pith. sign in

arxiv: 2606.23368 · v1 · pith:66IUQG6Jnew · submitted 2026-06-22 · 💻 cs.CE · physics.comp-ph

A kinetic-diffusion Monte Carlo-based particle-level fluid-kinetic decomposition for neutral transport simulations

Zhirui Tang , Niels Horsten , Giovanni Samaey This is my paper

Pith reviewed 2026-06-26 06:36 UTC · model grok-4.3

classification 💻 cs.CE physics.comp-ph
keywords hybrid fluid-kinetic modelkinetic-diffusion Monte Carloneutral transportplasma edge simulationMonte Carlo methodsasymptotic-preserving schemescharge exchangeNavier-Stokes closure
0
0 comments X

The pith

A particle-level KDMC hybrid decomposes neutral transport into fluid and kinetic parts without iterative coupling and yields at least 500 times speedup over pure kinetic Monte Carlo.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper constructs a hybrid fluid-kinetic model for neutral particles by performing distribution decomposition directly at the particle level inside the kinetic-diffusion Monte Carlo framework. This construction inherits asymptotic preservation from KDMC and eliminates the need for iterative coupling or explicit interface handling between fluid and kinetic regions. A Navier-Stokes-type fluid closure is obtained through Hilbert-Chapman-Enskog expansion and paired with a tunable reflective boundary condition. One-dimensional tests demonstrate that the resulting scheme runs at least 500 times faster than full kinetic Monte Carlo while keeping relative L2 errors near 10 percent when charge exchange dominates. In regimes where charge exchange is not dominant, accuracy becomes more sensitive to boundary treatment because of the fluid approximation's limitations near boundaries.

Core claim

The central claim is that a distribution-decomposition hybrid model built at the particle level on the kinetic-diffusion Monte Carlo method combines the efficiency of a fluid description with the accuracy of a kinetic description without requiring iterative coupling between the two. The fluid component is closed by a Navier-Stokes-type system derived via Hilbert-Chapman-Enskog expansion, which requires substantially fewer nonlinear iterations than the AFN model used in SOLPS-ITER while delivering comparable accuracy. In one-dimensional tests the model achieves at least 500 times speedup over pure kinetic Monte Carlo with relative L2 errors around 10 percent in charge-exchange-dominant cases;

What carries the argument

The kinetic-diffusion Monte Carlo (KDMC) particle-level distribution decomposition, which splits particles into fluid and kinetic components at each step while remaining asymptotic-preserving and free of iterative coupling.

If this is right

  • The hybrid scheme can be applied directly to reactor-relevant neutral transport problems that are currently limited by the cost of full kinetic Monte Carlo.
  • The Navier-Stokes fluid closure derived for KDMC matches the accuracy of the AFN model but reduces the number of nonlinear iterations required.
  • The tunable reflective boundary condition provides a direct knob for trading accuracy against speed in different collisionality regimes.
  • Because the decomposition occurs at the particle level, the method avoids unphysical assumptions and domain-decomposition interfaces that appear in other hybrid approaches.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Extending the boundary treatment refinement suggested in the paper could widen the range of regimes where the 10 percent error target is met without sacrificing the speedup.
  • The particle-level decomposition may transfer to other transport problems that combine diffusive and ballistic regimes, such as radiation or phonon transport.
  • Because the method is asymptotic-preserving, it could serve as a building block for multi-scale plasma-edge codes that automatically transition between fluid and kinetic descriptions.

Load-bearing premise

The fluid approximation remains accurate enough near boundaries when the regime is not charge-exchange dominant.

What would settle it

A non-charge-exchange-dominant test case in which relative L2 errors exceed 10 percent after the tunable reflective boundary condition is optimized would show that the fluid component's boundary limitation prevents the claimed accuracy level.

Figures

Figures reproduced from arXiv: 2606.23368 by Giovanni Samaey, Niels Horsten, Zhirui Tang.

Figure 1
Figure 1. Figure 1: Illustration of KDMC. (a) In the first time step of size [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Kinetic MC simulation with a fixed time step [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The one dimension simulation domain x ∈ [xL, xR] with the absorbing BC imposed at xL and the specular reflective BC at xR. 6.1. Boundary conditions for fluid system Let the simulation domain be defined as x ∈ [xL, xR]. For the fluid system, including Eqs (11), (15), and (16), the specular reflective BC corresponds to enforcing zero normal flux at the boundary. For instance, the particle flux density at xR … view at source ↗
Figure 4
Figure 4. Figure 4: A particle crosses the divertor plate (physical boundary) at [PITH_FULL_IMAGE:figures/full_fig_p014_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Plasma background used in the simulation. (a) Plasma density [PITH_FULL_IMAGE:figures/full_fig_p016_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Comparison of fluid models: the proposed fluid system, the single-density model, and the AFN model. Each row shows the neutral [PITH_FULL_IMAGE:figures/full_fig_p017_6.png] view at source ↗
Figure 6
Figure 6. Figure 6: From the first row, we see that all fluid models fit well with the reference simulated using standard kinetic MC [PITH_FULL_IMAGE:figures/full_fig_p018_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparison between proposed hybrid model with [PITH_FULL_IMAGE:figures/full_fig_p019_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Comparison between proposed hybrid model with [PITH_FULL_IMAGE:figures/full_fig_p021_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: The collision rate profiles of the periodic test case: (a) Recombination rate; (b) Ionization rate; (c) Charge-exchange rate. [PITH_FULL_IMAGE:figures/full_fig_p022_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Comparison between the proposed hybrid model and the scheme in [29] with a periodic test case. (a) the neutral density [PITH_FULL_IMAGE:figures/full_fig_p022_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Standard deviations of the proposed hybrid model and the scheme in [29] for the periodic test case: (a) standard deviation of density, [PITH_FULL_IMAGE:figures/full_fig_p023_11.png] view at source ↗
read the original abstract

Neutrals in the plasma edge are commonly modeled by kinetic equations, with quantities of interest given by macroscopic quantities such as density, velocity, and temperature. In reactor-relevant regimes, fully kinetic descriptions solved by Monte Carlo (MC) methods, although accurate, become computationally expensive, whereas fluid-limit approximations are computationally more efficient but may lose accuracy due to boundary effects or low-collisional regimes. Hybrid fluid-kinetic approaches aim to combine the strengths of both descriptions. However, existing simulation methods face challenges, including interface handling in domain decomposition, unphysical assumptions, and iterative coupling in distribution decomposition. In this work, we propose a distribution-decomposition hybrid model constructed at the particle level based on the kinetic-diffusion Monte Carlo (KDMC) method. The model inherits key properties of KDMC: it is asymptotic-preserving and does not require iterative coupling between the fluid and kinetic components. To improve the accuracy of the fluid-part quantities estimation, a Navier-Stokes-type fluid system is derived via Hilbert-Chapman-Enskog expansions, tailored for KDMC. In the considered one-dimensional tests, the resulting fluid system has comparable accuracy to the AFN model used in SOLPS-ITER while requiring substantially fewer nonlinear iterations. Additionally, a tunable reflective boundary condition is introduced that allows balancing accuracy and efficiency. The model exhibits at least 500 times speedup over the kinetic MC, while maintaining relative L2 errors around 10% in a charge exchange (CX)-dominant test case. In non-CX-dominant regimes, the accuracy becomes increasingly sensitive to boundary treatment due to the inherent limitations of the fluid approximation near the boundary, motivating further refinement of the KDMC boundary conditions.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper proposes a particle-level fluid-kinetic decomposition for neutral transport in plasma edge simulations, built on the kinetic-diffusion Monte Carlo (KDMC) framework. It derives a Navier-Stokes-type fluid system via Hilbert-Chapman-Enskog expansion tailored to KDMC, introduces a tunable reflective boundary condition to balance accuracy and efficiency, and reports that the hybrid model is asymptotic-preserving with no iterative coupling. In 1D tests, it claims at least 500x speedup over pure kinetic MC while keeping relative L2 errors around 10% in a charge-exchange-dominant case; accuracy is noted to become more sensitive to boundary treatment in non-CX regimes due to fluid limitations.

Significance. If the performance and accuracy claims generalize, the method offers a promising route to efficient hybrid neutral transport modeling that inherits KDMC's asymptotic-preserving property and avoids iterative fluid-kinetic coupling. The tailored fluid derivation and tunable BC address practical limitations of existing hybrids. Credit is due for the particle-level construction and the explicit acknowledgment of boundary sensitivity as a limitation motivating further work.

major comments (2)
  1. [Numerical results section] Numerical results (1D test cases): the 500x speedup and ~10% relative L2 error are reported only for the CX-dominant regime; no quantitative data (error bars, data-selection criteria, or parameter sweeps) are given for how the tunable reflective BC parameter affects these metrics when the fluid approximation is active near boundaries in non-CX-dominant regimes, which directly bears on whether the headline performance numbers hold beyond the single reported case.
  2. [Fluid derivation section] Fluid derivation and interface (Hilbert-Chapman-Enskog section): the claim that the NS-type system preserves the asymptotic-preserving property of KDMC at the particle level requires explicit verification that the tunable reflective BC does not violate the Chapman-Enskog ordering at the fluid-kinetic interface; the abstract flags increasing boundary sensitivity in non-CX regimes, but no moment comparisons or ordering checks with/without the tunable parameter are described.
minor comments (2)
  1. [Methods] Notation for the tunable reflective BC parameter should be introduced with a clear symbol and range in the methods section rather than only in the abstract.
  2. [Figures] Figure captions for the 1D test results should include the exact value of the tunable BC parameter used and the collisionality regime for each panel.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and positive assessment of the work's potential. Below we provide point-by-point responses to the major comments.

read point-by-point responses

  1. Referee: [Numerical results section] Numerical results (1D test cases): the 500x speedup and ~10% relative L2 error are reported only for the CX-dominant regime; no quantitative data (error bars, data-selection criteria, or parameter sweeps) are given for how the tunable reflective BC parameter affects these metrics when the fluid approximation is active near boundaries in non-CX-dominant regimes, which directly bears on whether the headline performance numbers hold beyond the single reported case.

    Authors: The reported performance metrics are indeed specific to the CX-dominant regime, as indicated in the abstract. The manuscript already highlights that accuracy in non-CX regimes is more sensitive to boundary treatment due to fluid limitations. While we did not include parameter sweeps or error bars for the BC in non-CX cases in the original submission, we agree this would strengthen the claims. In the revised manuscript, we will add quantitative analysis of the tunable BC parameter's impact, including any available data or additional tests. revision: yes

  2. Referee: [Fluid derivation section] Fluid derivation and interface (Hilbert-Chapman-Enskog section): the claim that the NS-type system preserves the asymptotic-preserving property of KDMC at the particle level requires explicit verification that the tunable reflective BC does not violate the Chapman-Enskog ordering at the fluid-kinetic interface; the abstract flags increasing boundary sensitivity in non-CX regimes, but no moment comparisons or ordering checks with/without the tunable parameter are described.

    Authors: The asymptotic-preserving property is maintained by the particle-level decomposition and the KDMC framework itself, with the fluid system derived to be consistent via the Hilbert-Chapman-Enskog expansion. The tunable reflective BC is introduced as a practical adjustment at the boundary. However, the current manuscript does not provide explicit moment comparisons or ordering verifications at the interface with respect to the tunable parameter. We will include such checks in the revision to explicitly confirm that the BC does not violate the ordering. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained.

full rationale

The Navier-Stokes-type fluid system is obtained from standard Hilbert-Chapman-Enskog expansions applied to the kinetic model, with no indication that parameters are fitted to the target outputs and then relabeled as predictions. The reported 500x speedup and 10% L2 error are simulation outcomes on test cases, not quantities defined by construction from the same inputs. The tunable reflective boundary condition is introduced as an auxiliary device for balancing accuracy and efficiency rather than a fitted parameter whose value is then used to define the claimed performance. No load-bearing self-citation or uniqueness theorem imported from prior author work appears in the provided text. The central claims therefore rest on independent derivations and numerical experiments rather than reducing to the inputs by definition.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

The approach rests on the asymptotic-preserving property of KDMC (treated as given) and the validity of Hilbert-Chapman-Enskog expansion for the fluid closure; the tunable reflective boundary condition introduces one free parameter whose value is chosen per simulation.

free parameters (1)
  • tunable reflective boundary condition parameter
    Introduced to balance accuracy and efficiency; its specific value is not derived from first principles and must be selected for each case.
axioms (2)
  • standard math Hilbert-Chapman-Enskog expansion yields a suitable Navier-Stokes-type closure for the KDMC fluid component
    Invoked to derive the fluid system tailored for KDMC.
  • domain assumption KDMC is asymptotic-preserving and requires no iterative fluid-kinetic coupling
    Inherited from prior KDMC literature and used as foundation for the hybrid model.

pith-pipeline@v0.9.1-grok · 5836 in / 1470 out tokens · 26902 ms · 2026-06-26T06:36:28.950629+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

41 extracted references · 36 canonical work pages

  1. [1]

    Reiter, M

    D. Reiter, M. Baelmans, P. Börner, The EIRENE and B2-EIRENE Codes, Fusion Science and Technology 47 (2005) 172–186. doi:10.13182/FST47-172

    work page doi:10.13182/fst47-172 2005

  2. [2]

    Multi-scale deep neural network (mscalednn) methods for oscillatory stokes flows in complex domains

    B. Mortier, M. Baelmans, G. Samaey, A Study of Source Term Estimators in Coupled Finite-V olume/Monte- Carlo Methods with Applications to Plasma Edge Simulations in Nuclear Fusion: Track-Length and Next- Event Methods, Communications in Computational Physics 36 (2024) 1186–1218. doi:10.4208/cicp. OA-2023-0224

    work page doi:10.4208/cicp 2024

  3. [3]

    Fleck, E

    J. Fleck, E. Canfield, A random walk procedure for improving the computational efficiency of the implicit Monte Carlo method for nonlinear radiation transport, Journal of Computational Physics 54 (1984) 508–523. doi:10.1016/0021-9991(84)90130-X

    work page doi:10.1016/0021-9991(84)90130-x 1984

  4. [4]

    Börgers, E

    C. Börgers, E. W. Larsen, M. L. Adams, The asymptotic diffusion limit of a linear discontinuous discretization of a two-dimensional linear transport equation, Journal of Computational Physics 98 (1992) 285–300. doi:10. 1016/0021-9991(92)90143-M

    1992

  5. [5]

    Q. Sun, I. D. Boyd, G. V . Candler, A hybrid continuum/particle approach for modeling subsonic, rarefied gas flows, Journal of Computational Physics 194 (2004) 256–277. doi:10.1016/j.jcp.2003.09.005

    work page doi:10.1016/j.jcp.2003.09.005 2004

  6. [6]

    Blommaert, N

    M. Blommaert, N. Horsten, P. Börner, W. Dekeyser, A spatially hybrid fluid-kinetic neutral model for SOLPS- ITER plasma edge simulations, Nuclear Materials and Energy 19 (2019) 28–33. doi:10.1016/j.nme.2019. 01.022. 24

    work page doi:10.1016/j.nme.2019 2019

  7. [7]

    Horsten, M

    N. Horsten, M. Groth, M. Blommaert, W. Dekeyser, I. P. Pérez, S. Wiesen, Application of spatially hybrid fluid–kinetic neutral model on JET L-mode plasmas, Nuclear Materials and Energy 27 (2021) 100969. doi:10. 1016/j.nme.2021.100969

    arXiv 2021

  8. [8]

    Van Uytven, W

    W. Van Uytven, W. Dekeyser, M. Blommaert, N. Horsten, Y . Marandet, M. Baelmans, Advanced spatially hybrid fluid-kinetic modelling of plasma-edge neutrals and application to ITER case using SOLPS-ITER, Contributions to Plasma Physics 62 (2022) e202100191. doi:10.1002/ctpp.202100191

    work page doi:10.1002/ctpp.202100191 2022

  9. [9]

    C. F. F. Karney, D. P. Stotler, B. J. Braams, Modeling of neutral plasma in a divertor in the fluid-kinetic transition, Contributions to Plasma Physics 38 (1998) 319–324. doi:10.1002/ctpp.2150380148

    work page doi:10.1002/ctpp.2150380148 1998

  10. [10]

    Degond, G

    P. Degond, G. Dimarco, Fluid simulations with localized boltzmann upscaling by direct simulation Monte-Carlo, Journal of Computational Physics 231 (2012) 2414–2437. doi:10.1016/j.jcp.2011.11.030

    work page doi:10.1016/j.jcp.2011.11.030 2012

  11. [11]

    Degond, S

    P. Degond, S. Jin, L. Mieussens, A smooth transition model between kinetic and hydrodynamic equations, Journal of Computational Physics 209 (2005) 665–694. doi:10.1016/j.jcp.2005.03.025

    work page doi:10.1016/j.jcp.2005.03.025 2005

  12. [13]

    Degond, G

    P. Degond, G. Dimarco, L. Mieussens, A multiscale kinetic–fluid solver with dynamic localization of kinetic effects, Journal of Computational Physics 229 (2010) 4907–4933. doi:10.1016/j.jcp.2010.03.009

    work page doi:10.1016/j.jcp.2010.03.009 2010

  13. [14]

    Crouseilles, P

    N. Crouseilles, P. Degond, M. Lemou, A hybrid kinetic/fluid model for solving the gas dynamics Boltzmann– BGK equation, Journal of Computational Physics 199 (2004) 776–808. doi:10.1016/j.jcp.2004.03.007

    work page doi:10.1016/j.jcp.2004.03.007 2004

  14. [15]

    Tiwari, A

    S. Tiwari, A. Klar, S. Hardt, A particle–particle hybrid method for kinetic and continuum equations, Journal of Computational Physics 228 (2009) 7109–7124. doi:10.1016/j.jcp.2009.06.019

    work page doi:10.1016/j.jcp.2009.06.019 2009

  15. [16]

    Kolobov, R

    V . Kolobov, R. Arslanbekov, V . Aristov, A. Frolova, S. Zabelok, Unified solver for rarefied and continuum flows with adaptive mesh and algorithm refinement, Journal of Computational Physics 223 (2007) 589–608. doi:10.1016/j.jcp.2006.09.021

    work page doi:10.1016/j.jcp.2006.09.021 2007

  16. [17]

    Valentinuzzi, Y

    M. Valentinuzzi, Y . Marandet, H. Bufferand, G. Ciraolo, P. Tamain, Two-phases hybrid model for neutrals, Nuclear Materials and Energy 18 (2019) 41–45. doi:10.1016/j.nme.2018.12.003

    work page doi:10.1016/j.nme.2018.12.003 2019

  17. [18]

    Horsten, G

    N. Horsten, G. Samaey, M. Baelmans, A hybrid fluid-kinetic model for hydrogenic atoms in the plasma edge of tokamaks based on a micro-macro decomposition of the kinetic equation, Journal of Computational Physics 409 (2020) 109308. doi:10.1016/j.jcp.2020.109308

    work page doi:10.1016/j.jcp.2020.109308 2020

  18. [19]

    Horsten, W

    N. Horsten, W. Dekeyser, M. Blommaert, G. Samaey, M. Baelmans, A hybrid fluid–kinetic neutral model based on a micro–macro decomposition in the SOLPS-ITER plasma edge code suite, Contributions to Plasma Physics 60 (2020) e201900132. doi:10.1002/ctpp.201900132

    work page doi:10.1002/ctpp.201900132 2020

  19. [21]

    M. A. Katsoulakis, A. J. Majda, A. Sopasakis, Multiscale Couplings In Prototype Hybrid Deterministic/Stochas- tic Systems: Part I, Deterministic Closures, Communications in Mathematical Sciences 2 (2004) 255–294. doi:10.4310/CMS.2004.v2.n2.a7

    work page doi:10.4310/cms.2004.v2.n2.a7 2004

  20. [22]

    Le Tallec, F

    P. Le Tallec, F. Mallinger, Coupling Boltzmann and Navier–Stokes Equations by Half Fluxes, Journal of Computational Physics 136 (1997) 51–67. doi:10.1006/jcph.1997.5729

    work page doi:10.1006/jcph.1997.5729 1997

  21. [23]

    Mortier, M

    B. Mortier, M. Baelmans, G. Samaey, A Kinetic-Diffusion Asymptotic-Preserving Monte Carlo Algorithm for the Boltzmann-BGK Model in the Diffusive Scaling, SIAM Journal on Scientific Computing 44 (2022) A720– A744. doi:10.1137/20M1381526. 25

    work page doi:10.1137/20m1381526 2022

  22. [24]

    Efficient asymptotic-preserving schemes for some multiscale kinetic equations

    S. Jin, Efficient Asymptotic-Preserving (AP) Schemes For Some Multiscale Kinetic Equations, SIAM Journal on Scientific Computing 21 (1999) 441–454. doi:10.1137/S1064827598334599

    work page doi:10.1137/s1064827598334599 1999

  23. [25]

    Dimarco, L

    G. Dimarco, L. Pareschi, Hybrid Multiscale Methods II. Kinetic Equations, Multiscale Modeling & Simulation 6 (2008) 1169–1197. doi:10.1137/070680916

    work page doi:10.1137/070680916 2008

  24. [26]

    Crestetto, N

    A. Crestetto, N. Crouseilles, M. Lemou, A particle micro-macro decomposition based numerical scheme for collisional kinetic equations in the diffusion scaling, Communications in Mathematical Sciences 16 (2018) 887–911. doi:10.4310/CMS.2018.v16.n4.a1

    work page doi:10.4310/cms.2018.v16.n4.a1 2018

  25. [27]

    Løvbak, G

    E. Løvbak, G. Samaey, Accelerated Simulation of Boltzmann-BGK Equations near the Diffusive Limit with Asymptotic-Preserving Multilevel Monte Carlo, SIAM Journal on Scientific Computing 45 (2023) A1862– A1889. doi:10.1137/22M1498498

    work page doi:10.1137/22m1498498 2023

  26. [28]

    Mortier, P

    B. Mortier, P. Robbe, M. Baelmans, G. Samaey, Multilevel asymptotic-preserving Monte Carlo for kinetic- diffusive particle simulations of the Boltzmann-BGK equation, Journal of Computational Physics 450 (2022) 110736. doi:10.1016/j.jcp.2021.110736

    work page doi:10.1016/j.jcp.2021.110736 2022

  27. [29]

    Mortier, V

    B. Mortier, V . Maes, G. Samaey, Estimation as a post-processing step for random walk approximations of the Boltzmann-BGK model, Contributions to Plasma Physics 62 (2022). doi:10.1002/ctpp.202100197

    work page doi:10.1002/ctpp.202100197 2022

  28. [30]

    I. Lux, L. Koblinger, Monte Carlo Particle Transport Methods: Neutron and Photon Calculations, 1 ed., CRC Press, 2018. doi:10.1201/9781351074834

    work page doi:10.1201/9781351074834 2018

  29. [31]

    Z. Tang, J. Koellermeier, E. Løvbak, G. Samaey, Analysis of kinetic-diffusion Monte Carlo simulation and source term estimation scheme in nuclear fusion applications, 2025. doi:10.48550/arXiv.2512.23580. arXiv:2512.23580

    work page doi:10.48550/arxiv.2512.23580 2025

  30. [32]

    V . Maes, W. Dekeyser, J. Koellermeier, M. Baelmans, G. Samaey, Hilbert expansion based fluid models for kinetic equations describing neutral particles in the plasma edge of a fusion device, Physics of Plasmas 30 (2023) 063907. doi:10.1063/5.0146158

    work page doi:10.1063/5.0146158 2023

  31. [33]

    Horsten, W

    N. Horsten, W. Dekeyser, G. Samaey, M. Baelmans, Comparison of fluid neutral models for one-dimensional plasma edge modeling with a finite volume solution of the Boltzmann equation, Physics of Plasmas 23 (2016) 012510. doi:10.1063/1.4940309

    work page doi:10.1063/1.4940309 2016

  32. [34]

    Representation in

    B. Lapeyre, É. Pardoux, R. Sentis, Introduction to Monte-Carlo Methods for Transport and Diffusion Equations, Oxford University PressOxford, 2003. doi:10.1093/oso/9780198525929.001.0001

    work page doi:10.1093/oso/9780198525929.001.0001 2003

  33. [35]

    Horsten, G

    N. Horsten, G. Samaey, M. Baelmans, Development and assessment of 2D fluid neutral models that in- clude atomic databases and a microscopic reflection model, Nuclear Fusion 57 (2017) 116043. doi:10.1088/ 1741-4326/aa8009

    2017

  34. [36]

    Cercignani, R

    C. Cercignani, R. Illner, M. Pulvirenti, The Mathematical Theory of Dilute Gases, volume 106 ofApplied Mathematical Sciences, Springer New York, New York, NY , 1994. doi:10.1007/978-1-4419-8524-8

    work page doi:10.1007/978-1-4419-8524-8 1994

  35. [37]

    Harris, An Introduction to the Theory of the Boltzmann Equation, Dover Books on Physics, Dover Publica- tions, Mineola, N.Y ., 2004

    S. Harris, An Introduction to the Theory of the Boltzmann Equation, Dover Books on Physics, Dover Publica- tions, Mineola, N.Y ., 2004

    2004

  36. [39]

    MacNamara, G

    S. MacNamara, G. Strang, Operator Splitting, in: R. Glowinski, S. J. Osher, W. Yin (Eds.), Splitting Methods in Communication, Imaging, Science, and Engineering, Springer International Publishing, Cham, 2016, pp. 95–114. doi:10.1007/978-3-319-41589-5_3. 26

    work page doi:10.1007/978-3-319-41589-5_3 2016

  37. [40]

    Dimarco, L

    G. Dimarco, L. Pareschi, G. Samaey, Asymptotic-Preserving Monte Carlo Methods for Transport Equations in the Diffusive Limit, SIAM Journal on Scientific Computing 40 (2018) A504–A528. doi:10.1137/17M1140741

    work page doi:10.1137/17m1140741 2018

  38. [41]

    Steel, V

    T. Steel, V . Maes, G. Samaey, Fluid boundary conditions in kinetic-diffusion Monte Carlo, 2025. doi:10.48550/ arXiv.2509.03942.arXiv:2509.03942

    arXiv 2025

  39. [42]

    Kukushkin, H

    A. Kukushkin, H. Pacher, V . Kotov, D. Reiter, D. Coster, G. Pacher, Effect of conditions for gas recirculation on divertor operation in ITER, Nuclear Fusion 47 (2007) 698. doi:10.1088/0029-5515/47/7/021

    work page doi:10.1088/0029-5515/47/7/021 2007

  40. [43]

    B. P. Welford, Note on a Method for Calculating Corrected Sums of Squares and Products, Technometrics 4 (1962) 419–420. doi:10.1080/00401706.1962.10490022

    work page doi:10.1080/00401706.1962.10490022 1962

  41. [44]

    Horsten, G

    N. Horsten, G. Samaey, M. Baelmans, Hybrid fluid–kinetic neutral model based on a kinetic correction on a fluid solution, Contributions to Plasma Physics 58 (2018) 703–709. doi:10.1002/ctpp.201700179. 27

    work page doi:10.1002/ctpp.201700179 2018