CoChat

Plasma propulsion modeling with particle-based algorithms

Open access
Taccogna, F., Cichocki, F., Eremin, D., Fubiani, G., Garrigues, L. · 2023 · Journal of Applied Physics

The DOI matches its Crossref record; 344 of the 344 references it cites were found in a registry or an index; 78 of the 190 sources its claims cite were checked for support.

Abstract

Perspective overview of PIC/MCC models for plasma propulsion and recent computational progress including plasma chemistry Monte Carlo modules for molecular and alternative propellants.

Overview

CoChat Summary

TL;DR

This perspective paper reviews particle-in-cell/Monte Carlo collision modeling of electric propulsion, covering electrostatic and electromagnetic thrusters, plasma plumes, plasma-wall interaction, and computational acceleration methods. It argues that particle-based models are essential for kinetic, non-local thruster physics but remain limited by computational cost, input-data gaps, and simplified electron/plume treatments.

Key findings

  • PIC/Monte Carlo models are presented as the main kinetic tool for low-pressure, partially magnetized electric-thruster plasmas, with ES thrusters requiring Poisson’s equation and EM thrusters requiring a larger set of Maxwell’s equations.
  • Computational demands are severe: for pulsed arc thrusters, plasma density peaks at \(10^{22}\ \mathrm{m^{-3}}\), giving \(\lambda_{D,e}\approx O(10^{-7})\ \mathrm{m}\) and \(\Delta t\approx O(10^{-13})\ \mathrm{s}\), requiring about \(10^8\) time steps for a few \(\mu\)s; other ES thrusters typically reach \(10^{18}\ \mathrm{m^{-3}}\), \(\lambda_{D,e}\approx O(10^{-5})\ \mathrm{m}\), \(\Delta t\approx O(10^{-11})\ \mathrm{s}\), and about \(10^7\) time steps.
  • New numerical methods report large gains: sparse-grid ES PIC achieved a computational-time gain of 6.5 for the same resources in a Hall-thruster ECDI case; reduced-dimensional-order PIC showed speed-ups up to 50 versus 2D and an expected quasi-3D speed-up of about 2500 versus full 3D.
  • Plasma-wall and plume studies highlight important effects: ion flux to unshielded walls can be up to 40% of total ion production; including non-neutral regions near a spacecraft increased ion backflow current by approximately 20%; reversed propellant injection increased utilization efficiency by a maximum of 30%; one ECR magnetic-field shape increased extraction efficiency by 50%.

Methods

  • Design: a perspective/review article, not a primary experimental or numerical study; it synthesizes particle-based modeling approaches for plasma propulsion.
  • Data/sample: literature and prior simulation/experimental studies across ES devices (E×B, pulsed arc, electrospray, gridded ion engines) and EM devices (RF inductive, helicon, ECR, MPDT), plus plume and spacecraft-interaction models.
  • Analysis: narrative overview of PIC/MCC algorithms, plasma-wall interaction models, hybrid versus full-PIC plume approaches, HPC techniques, and emerging methods such as sparse grids, implicit/energy-conserving PIC, reduced-order PIC, and deep-learning-based PIC.

Limitations

  • The paper states that explicit PIC is constrained by Debye-length, plasma-period, and CFL restrictions, making multidimensional high-density simulations computationally expensive; numerical tricks such as enlarged permittivity or mass/size scaling are still used.
  • It states that hybrid plume models with polytropic or Boltzmann electrons miss kinetic electron cooling, anisotropy, and peripheral/backflow plasma behavior, and that low-energy data for secondary electron emission, sputtering, neutral reflection, and ion recombination are lacking.
  • Evident from the text: as a perspective/review, it presents no new primary data, no single validation study, and its forward-looking claims about future methods are authors’ assessments rather than demonstrated results.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1063/5.0153862

Cites

344 of 344 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

FIG. 1 Scheme of a typical PIC-Monte Carlo loop
FIG. 1 Scheme of a typical PIC-Monte Carlo loopFrom the PDF
FIG. 2 Three dimensional simulation of beamlets extraction, acceleration and coalescence in a GIE with multiple apertures, obtained with a hybrid model. Subplots (a,c,e) and (b,d,f) refer to different thruster cross sections. Reproduced with permission from Perales et al. 72 , Plasma Sources Science and Technology 30 , 105023 (2021). Copyright 2021 IOP Publishing Ltd.
FIG. 2 Three dimensional simulation of beamlets extraction, acceleration and coalescence in a GIE with multiple apertures, obtained with a hybrid model. Subplots (a,c,e) and (b,d,f) refer to different thruster cross sections. Reproduced with permission from Perales et al. 72 , Plasma Sources Science and Technology 30 , 105023 (2021). Copyright 2021 IOP Publishing Ltd.From the PDF
FIG. 3 Axial x - azimuthal y distributions of the azimuthal electric field E y and of the ion density n i at a given time step. The minimum and maximum values for E y are -5 × 10 4 and 5 × 10 4 V/m, respectively, and for n i are 0 and 5 × 10 17 m -3 . Adapted with permission from Boeuf et al. . 90 , Phys. Plasmas 25 , 061204 (2018). Copyright 2018 AIP Publishing LLC.
FIG. 3 Axial x - azimuthal y distributions of the azimuthal electric field E y and of the ion density n i at a given time step. The minimum and maximum values for E y are -5 × 10 4 and 5 × 10 4 V/m, respectively, and for n i are 0 and 5 × 10 17 m -3 . Adapted with permission from Boeuf et al. . 90 , Phys. Plasmas 25 , 061204 (2018). Copyright 2018 AIP Publishing LLC.From the PDF
FIG. 4 1D azimuthal FFT of the azimuthal electric field E y, averaged over all radial positions and over three temporal intervals obtained by seven ES PIC models implemented by different research groups. MTSI and ECDI resonances are indicated by arrows. Reproduced with permission from Villafana et al. 99 , Plasma Sources Science and Technology 30 , 075002 (2021). Copyright 2021 IOP Publishing Ltd.
FIG. 4 1D azimuthal FFT of the azimuthal electric field E y, averaged over all radial positions and over three temporal intervals obtained by seven ES PIC models implemented by different research groups. MTSI and ECDI resonances are indicated by arrows. Reproduced with permission from Villafana et al. 99 , Plasma Sources Science and Technology 30 , 075002 (2021). Copyright 2021 IOP Publishing Ltd.From the PDF
FIG. 5 The ( r , z ) profiles of electron temperature, electron density, ionization rate, and electrostatic potential in the µ 10 ECR thruster. Reproduced with permission from Yamashita et al. 116 , Physics of Plasmas 26 , 073510 (2019), Copyright 2019 AIP Publishing.
FIG. 5 The ( r , z ) profiles of electron temperature, electron density, ionization rate, and electrostatic potential in the µ 10 ECR thruster. Reproduced with permission from Yamashita et al. 116 , Physics of Plasmas 26 , 073510 (2019), Copyright 2019 AIP Publishing.From the PDF
Image 6
From the PDF
FIG. 7 Elastic backscattering lobes for Al surface with an incident electron energy E p = 40 eV and two different incident angle: (a) θ p = 10° and (b) θ p = 45°. Reproduced with permission from Villemant et al. 203 , IEPC-2017-366 (2017), Copyright 2017 Electric Rocket Propulsion Society.
FIG. 7 Elastic backscattering lobes for Al surface with an incident electron energy E p = 40 eV and two different incident angle: (a) θ p = 10° and (b) θ p = 45°. Reproduced with permission from Villemant et al. 203 , IEPC-2017-366 (2017), Copyright 2017 Electric Rocket Propulsion Society.From the PDF
FIG. 8 Trajectories of neutrals inside a cylindrical discharge chamber. Propellant atoms are injected from the left boundary, and get ionized by an isothermal electrons fluid. The dashed red line refers to the trajectory of a specularly reflected neutral, the black solid line to a purely diffused neutral, and the blue dash-dot line to a neutral reflected according to Schamberg's model. Reproduced with permission from Domínguez-Vázquez et al. 219 , Plasma Sources Science and Technology 30 , 085004 (2021), Plasma Sources Science and Technology. Copyright 2021 IOP Publishing Ltd.
FIG. 8 Trajectories of neutrals inside a cylindrical discharge chamber. Propellant atoms are injected from the left boundary, and get ionized by an isothermal electrons fluid. The dashed red line refers to the trajectory of a specularly reflected neutral, the black solid line to a purely diffused neutral, and the blue dash-dot line to a neutral reflected according to Schamberg's model. Reproduced with permission from Domínguez-Vázquez et al. 219 , Plasma Sources Science and Technology 30 , 085004 (2021), Plasma Sources Science and Technology. Copyright 2021 IOP Publishing Ltd.From the PDF
FIG. 9 Electric potential in the surrounding of a cubic satellite, predicted by a hybrid model, with polytropic electrons modeled with γ = 1 . 1 and non-neutral regions taken into account. Chargeexchange collisions for ions are included. Adapted with permission from Cichocki et al. , 229 , Plasma Sources Science and Technology 26 , 125008 (2017). Copyright 2017 IOP Publishing.
FIG. 9 Electric potential in the surrounding of a cubic satellite, predicted by a hybrid model, with polytropic electrons modeled with γ = 1 . 1 and non-neutral regions taken into account. Chargeexchange collisions for ions are included. Adapted with permission from Cichocki et al. , 229 , Plasma Sources Science and Technology 26 , 125008 (2017). Copyright 2017 IOP Publishing.From the PDF
Image 10
From the PDF
FIG. 11 Ion (top) and electron (bottom) relative density differences, between a quasi-neutral hybrid PIC simulation with isothermal Boltzmann's electrons, and a full PIC simulation (simulation B of Ref.245). Adapted with permission from Wang et al. 245 , Physics of Plasmas 26 , 103502 (2019). Copyright 2019, AIP Publishing.
FIG. 11 Ion (top) and electron (bottom) relative density differences, between a quasi-neutral hybrid PIC simulation with isothermal Boltzmann's electrons, and a full PIC simulation (simulation B of Ref.245). Adapted with permission from Wang et al. 245 , Physics of Plasmas 26 , 103502 (2019). Copyright 2019, AIP Publishing.From the PDF
FIG. 12 Electric potential in a 2D axisymmetric plasma plume simulation for different domain extensions, and using open boundary conditions for electrons. Reproduced with permission from Li et al. 247 , Plasma Sources Science and Technology 28 , 034004 (2019). Copyright 2019, IOP Publishing.
FIG. 12 Electric potential in a 2D axisymmetric plasma plume simulation for different domain extensions, and using open boundary conditions for electrons. Reproduced with permission from Li et al. 247 , Plasma Sources Science and Technology 28 , 034004 (2019). Copyright 2019, IOP Publishing.From the PDF
FIG. 13 Ion density predicted with a hybrid plume model with polytropic electrons (inertialess, unmagnetized and collisionless) and γ = 1 . 3, in a complex geometry scenario, corresponding to the DAWN spacecraft. Charge exchange collisions for ions are included. Courtesy from S.J. Araki 234 , IEEE Transactions on Plasma Science 47 , 4898-4908 (2019).
FIG. 13 Ion density predicted with a hybrid plume model with polytropic electrons (inertialess, unmagnetized and collisionless) and γ = 1 . 3, in a complex geometry scenario, corresponding to the DAWN spacecraft. Charge exchange collisions for ions are included. Courtesy from S.J. Araki 234 , IEEE Transactions on Plasma Science 47 , 4898-4908 (2019).From the PDF
FIG. 14 Antenna gain radiation pattern for two scenarios, in the antenna emission plane ( ϕ = 90 deg.) and in the plane perpendicular to the emission ( ϕ = 0 deg.) with (ON) and without (OFF) the plasma. Reproduced with permission from De Mejanes et al. , Journal of Applied Physics 131 , 243303 (2022). Copyright 2022 AIP Publishing LLC.
FIG. 14 Antenna gain radiation pattern for two scenarios, in the antenna emission plane ( ϕ = 90 deg.) and in the plane perpendicular to the emission ( ϕ = 0 deg.) with (ON) and without (OFF) the plasma. Reproduced with permission from De Mejanes et al. , Journal of Applied Physics 131 , 243303 (2022). Copyright 2022 AIP Publishing LLC.From the PDF
FIG. 15 Scheme of a coupled PIC-MCC / TPMC simulation for a HT discharge simulation. The plasma (gas) modules see respectively the neutral gas (plasma) as fixed non-uniform backgrounds with density ng ( n e ), Reproduced with permission from Taccogna et al. 57 , Frontiers in Physics 10 (2022). Copyright 2022 Frontiers media SA.
FIG. 15 Scheme of a coupled PIC-MCC / TPMC simulation for a HT discharge simulation. The plasma (gas) modules see respectively the neutral gas (plasma) as fixed non-uniform backgrounds with density ng ( n e ), Reproduced with permission from Taccogna et al. 57 , Frontiers in Physics 10 (2022). Copyright 2022 Frontiers media SA.From the PDF
FIG. 16 Collisional energy loss per electron-ion pair created, W s (eV), versus the electron temperature T e (eV) in Xe and air relevant species (N, N2, O, O2 and an air mixture 0.5 N2 / 0.5 O) 57 and in atomic (I) and molecular (I2 ) iodine 272 . Atomic species are shown by solid lines, while molecular species by dashed lines.
FIG. 16 Collisional energy loss per electron-ion pair created, W s (eV), versus the electron temperature T e (eV) in Xe and air relevant species (N, N2, O, O2 and an air mixture 0.5 N2 / 0.5 O) 57 and in atomic (I) and molecular (I2 ) iodine 272 . Atomic species are shown by solid lines, while molecular species by dashed lines.From the PDF
FIG. 17 Xe excitation rate coefficients 282 from ground to any excited state (exc.tot.) and from ground to metastable states (exc.to 1s5, exc.to 1s3), and Xe ionization rate coefficients of the ground state 282 (ion.gnd) and of the two metastable states 283 (ion.1s5, ion.1s3). Paschen notation is here considered.
FIG. 17 Xe excitation rate coefficients 282 from ground to any excited state (exc.tot.) and from ground to metastable states (exc.to 1s5, exc.to 1s3), and Xe ionization rate coefficients of the ground state 282 (ion.gnd) and of the two metastable states 283 (ion.1s5, ion.1s3). Paschen notation is here considered.From the PDF
FIG. 18 Axial x - azimuthal y simulations of the HT ECDI for a computational domain of 512 × 512 grid cells. 2D profiles of ion density at a given timestep, with (left) standard ES PIC, and (right) sparse grid ES PIC. The maximum ion density is 2 . 2 × 10 17 m -3 and 2 × 10 17 m -3 , in left and right figures, respectively. The number of particles-per-cell is the same (400), and the total number of particles is 10 8 and 5 . 2 × 10 6 , respectively. Reproduced with permission from L. Garrigues et al. 303 , Journal of Applied Physics 129 , 153304 (2021). Copyright 2021 AIP Publishing LLC.
FIG. 18 Axial x - azimuthal y simulations of the HT ECDI for a computational domain of 512 × 512 grid cells. 2D profiles of ion density at a given timestep, with (left) standard ES PIC, and (right) sparse grid ES PIC. The maximum ion density is 2 . 2 × 10 17 m -3 and 2 × 10 17 m -3 , in left and right figures, respectively. The number of particles-per-cell is the same (400), and the total number of particles is 10 8 and 5 . 2 × 10 6 , respectively. Reproduced with permission from L. Garrigues et al. 303 , Journal of Applied Physics 129 , 153304 (2021). Copyright 2021 AIP Publishing LLC.From the PDF
Image 19
From the PDF

Verification

Claims the paper makes

14 Valid · 44 Not valid · 34 Not checked · 63 Own results and interpretations

57 more claims · 15 with their cited sources checked

Create a free account to see all 155 claims and to verify the 4 references still open.

See all claims — free account
  • Probably not a substantial claim. 48
  • It may state two separate claims as one. 67
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Citations

Cites

Items 201–344 of 344

Research works printed but not linked: 20

  1. 2005, Plasma Physics via Computer Simulation
  2. 2013, The DSMC Method
  3. 2021, Python Programming and Numerical Methods. A Guide for Engineers and Scientists
  4. 1988, Computer Simulation Using Particles
  5. 2010, Plasma Simulations by Example
  6. 2005, Computational Electrodynamics: The Finite-Difference Time-Domain Method
  7. 2007, Perfectly Matched Layer (PML) for Computational Electromagnetics
  8. 2018, Energy dissipation in helicon plasma at the near field of an antenna, Probl. At. Sci. Technol., 6, 86
  9. 2023, Density profile transition and high-energy electron transport in a magnetically expanding radio frequency plasma, Phys. Plasmas, 30, 013509, 10.1063/5.0126901 · DOI · Paper page
  10. 1957, On a singularity of the field of an electromagnetic wave propagated in an lnhomogeneous plasma, Sov. Phys. JETP, 4, 544
  11. 1996, Electron Cyclotron Resonance Ion Sources and ECR Plasmas
  12. D. Sydorenko , “Particle-in-cell simulations of electron dynamics in low pressure discharges with magnetic fields,” Ph.D. thesis (Department of Physics and Engineering Physics, University of Saskatchewan, 2006), unpublished thesis.
  13. 2006, Gaseous Electronics. Theory and Practice
  14. 2012, Gaseous Electronics. Tables, Atoms, and Molecules
  15. 2022, Programming Massively Parallel Processors: A Hands-on Approach
  16. 2014, Professional CUDA C Programming
  17. 2015, Heterogeneous Computing with OpenCL 2.0
  18. 2016, Parallel Programming with OpenACC
  19. 1981, A direct method for implicit particle-in-cell simulation, Commun. Plasma Phys. Control. Fusion, 6, 225
  20. 2021, GPU Porting of Scalable Implicit Solver with Green’s Function-Based Neural Networks by Openacc

Cited by

See all 29

Metadata

Authors
Taccogna, F., Cichocki, F., Eremin, D., Fubiani, G., & Garrigues, L.
Affiliations
  • Institute for the Science and Technology of Plasmas (IT)
  • National Agency for New Technologies, Energy and Sustainable Economic Development (IT)
  • Ruhr University Bochum (DE)
  • Centre National de la Recherche Scientifique (FR)
Venue
Journal of Applied Physics
Published
2023-10-19
Volume
134
Issue
15
ISSN
  • 0021-8979 (ISSN-L)
  • 1089-7550
OpenAlex
W4387781340
Cited by
29
References
327
Citations by year
2026: 6 · 2025: 12 · 2024: 11
FWCI
2.15
Type
journal article
Language
en
Open access
Locations
  • Journal of Applied Physics (publishedVersion)
  • HAL (submittedVersion, other-oa)
Funders
  • Ministero dell'Università e della Ricerca (PON ARS01-00141)
  • Agence Nationale de la Recherche (ANR) under the projet (ANR-22-CE46-0012)
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Low-Pressure Plasma Processing and Diagnostics
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Vacuum and Plasma Arcs
  • Physical Sciences › Mathematics › Applied Mathematics › Gas Dynamics and Kinetic Theory
Keywords
plasma propulsion, particle-in-cell simulation, Monte Carlo collisions, electrostatic thrusters, electron cyclotron resonance thruster, plasma plume, plasma-wall interaction, plasma chemistry, alternative propellants

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-12, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-12 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.