cell benchmark for low-temperature partially magnetized plasmas

“…The code is parallelized via MPI through a domain decomposition. It has been verified with the 1D helium benchmark of Turner et al [19] (further details in [28,29]) and extensively used to simulate the radial-azimuthal plane of a Hall Thruster [28,30,31]. For this benchmark, the code was adapted to the axial-azimuthal plane and accelerated via a load-balancing algorithm (adjusting periodically the size of each processor domain to have approximately the same number of particles inside each processor).…”

Collisionless ion modeling in Hall thrusters: Analytical axial velocity distribution function and heat flux closures

“…As a mean to assess the quality of our results, we have performed three Particle-In-Cell Monte Carlo-Collisions (PIC-MCC) simulations, with the code LPPic that was already verified via the 1D Helium benchmark of Turner et al 38 and extensively used to simulate the radial-azimuthal plane of a Hall Thruster. 17,39 The code was adapted to simulate the axial-azimuthal plane and we present here three cases with a simulation model similar to the one used in the 2D benchmark of Charoy et al 40 In this case, an axial electric field is created by a potential difference imposed between an anode at 200 V and a cathode at 0 V (see Fig. 15).…”

Latest progress in Hall thrusters plasma modelling

“…Finally, a regime of relaxation sheath oscillations observed in 1D-radial model (Sydorenko et al 2009) has been also observed in the 2D(r,q) model of Tavant et al (2018). In particular, three regimes have been identified when varying the effective SEE coefficient.…”

“…Secondly, short wavelength instabilities with a significantly large component along the magnetic field reduce to the unmagnetized (ion-sound) form as seen from the analytical solution of the dispersion relation. Recent models (Héron-Adam 2013, Hara-Cho 2017, Croes et al 2017, Jahnunen et al 2018b, Tavant et al 2018, Taccogna et al 2019 have added the radial coordinate resolving the direction along the magnetic field and accounting for the sheath boundary conditions. They have shown that the full ion sound regime is not realized, but a more effective regime of discrete cyclotron resonance instabilities (again inverse cascade towards low wavenumbers) occur.…”

“…Another important aspect that these models can handle is the coupling between SEE and ExB EDI assessing the relative importance of these two phenomena on the electron anomalous transport. It has been recently shown (Tavant et al 2018, Taccogna et al 2019) that the electron radial heating due to ExB EDI can induce a larger SEE, but this, in turn, can help the ExB EDI saturation by cooling the electrons. As a result, it was observed that the anomalous electron cross-field mobility in the bulk plasma (induced by ExB EDI) decreases by 20% as the effective SEE coefficient increases.…”