Pith. sign in

REVIEW 5 major objections 4 minor 49 references

A Concurrent Multiscale Framework Coupling Direct Simulation Monte Carlo and Molecular Dynamics

T0 review · 5 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A cell-based coupling of DSMC and molecular dynamics, built on lifting and restricting operators, replaces simplified gas-surface models with atomistic detail and cuts the mean surface-temperature error in a hypersonic test case from 730%…

desk verdict Genuinely useful coupling machinery with an under-validated, likely mis-implemented temporal bridge; deserves review but not acceptance without fixing Eq. 19. read the letter →

arxiv 2506.01924 v1 pith:QS4UBGZS submitted 2025-06-02 physics.comp-ph cond-mat.mes-hallcond-mat.stat-mech

Tim Linke , Dane Sterbentz , Niels Gr{o}nbech-Jensen , Jean-Pierre Delplanque , Jonathan Belof This is my paper
classification physics.comp-phcond-mat.mes-hallcond-mat.stat-mech
keywords multiscalemodelingconcurrentcouplingdirectsimulationMonteCarlomoleculardynamicsequation-freecomputationgas-surfaceinteractionshypersonicflowliftingandrestrictingoperators
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper describes a way to run Direct Simulation Monte Carlo (DSMC) and molecular dynamics (MD) together in the same simulation, with a cell-by-cell decision about where atomistic detail is needed. The central claim is that two new translation operators—a lifting operator that turns DSMC cell data into an MD initial state and a restricting operator that turns MD results back into DSMC particles—let DSMC replace its simplified gas-surface interaction models with full atomistic collisions without fixing a coupling region in advance. The method is verified by showing that mass is conserved and temperature fluctuations stay within theoretical finite-size bounds. In a Mach 5.84 flow over a small thermocouple wire, the coupled method predicts a mean surface temperature of 500.2 K against the experimental mean of 523.2 K, a 4.4% error, where conventional DSMC gives 4342.7 K, a 730% error. The paper argues this opens up problems that DSMC could not previously handle accurately, such as gas-surface energy transfer and surface chemistry.

What carries the argument

The load-bearing machinery is the pair of lifting and restricting operators that bridge DSMC and MD. On lifting, the DSMC cell's aspect ratio and number density fix the MD box, and each DSMC particle is replaced by a circular patch of atoms on a close-packed lattice whose radius comes from the asymptotic lattice-point count $N(r)\approx 2\pi r^2/(\sqrt{3}a^2)$; choosing the lattice spacing equal to the interatomic potential cut-off keeps the configuration near an energy minimum. On restricting, the MD atoms are sorted by a nearest-neighbor criterion and grouped exactly $f_{MD}^{num}$ at a time, with position and velocity averaged into one DSMC particle, which cancels the extra information the lattice introduced. The temporal bridge uses a thermostatted bulk layer and a Stefan-Boltzmann radiative-loss term applied over the DSMC timestep of $10^{-5}$ s while the MD run lasts only 50,000 steps of $10^{-15}$ s. This pair of operators carries the argument because it is what lets a coarse stochastic method and a fine deterministic method exchange conserved quantities without a fixed coupling zone.

What would settle it

Run the same hypersonic wire case with the MD instance stopped after 20,000, 50,000, and 100,000 timesteps, and check whether the temperature returned to DSMC changes; if the coupled mean surface temperature moves by much more than the claimed 4.4% error across these durations, the steady-state bridging assumption is not converged. A second check is to turn off the analytical radiative-loss thermostat in the bulk wire layer: if the predicted surface temperature then departs from the experimental mean by far more than 4.4%, the accuracy gain is carried mainly by the temporal-balancing formula rather than by the atomistic gas-surface collision description.

Watch

Extended reading notes

Core claim

On its own terms, the paper's discovery is that a concurrent DSMC-MD coupling can be made general by deciding per DSMC cell whether to run an MD instance, rather than fixing an overlap buffer or a database of precomputed MD results. The lifting operator rescales a DSMC cell to an MD box of the same number density, decomposes each simulated DSMC particle into a cluster of MD atoms arranged on a hexagonal close-packed lattice, and assigns velocities from the DSMC phase space. The restricting operator sorts the MD atoms by nearest neighbor, regroups exactly the same number of atoms per DSMC particle, and returns a center-of-mass position and average velocity to DSMC. Applied to the hypersonic thermocouple example, this replaces the DSMC surface temperature model and reduces the mean surface-temperature error from 730% to 4.4%, while the steady cooling rate matches experiment within 8.2%. The paper also reports that the method fails to reproduce the transient initial heating and the late sharp cool-off in the experiment, attributing those to unsteady inflow effects not included in the simulation.

Load-bearing premise

The long-time accuracy depends on the assumption that a few tens of picoseconds of MD, run to a quasi-steady state with an analytical radiative-cooling thermostat, represent what the gas-surface interaction would do over the full $10^{-5}$ s DSMC timestep.

Editorial extensions

If this is right

  • Selecting MD cells on a per-cell basis removes the need to know in advance where atomistic detail is required.
  • DSMC surface treatment can be upgraded from accommodation-coefficient models to deterministic atomistic collisions, improving mean surface temperature prediction from a 730% error to a 4.4% error in the hypersonic example.
  • The steady-state cooling rate of the wire is reproduced within 8.2% of experiment, giving confidence in heat-transfer predictions for steady hypersonic flows.
  • The same coupling is claimed to apply to evaporation and condensation, gas-surface chemistry validation, foreign-particle impacts, fusion-relevant rarefied plasmas, and micro- and nanofluidic surface phenomena.
  • Because the restricting operator returns a single center-of-mass particle, the DSMC particle count and mass are conserved exactly across coupling steps.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A direct extension would be to run the MD instances only occasionally and use them to refresh gas-surface accommodation coefficients in DSMC, turning the concurrent scheme into a self-refreshing database.
  • The reported miss of the initial transient suggests the method, as presented, is best trusted for quasi-steady surface response; a testable next step is whether time-varying inlet pressure reproduces the initial heating.
  • Because the lifting operator seeds atoms on a lattice whose spacing equals the potential cut-off, the results may be sensitive to the chosen interatomic potential; a sensitivity study across cut-off radii would clarify how much of the accuracy gain is method versus potential.
  • The cell-based decision rule could be driven by a local criterion such as the Knudsen number or a non-equilibrium measure, making the framework adaptive rather than dependent on a user-chosen cell list.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. The manuscript presents a concurrent, cell-based coupling of Direct Simulation Monte Carlo (DSMC) and molecular dynamics (MD). DSMC is the macroscopic driver; selected DSMC cells are replaced, during a DSMC step, by MD simulations. The lifting operator maps cell geometry, particle positions, velocities, and f_num to an MD domain, while the restricting operator groups MD atoms into DSMC particles and returns averaged positions and velocities. The method is implemented in SPARTA and LAMMPS. The authors verify mass and temperature behavior on a small atomic-oxygen system and then apply the method to a Mach 5.84 hypersonic flow over a thermocouple wire, comparing DSMC-MD, standard DSMC, and experiment. They report a mean surface temperature of 500.2 K for DSMC-MD versus 523.2 K experimental (4.4% error), versus 4342.7 K for standard DSMC (730% error). They conclude that the framework improves accuracy by orders of magnitude and discuss limitations.

Significance. Cell-based concurrent DSMC-MD coupling is a useful idea that generalizes existing buffer-zone and indirect coupling approaches. The lifting and restricting operators are clearly described, the implementation in open-source codes is a strength, and the hypersonic example demonstrates the potential for replacing phenomenological gas-surface models with atomistic detail. If the temporal-bridging and baseline-comparison issues are resolved, the method could become a valuable tool. However, the accuracy gain as reported is not yet convincing: the radiative loss in the MD domain uses the same model it claims to replace, the DSMC baseline appears improperly resolved, and the verification tests are too weak to establish operator correctness.

major comments (5)
  1. [Sec. 5.1, Eqs. (16)-(19)] The temporal bridging is internally inconsistent and unvalidated. The text states that Delta-t in Eq. (19) is the DSMC timestep (10^-5 s) and then says that at the beginning of each MD time step the wire atoms are initialized to T_target = T_surf - Delta-t*sigma*epsilon*T^4/(rho*c*d). If the decrement intended for the full DSMC step is applied at each of the 50,000 MD steps (1 fs each), the cumulative radiative cooling exceeds the physical value by roughly a factor of 50,000; if it is applied only once per coupling call, the sentence is wrong. No sensitivity study is provided for rho, c, d, epsilon, the thermostat relaxation time, or the averaging window, and no evidence is given that a 50 ps MD run reaches the same steady state as the DSMC cell would over 10^-5 s. The 4.4% mean agreement in Fig. 12 is therefore not yet a validated accuracy claim for the lifting and restricting operators; it is at least partly determined by this imposed energy balance.
  2. [Sec. 5.1, simulation set-up] The DSMC timestep is inconsistent with the stated cell-transit condition. With V=1001.14 m/s and Delta-t=10^-5 s, a particle travels about 10.0 mm per timestep, which is roughly 17 cell lengths given Delta-x=0.6 mm. This directly contradicts the statement that Delta-t is chosen so that particles do not traverse an entire cell within one timestep. The baseline DSMC run is therefore not a properly resolved DSMC simulation, and the reported 730% error in Sec. 5.2 is not a meaningful benchmark for the accuracy gain. A corrected comparison with a DSMC timestep satisfying the CFL condition is needed before the improvement can be attributed to the coupling method.
  3. [Sec. 5.1, Eqs. (14) and (19)] The radiative cooling term in the MD domain uses the same Stefan-Boltzmann gray-body law and the same emissivity (derived from Eq. (15)) as the DSMC surface model that the coupling is intended to replace. The close agreement between the DSMC-MD mean temperature (500.2 K) and experiment (523.2 K) does not, by itself, validate the MD description; it shows that an analytic radiation law with the same parameters is being enforced in the MD energy balance. The claim in Sec. 5.2 that the MD domain bypasses the inaccurate surface temperature model is therefore overstated: the accommodation and collision dynamics are indeed atomistic, but the radiative cooling is not independent.
  4. [Sec. 4.1-4.2] The verification of the operators is too weak to support the claim that mass and temperature are correctly transferred. In Sec. 4.1, mass conservation is checked only by observing that the average particle count per cell remains near 15; however, the restricting operator groups exactly f_MD_num atoms into one DSMC particle, so mass conservation is enforced by construction and the test cannot detect errors in the lifting/restricting operators. In Sec. 4.2, the temperature check compares the measured standard deviation to the theoretical value from Eq. (13), but this does not verify that velocity distributions, energy, or higher moments are preserved. The statement in Sec. 4.1 that these results 'firmly prove a conservation of mass' is not supported by the presented evidence.
  5. [Sec. 3.1, text after Eq. (12)] The assertion that setting the hcp lattice spacing a equal to the interatomic cutoff distance r_c 'ensures an energy minimum' is not generally valid for a Lennard-Jones or embedded-atom potential. For a truncated potential, the energy at r_c is not necessarily a minimum, and for a smooth cutoff the force may not vanish there. The lattice spacing is a free parameter that affects the initial configuration energy and hence the MD dynamics; an energy minimization or a direct check of the initial configuration energy should be provided.
minor comments (4)
  1. [Sec. 4.1-4.2] The text is inconsistent about the coupling frequency: Sec. 4.1 says three randomly selected cells are fully coupled at every DSMC timestep, whereas Sec. 4.2 says 'the simulations with a coupling of DSMC-MD for every cell every 10 timesteps.' Please clarify which protocol corresponds to Figures 6-8.
  2. [Sec. 6, Conclusion] The claim of an improvement of 'three order of magnitudes' is not supported by the reported numbers: 730% versus 4.4% is a factor of about 166, not 1000. Please correct the wording.
  3. [Sec. 5.1, Eq. (19)] The area-to-volume ratio used in the radiative loss formula is given as 1/d, which is not the correct geometric factor for a cylindrical wire; for a cylinder of radius r, A/V would be 2/r (or a related expression depending on how the simulated thickness is defined). This affects the magnitude of the radiative decrement and should be stated explicitly.
  4. [Sec. 5.1, thermostat reference] The Nose-Hoover thermostat is cited to Shinoda et al. (2004); the original references are Nosé (1984) and Hoover (1985), which should be cited instead.

Circularity Check

2 steps flagged · score 3.0 of 10

Mass-conservation verification is by construction and the MD radiative-cooling term reuses the DSMC surface model, but the central lifting/restricting operators and MD heating dynamics carry independent content.

  1. self definitional [Section 3.2 (restricting operator) and Section 4.1 (Conservation of Mass)]
    ""To maintain the same mass of one DSMC particle, the same number of MD atoms is grouped to form a DSMC particle as those initially created during the lifting step." ... "These results firmly prove a conservation of mass of the cell-based coupling approach.""

    Mass conservation is baked into the definition of the restricting operator: each DSMC particle is decomposed into exactly f_MD_num atoms on lifting and re-formed from exactly f_MD_num atoms on restriction. The total number of DSMC particles, and hence the total mass, is therefore conserved by construction. The 100 ms numerical test cannot fail this invariant except through an implementation bug; presenting it as a verification that 'proves' mass conservation is a self-definitional check, not an independent test of the coupling physics.

  2. fitted input called prediction [Section 5.1, Equations (14)-(19) and the paragraph beginning 'The balance between atom-surface interactions...']
    ""The balance between atom-surface interactions and radiation heat loss is designed to emulate the DSMC surface model closely to enable a direct comparison." ... "Rewriting the Stefan-Boltzmann law (Eq. 14) ... ΔT= Δt σϵT^4/(ρcd).""

    The MD 'prediction' of the thermocouple surface temperature uses the same Stefan-Boltzmann gray-body law (Eq. 14) and the same emissivity (from Eq. 15, calibrated to the experimental wire conductivity) as the DSMC surface temperature model that the framework is claimed to bypass. The radiative-cooling side of the MD energy balance is therefore an input borrowed from the baseline model rather than an atomistic prediction. The reported 4.4% mean-temperature agreement is hence not fully independent of the DSMC model being compared against, because one side of the energy balance is shared by construction. The MD-computed heating is independent, so this is a partial reduction, not a complete equivalence.

full rationale

The paper's central claim—a cell-based concurrent DSMC-MD coupling with novel lifting and restricting operators—is not circular. The operators are defined through explicit geometric and statistical mappings (Eqs. 3-12), and the temperature verification in Sec. 4.2 is an independent statistical check against the analytic finite-size fluctuation formula (Eq. 13). The framework is implemented in external open-source codes (SPARTA and LAMMPS) and compared with experimental data [41], so no load-bearing self-citation chain is present; co-author citations such as [24] and [38] are peripheral. Two partial circularities are identified but neither collapses the central claim. First, the mass-conservation 'verification' is tautological, since the restricting operator is defined to regroup exactly f_MD_num atoms per DSMC particle, making mass conservation an identity of the algorithm rather than a demonstrated conservation law. Second, the hypersonic demonstration's radiative-loss term is the same Stefan-Boltzmann law with the same emissivity used by the DSMC surface model it replaces, so part of the predicted surface temperature is an input to the MD simulation; the MD atomistic heating term, however, is independent and provides genuine content. The temporal-bridging choice described around Eq. 19—applying the DSMC-scale radiative decrement in the MD run—is a validation and correctness risk rather than a circularity, as it concerns whether the bridging assumption is physically consistent. Overall, the core coupling derivation is self-contained, and the circular steps affect only a supporting verification and the independence of the accuracy comparison, giving a score of 3.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central result depends on several user-chosen parameters and strong modeling assumptions, most notably the temporal bridging between MD and DSMC and the analytical radiative loss term.

free parameters (6)
  • f_num_MD = Not reported; examples use 15 and 19
    Number of MD atoms representing one DSMC particle; chosen by user, affects MD domain size and statistical accuracy.
  • z_MD = Not reported
    MD domain thickness in the out-of-plane direction; chosen by user to conserve number density.
  • lattice_spacing_a = Set equal to cutoff distance r_c
    Initial spacing of the hexagonal close-packed lattice; chosen to minimize configurational energy, affects MD dynamics.
  • equilibration_time = 10^4 timesteps for thermostat, 50,000 timesteps for coupled run
    MD run duration; chosen conservatively to reach equilibrium, determines the reported temperature.
  • balanced_layer_thickness = At least 10 atoms
    Thickness of the region where temperature is measured; affects captured temperature gradient.
  • emissivity_prefactor = 1e-4
    Semi-theoretical prefactor in Atallah's relationship for emissivity; from literature, used in the radiative loss model.
assumptions (6)
  • standard math SPARTA and LAMMPS correctly implement DSMC and MD dynamics.
    The paper relies on these codes for the actual simulations without verifying their internals.
  • domain assumption Interatomic potentials (Lennard-Jones for air, EAM for nickel, Lorentz-Berthelot mixing) accurately model the physical interactions.
    Section 5.1 defines the potentials; the accuracy of the application depends on these choices.
  • domain assumption Conserving number density between DSMC and MD domains is sufficient to maintain physical consistency.
    Section 3.1 derives the MD domain size from number density conservation, assuming this is the key invariant.
  • ad hoc to paper A hexagonal close-packed lattice with spacing equal to the cutoff distance is an energy-minimizing and physically valid initialization.
    Section 3.1 justifies this to avoid atomic overlaps; no comparison to other initialization methods is provided.
  • ad hoc to paper A short MD run reaching steady state can represent flow conditions over a much longer DSMC timestep.
    Section 5.1 bridges time scales by assuming equilibrium after 50,000 MD steps, a critical premise for the coupling.
  • ad hoc to paper The Stefan-Boltzmann based radiative loss term applied each MD step emulates the wire's radiative cooling.
    Section 5.1 introduces this to match the DSMC surface model; it is an analytical add-on, not derived from MD.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Concurrent Multiscale Framework Coupling Direct Simulation Monte Carlo and Molecular Dynamics." pith.science (2026). https://pith.science/paper/QS4UBGZS

@misc{pith2026250601924,
  author       = {Pith},
  title        = {Pith review of: A Concurrent Multiscale Framework Coupling Direct Simulation Monte Carlo and Molecular Dynamics},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QS4UBGZS}},
  note         = {Machine review of arXiv:2506.01924}
}
read the original abstract

We present a new method to couple the Direct Simulation Monte Carlo (DSMC) algorithm with molecular dynamics (MD). The coupling approach generalizes prior coupling methods using a cell-based decision. The approach is supported by a lifting and restricting operator, which translate length and time scales between DSMC and MD. We verify the framework on basic conservation laws, and demonstrate its usability on a hypersonic flow example. Its accuracy gain is discussed in light of conventional DSMC simulations. Advantages and limitations are discussed, and additional use cases are highlighted.

Figures

Figures reproduced from arXiv: 2506.01924 by the authors.

Figure 1
Figure 1. Concurrent coupling approaches. The direct coupling (left) uses an overlapping region between [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Equation-free coupling. The cell-based approach allows MD to supplement constitutive relations [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Decomposition of DSMC particles to MD atoms. Atoms are initialized in a circular region across [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (10 more)
Figure 4
Figure 4. Figure 4: The lifting operator. The DSMC geometry and its particles are mapped to an initial MD domain [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: The restricting operator. Atomistic insights gained in the MD domain are provided to the DSMC [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: DSMC particle count per cell. Fluctuations in number density are observed for DSMC (blue) and [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: Temperature per cell. Fluctuations are observed in DSMC (blue) and DSMC-MD cells (gold) with [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]
Figure 8
Figure 8. Figure 8: Average temperature in DSMC. The measured temperature is observed over time for a DSMC [PITH_FULL_IMAGE:figures/full_fig_p015_8.png]
Figure 9
Figure 9. Figure 9: Experimental Set Up of Stagnation Temperature Measurements in a Hypersonic Flow (Not to [PITH_FULL_IMAGE:figures/full_fig_p017_9.png]
Figure 10
Figure 10. Figure 10: The molecular dynamics domain. Air (green) and nickel (red) atoms are divided into three [PITH_FULL_IMAGE:figures/full_fig_p019_10.png]
Figure 11
Figure 11. Figure 11: Temperature evolution in the hypersonic flow. The DSMC simulation cannot accurately capture [PITH_FULL_IMAGE:figures/full_fig_p021_11.png]
Figure 12
Figure 12. Figure 12: A detailed comparison between the coupled DSMC-MD approach and the experimental data. [PITH_FULL_IMAGE:figures/full_fig_p022_12.png]
Figure 13
Figure 13. Figure 13: Comparison of cooling rates between three approaches. The experimental data is compared to a [PITH_FULL_IMAGE:figures/full_fig_p024_13.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

49 extracted references · 49 canonical work pages

  1. [1]

    Votta,Hypersonic high altitude aerothermodynamics of a space re-entry vehicle, Aerospace Science and Technology, 2013

    R. Votta,Hypersonic high altitude aerothermodynamics of a space re-entry vehicle, Aerospace Science and Technology, 2013

    work page 2013

  2. [2]

    Wang & Z

    M. Wang & Z. Li,Simulations for the gas flows in microgeometries using the direct simulation Monte Carlo method, International Journal of Heat and Fluid Flow, 2004

    work page 2004

  3. [3]

    S. Rauf, K. Bera, J. Kenney, P. Kothnur,Modeling of plasma processing reactors: review and perspective, Journal of Micro/Nanopatterning, Materials and Metrology 22, 2023

    work page 2023

  4. [4]

    Kawagoe,A study on pressure-driven gas transport in porous media: from nanoscale to microscale, Microfluidics and Nanofluidics, 2016

    Y. Kawagoe,A study on pressure-driven gas transport in porous media: from nanoscale to microscale, Microfluidics and Nanofluidics, 2016

    work page 2016

  5. [5]

    Gallis & J

    M. Gallis & J. Harvey,Modelling of chemical reactions in hypersonic rarefied flow with the direct simulation Monte Carlo method, Journal of Fluid Mechanics, 1996

    work page 1996

  6. [6]

    Lee,Fundamentals of Theoretical Plasma Physics: Mathematical Description of Plasma Waves, World Scientific Publishing Company, 2019

    H.J. Lee,Fundamentals of Theoretical Plasma Physics: Mathematical Description of Plasma Waves, World Scientific Publishing Company, 2019

    work page 2019

  7. [7]

    Sone,Molecular Gas Dynamics, Springer, 2007

    Y. Sone,Molecular Gas Dynamics, Springer, 2007

    work page 2007

  8. [8]

    Sone,Kinetic Theory and Fluid Dynamics, Springer, 2002

    Y. Sone,Kinetic Theory and Fluid Dynamics, Springer, 2002

    work page 2002

Show all 49 references
  1. [9]

    Bird,Molecular Gas Dynamics and the Direct Simulation of Gas Flows

    G.A. Bird,Molecular Gas Dynamics and the Direct Simulation of Gas Flows. New York, NY, USA: Oxford Science Publications, 1994

    1994

  2. [10]

    Kardar,Statistical Physics of Particles, Cambridge University Press, 2007

    M. Kardar,Statistical Physics of Particles, Cambridge University Press, 2007

    2007

  3. [11]

    Mott-Smith,The Solution of the Boltzmann Equation for a Shock Wave, Physical Review, 1951

    H. Mott-Smith,The Solution of the Boltzmann Equation for a Shock Wave, Physical Review, 1951

    1951

  4. [12]

    Chapman, T

    S. Chapman, T. Cowling,The Mathematical Theory of Non-Uniform Gases, Cambridge University Press, 1990

    1990

  5. [13]

    S. J. Plimpton, S. G. Moore, A. Borner, A. K. Stagg, T. P. Koehler, J. R. Torczynski, M. A. Gallis,Direct Simulation Monte Carlo on petaflop supercomputers and beyond, Physics of Fluids, 31, 2019

    2019

  6. [14]

    Koura & H

    K. Koura & H. MatsumotoVariable soft sphere molecular model for air species, Phys. Fluids 4, (1992)

    1992

  7. [15]

    M. A. Gallis, T. P. Koehler, J. R. Torczynski, and S. J. Plimpton,Direct simulation Monte Carlo investigation of the Richtmyer-Meshkov instability, Physics of Fluids 27, 2015

    2015

  8. [16]

    M. A. Gallis, T. P. Koehler, J. R. Torczynski, and S. J. Plimpton,Direct simulation Monte Carlo investigation of the Rayleigh-Taylor instability, Physical Review Fluids 1, 2016. 27

    2016

  9. [17]

    M. A. Gallis, T. P. Koehler, J. R. Torczynski, S. J. Plimpton, and G. Papadakis, Molecular-level simulations of turbulence and its decay, Physical Review Letters 118, 2017

    2017

  10. [18]

    M. A. Gallis, T. P. Koehler, J. R. Torczynski, S. J. Plimpton, and G. Papadakis,Gas- kinetic simulation of sustained turbulence in minimal Couette flow, Physical Review Fluids 3, 2018

    2018

  11. [19]

    Garcia,Nonequilibrium fluctuations studied by rarefied-gas simulation, Physical Re- view A 34, 1986

    A. Garcia,Nonequilibrium fluctuations studied by rarefied-gas simulation, Physical Re- view A 34, 1986

    1986

  12. [20]

    Maxwell, J.C.Illustrations of the dynamical theory of gases. Part I. On the motions and collisions of perfectly elastic spheres, The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 19, 1860

  13. [21]

    and Lampis, M.,Kinetic Models for Gas-Surface Interaction, Transport Theory and Statistical Physics 1, 1971

    Cercignani, C. and Lampis, M.,Kinetic Models for Gas-Surface Interaction, Transport Theory and Statistical Physics 1, 1971

    1971

  14. [22]

    Padilla, I

    J. Padilla, I. Boyd,Assessment of Gas-Surface Interaction Models in DSMC Analysis of Rarefied Hypersonic Flow, 39th AIAA Thermophysics Conference, 2007

    2007

  15. [23]

    Zhou,Hydrodynamic Instabilities and Turbulence: Rayleigh- Taylor, Richtmyer- Meshkov, and Kelvin-Helmholtz Mixing, Cambridge University Press, 2024

    Y. Zhou,Hydrodynamic Instabilities and Turbulence: Rayleigh- Taylor, Richtmyer- Meshkov, and Kelvin-Helmholtz Mixing, Cambridge University Press, 2024

    2024

  16. [24]

    D. M. Sterbentz, C. F. Jekel, D. A. White, S. Aubry, H. E. Lorenzana, J. L. Belof, Design optimization for Richtmyer-Meshkov instability suppression at shock-compressed material interfaces, Physics of Fluids 34 (8), 2022

    2022

  17. [25]

    Thompson, L

    A. Thompson, L. Swiller, C. Trott, S. Foiles, G. Tucker,Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials, Journal of Computational Physics, 2015

    2015

  18. [26]

    P. Sun, J. Ding, S. Huang, X. Luo, W. Cheng,Microscopic Richtmyer-Meshkov insta- bility under strong shock, Physics of Fluids 32, 2020

    2020

  19. [27]

    Zhakhovskii, S

    V. Zhakhovskii, S. Zybin, S. Abarzhi, K. Nishihara,Atomistic Dynamics of the Richtmyer-Meshkov Instability in Cylindrical and Planar Geometries, Proceedings of the Conference of the American Physical Society Topical Group on Shock Compression of Condensed Matter, 2005

    2005

  20. [28]

    Longshaw, R

    S. Longshaw, R. Pillai, L. Gibelli, D. Emerson, D. Lockerby,Coupling Molecular Dy- namics and Direct Simulation Monte Carlo using a general and high-performance code coupling library, Computers and Fluids, 2020

    2020

  21. [29]

    K. Gu, C. Watkins, J. Koplik,Atomistic hybrid DSMC/NEMD method for nonequilib- rium multiscale simulations, Journal of Computational Physics, 2010. 28

    2010

  22. [30]

    Nedea, A

    S. Nedea, A. Frijns, A. Van Steenhoven, A. Markvoort, P. Hilbers,Hybrid method coupling molecular dynamics and monte carlo simulations to study the properties of gases in microchannels and nanochannels, Physical Review E - Statistical, Nonlinear, and Soft Matter Physics, 72, 2005

    2005

  23. [31]

    Watvisave, B

    D. Watvisave, B. Puranik, U. Bhandarkar,A hydrid MD-DSMC coupling method to investigate flow characteristics of micro-devices, Journal of Computational Physics, 2015

    2015

  24. [32]

    Yamanishi, Y

    N. Yamanishi, Y. Matsumoto, K. Shobatake,Multistage gas-surface interaction model for the direct simulation Monte Carlo method, Physics of Fluids, 11, 1999

    1999

  25. [33]

    Liang, W

    T. Liang, W. Ye,An Efficient Hydrid DSMC/MD Algorithm for Accurate Modeling of Micro Gas Flows, Communications of Computational Physics, 15, 2014

    2014

  26. [34]

    Zeifman, B

    M. Zeifman, B. Garrison, L. Zhigilei,Direct Simulation Monte Carlo Calculation: Strategies for Using Complex Initial Conditions, MRS Online Proceedings Library 731, 2002

    2002

  27. [35]

    Weinan, B

    E. Weinan, B. Engquist, X. Li, W. Ren, E. Vanden-Eijnden.Heterogeneous multiscale methods: a review, Communications in computational physics, Vol. 2 No. 3, 2007

    2007

  28. [36]

    C. Gear, J. Hyman, G. Kevrekidid, I. Kevrekidis, O. Runborg, C. Theodoropoulos. Equation-free, coarse-grained multiscale computation: Enabling microscopic simulators to perform system-level analysis, Comm. Math. Sci., Vol. 1 No. 4, 2003

    2003

  29. [37]

    P. Lax, R. Phillips,The asymptotic distribution of lattice points in euclidean and non- euclidean spaces, Journal of Functional Analysis 46, 1982

    1982

  30. [38]

    Beazley, P

    D. Beazley, P. Lomdahl, N. Grønbech-Jensen, P. Tamayo,A high performance com- munications and memory cashing scheme for molecular dynamics on the CM-5, The 8th International Parallel Processing Symposium, Cancún, Mexico, IEEE Computer Society Press, p.800, 1994

    1994

  31. [39]

    Thompson, H

    A. Thompson, H. Aktulga, R. Berger, D. Bolintineanu, W. Brown, P. Crozier, P. in ’t Veld, A. Kohlmeyer, S. Moore, T. Nguyen, R. Shan, M. Stevens, J. Tranchida, C. Trott, S. J. Plimpton,LAMMPS - a flexible simulation tool for particle-based materials modeling at the atomic, mes...

    2022

  32. [40]

    Hadjiconstantinou, A

    N. Hadjiconstantinou, A. Garica, M. Bazant, G. He,Statistical error in particle simu- lations of hydrodynamic phenomena, Journal of Computational Physics, 2003

    2003

  33. [41]

    Widodo, D

    A. Widodo, D. Buttsworth,Stagnation temperature in a cold hypersonic flow produced by a light free piston compression facility, Experiments in Fluids 54, 2013. 29

    2013

  34. [42]

    Buttsworth,Ludwieg Tunnel Facility with Free Piston Compression Heating for Su- personic and Hypersonic Testing, Proceedings of the 9th Australian Space Science Con- ference, 2010

    D. Buttsworth,Ludwieg Tunnel Facility with Free Piston Compression Heating for Su- personic and Hypersonic Testing, Proceedings of the 9th Australian Space Science Con- ference, 2010

    2010

  35. [43]

    C. Shu, X. Mao, Y. Chew,Particle number per cell and scaling factor effect on accuracy of DSMC simulation of micro flows, International Journal of Numerical Methods for Heat & Fluid Flow, Vol. 15 No. 8, 2005

    2005

  36. [44]

    Atallah,A relationship between emissivity and thermal conductivity of metals, British Journal of Applied Physics 17, 1966

    S. Atallah,A relationship between emissivity and thermal conductivity of metals, British Journal of Applied Physics 17, 1966

    1966

  37. [45]

    Stoller, A

    R. Stoller, A. Tamm, L. Béland, G. Samolyuk, G. Stocks, A. Caro, L. Slipchenko, Y. Osetsky, A. Aabloo, M. Klintenberg, Y. Wang,Impact of Short-Range Forces on Defect Production from High-Energy Collisions, Journal of Chemical Theory and Computation 12, 2016

    2016

  38. [46]

    R. Chen, W. Yuen,Short-Time Oxidation Behavior of Low-Carbon, Low-Silicon Steel in Air at 850–1180°C: II. Linear to Parabolic Transition Determined Using Existing Gas-Phase Transport and Solid-Phase Diffusion Theories, Oxidation of Metals 73, 2010

    2010

  39. [47]

    Lorentz,Ueber die Anwendung des Satzes vom Virial in der kinetischen Theorie der Gase, Annalen der Physik 248, 1881

    H. Lorentz,Ueber die Anwendung des Satzes vom Virial in der kinetischen Theorie der Gase, Annalen der Physik 248, 1881

  40. [48]

    Berthelot,Sur le mélange des gaz, Comptes rendus hebdomadaires des séances de l’Académie des Sciences 126, 1898

    D. Berthelot,Sur le mélange des gaz, Comptes rendus hebdomadaires des séances de l’Académie des Sciences 126, 1898

  41. [49]

    Shinoda, Wataru and Shiga, Motoyuki and Mikami, Masuhiro,Rapid estimation of elastic constants by molecular dynamics simulation under constant stress, Phys. Rev. B, Vol. 69 No. 13, 2004. 30

    2004

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.