Pith. sign in

REVIEW 3 major objections 4 minor 49 references

DSMC: A Statistical Mechanics Perspective

T0 review · 3 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read The paper argues that DSMC's stochastic fluctuations are physically correct thermodynamic data, not numerical noise, and that statistical mechanics supplies the tools to exploit and test them.

desk verdict A useful DSMC perspective whose central statistical-mechanics pitch is solid, but Section 5's logical chain about detailed balance, ergodicity, and the Second Law is wrong and needs a fix. read the letter →

arxiv 2501.07785 v1 pith:YBESVPKU submitted 2025-01-14 cond-mat.stat-mech

Alejandro L. Garcia This is my paper · ORCID
classification cond-mat.stat-mech MSC 82B8082C4082C22
keywords DirectSimulationMonteCarlostatisticalmechanicshydrodynamicfluctuationsfluctuation-dissipationtheoremGreen-KuborelationsdetailedbalancetransportcoefficientsBrownianmotion
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 makes the case that Direct Simulation Monte Carlo (DSMC) is a numerical method for statistical mechanics, not merely a solver for the Boltzmann equation. Its central move is to treat the stochastic scatter in DSMC as physical thermodynamic fluctuation, rescaled by the number of real molecules each simulator particle represents, and therefore as a source of hydrodynamic information rather than an annoyance. From this viewpoint, equilibrium simulations can measure viscosity, sound speed, and other transport coefficients through the fluctuation-dissipation relation, and non-equilibrium fluctuation phenomena such as long-ranged correlations and giant fluctuations become observable. The same statistical-mechanics tools are used to derive the algorithm's cell-size and time-step errors, and detailed balance is proposed as a guardrail for DSMC implementations, including those augmented by machine learning. A reader should care because the paper reframes what counts as signal in a DSMC run.

What carries the argument

The argument runs on four statistical-mechanics devices. The collision virial $\Theta=\langle \Delta v_\alpha \cdot r_{\alpha\beta}\rangle$, averaged over collision pairs, vanishes in DSMC because collision partners are chosen with no position bias, which is why DSMC gives the ideal gas law even at finite cell size. Green-Kubo and Einstein-Helfand relations convert equilibrium fluctuations of momentum current into transport coefficients; applying the hard-sphere decomposition of the stress autocorrelation to DSMC yields the finite-cell viscosity correction $\eta = \eta_K + \eta_C$ with $\eta_C \sim \ell^2/\lambda^2$, and a similar $\tau^2$ error for finite time steps. The fluctuation-dissipation theorem, with the coarse-graining factor $F_N$ as the variance amplification scale, is what licenses reading DSMC noise as physical fluctuations. Detailed balance, the condition $p_j W_{ji} = p_i W_{ij}$, is the guardrail that connects microscopic reversibility to the Second Law and provides an equilibrium statistical test for implementations.

What would settle it

Run equilibrium DSMC at fixed physical parameters while shrinking cell size and time step; if the $F_N$-rescaled variance of density and velocity fluctuations, or the location and width of the Brillouin peak in the dynamic structure factor, fails to converge to the fluctuating-hydrodynamics or light-scattering prediction, then the central claim is refuted. The same test applied to a machine-learned collision model would show whether its fluctuations satisfy fluctuation-dissipation.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that DSMC occupies a middle ground between molecular dynamics and kinetic theory: it samples the same thermal fluctuations as MD, but with a controllable coarse-graining factor $F_N$. After rescaling by $F_N$, the hydrodynamic fluctuations in DSMC are physically correct and obey the fluctuation-dissipation theorem, so the noise that early treatments dismissed is instead a faithful record of Brownian motion, Brillouin scattering, non-equilibrium long-range correlations, and the influence of thermal noise on the dissipation range of turbulence. The paper further establishes, via the collision virial and Green-Kubo analysis, that the ideal-gas equation of state in DSMC follows from the vanishing of the collision virial, and that finite cell size and time step contaminate transport coefficients by $\ell^2$ and $\tau^2$ errors respectively, with a closed-form viscosity correction. Finally, it argues that detailed balance is the necessary condition for a DSMC implementation to respect the Second Law, making equilibrium fluctuation tests a sensitive canary for algorithmic errors and a necessary checkpoint for machine-learned collision models.

Load-bearing premise

The argument assumes that the discrete cell-and-time-step collision dynamics of DSMC reproduce the fluctuation-dissipation theorem faithfully enough that, after rescaling by $F_N$, the amplified fluctuations are the same physics as real molecular fluctuations and not an artifact of the discretization.

Editorial extensions

If this is right

  • Equilibrium DSMC simulations can be used as a measurement device: fitting the Brillouin peak in the density-fluctuation spectrum gives sound speed and viscosity the way light-scattering experiments do.
  • The closed-form cell-size correction turns a practical rule of thumb into a quantitative accuracy criterion: transport coefficients inherit an $\ell^2/\lambda^2$ error, so collision cells well below one mean free path are needed for accurate viscosity.
  • Since DSMC fluctuations are physical after rescaling, comparisons of simulated dynamic structure factors with theory and laboratory scattering data become a legitimate validation test for collision and internal-energy relaxation models.
  • Detailed balance is a necessary condition for DSMC dynamics to satisfy the Second Law, so equilibrium checks such as the Poisson distribution of cell particle numbers serve as a sensitive canary for bugs and bias in implementations.
  • Machine-learned DSMC components must also pass detailed balance and equilibrium fluctuation tests, or the trained models risk learning unphysical dynamics and corrupting the data they are meant to accelerate.

Reading between the lines

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

  • Editorial inference: if the fluctuation-dissipation relation survives coarse-graining in the way the paper claims, then DSMC could double as a tunable laboratory for fluctuating hydrodynamics, where $F_N$ acts as a controllable noise-amplification dial; a direct test would be to check whether the rescaled fluctuation spectrum remains invariant as cell size approaches the mean free path.
  • Editorial inference: the virial argument implies a design rule for dense-gas particle schemes: any algorithm that aims beyond the ideal gas law must break the position-symmetry of DSMC collision-pair selection, either through post-collision displacements or position-dependent collision probabilities.
  • Editorial inference: the detailed-balance canary, as stated, is only a necessary test, so a stronger falsifiable extension would be to compare non-equilibrium fluctuation spectra from machine-learned DSMC against known analytic long-range correlation predictions; failure there would indicate the learned collision rule violates fluctuation-dissipation even if equilibrium Poisson statistics pass.
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

3 major / 4 minor

Summary. This perspective paper argues that DSMC should be viewed not only as a numerical solver for the Boltzmann equation but as a tool for statistical mechanics. It revisits three topics: the equation of state of DSMC (the collision virial vanishes by symmetry, yielding the ideal gas law), transport-coefficient errors due to finite cell size and time step (derived via Green-Kubo/Einstein-Helfand arguments), and the physical nature of DSMC fluctuations (after accounting for the coarse-graining factor F_N). The paper then proposes that equilibrium fluctuation relations can serve as a 'canary' test for algorithmic bias, particularly for machine-learning-based DSMC implementations, and attempts to ground this proposal in a logical chain involving detailed balance, microscopic reversibility, ergodicity, and the Second Law.

Significance. If the central claims hold, the paper provides a valuable perspective that reframes DSMC's stochastic noise as physical signal, enabling equilibrium determinations of transport coefficients and diagnostic checks of simulation correctness. The paper is concise and brings together several established results, including the virial cancellation, the cell-size correction to viscosity in Eq. (10), the statistical error estimate in Eq. (11), and the use of the dynamic structure factor. The proposed canary test is a concrete, falsifiable diagnostic idea. However, the Section 5 logical chain is seriously flawed, and because that chain is the stated support for the ML-guardrail recommendation, the manuscript needs substantial revision before its central application can be accepted.

major comments (3)
  1. [Section 5, Eq. (12)] The logical chain following Eq. (12) is not correct as stated. 'Ergodicity (p_i = p_j)' conflates a uniform stationary distribution with ergodicity, which is a property of time averages or of chain irreducibility, not equality of probabilities. Detailed balance can hold with a non-uniform stationary measure (e.g., the Maxwell-Boltzmann distribution in a gas), and in that case W_ij ≠ W_ji. Furthermore, the Shannon entropy S = −k Σ p_i ln p_i is not generally nondecreasing for a Markov chain whose stationary distribution is not uniform; the nondecreasing functional is the relative entropy with respect to the stationary measure. Because the canary test and the ML-guardrail recommendation in this section rest on these relations, the argument must be rewritten using the correct detailed-balance and entropy-production conditions for DSMC's actual stationary measure.
  2. [Section 4] The paper's strongest claim — that hydrodynamic fluctuations in DSMC are physically correct — is stated in a single sentence with a citation to [31]: 'After accounting for this amplification factor, we find that hydrodynamic fluctuations in DSMC are physically correct [31].' Since the subsequent proposals (measuring transport coefficients from equilibrium fluctuations and using fluctuations as a canary) depend on the fluctuation spectrum matching fluctuating hydrodynamics, the manuscript should state the conditions under which this equivalence holds (e.g., limits on cell size, time step, and F_N) and at least summarize the evidence from [31] and related work. Without this, the scope of the central claim is not assessable from the manuscript itself.
  3. [Section 5, detailed balance paragraph] The sentence 'Detailed balance is a necessary condition for microscopic reversibility, which is a sufficient condition for the dynamics to obey the Second Law' is also problematic. Under the paper's definition of microscopic reversibility (W_ij = W_ji), detailed balance follows only if the stationary distribution is uniform; in DSMC the relevant stationary distribution is not uniform over the full state space. The canary test is described as 'necessarily but not sufficient,' but the necessity claim is not established by the preceding argument, since the required logical relations are misstated. Please replace this paragraph with a correct statement of the relations among detailed balance, microscopic reversibility, and entropy production for the DSMC master equation.
minor comments (4)
  1. [Section 3, Eq. (6)] The second double sum in Eq. (6) is written as Σ_i Σ_{i=j}; this appears to be a typo for a sum over distinct pairs (i ≠ j), since the self-term F_ii is zero.
  2. [Section 3, after Eq. (10)] The text contains the typo 'Chapmann-Enskog'; it should read 'Chapman-Enskog'.
  3. [Section 6] The code name is spelled 'LAMPPS' in the text but 'LAMMPS' in reference [43]; please make the spelling consistent.
  4. [General] Because Sections 2–4 summarize previously published results in a very compressed way, a brief scope statement indicating which parts are new and which are reviews of prior work would help readers who are not familiar with the cited DSMC literature.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation found; self-citations are supported by external evidence, and Section 5's logical issues are correctness concerns, not circularity.

full rationale

The paper is a perspective that re-derives and cites prior published results, but no step makes a prediction equivalent to its input by construction. Section 3 derives the cell-size viscosity correction from standard Einstein-Helfand and Green-Kubo relations (Eqs. 4-7) plus Wainwright's hard-sphere analysis; the cited self-result [7] is re-derived in the text, not assumed. The central claim that DSMC hydrodynamic fluctuations are physically correct relies on self-citation [31], but the paper also anchors it to external light-scattering experiments, Bird's independent statements, and DSMC reproductions of long-ranged correlations [26,30,37], so this is real evidence rather than a self-referential loop. Section 5's identification of ergodicity with a uniform stationary distribution (pi = pj) and the claimed 'iff' relations for the Second Law are mathematically questionable, but they are an asserted logical chain, not an input-output equivalence; moreover, the canary test is explicitly qualified as 'necessarily but not sufficient.' No fitted parameter is relabeled as a prediction, and no equation reduces to its own input by construction. I therefore find no significant circularity.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper's central argument rests on DSMC-specific modeling assumptions (position-independent collision selection, physical fluctuation scaling) and on standard statistical mechanics results. No free parameters are fitted and no new entities are postulated.

assumptions (3)
  • domain assumption DSMC collision partners are selected uniformly within a cell, independent of particle positions, so the average collision virial vanishes.
    Stated in Section 2; this defines the DSMC collision rule and underlies the ideal gas law result.
  • domain assumption Hydrodynamic fluctuations in DSMC are physically correct once the amplification factor FN is accounted for.
    Stated in Section 4, citing Garcia and Penland [31]; it is the basis for treating DSMC noise as physical.
  • standard math The Second Law holds for DSMC dynamics iff the equilibrium distribution is uniform (ergodicity), equivalent to microscopic reversibility, as per Thomsen [44].
    Section 5 uses these logical relations to argue that detailed balance is a guardrail against Maxwell demons.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DSMC: A Statistical Mechanics Perspective." pith.science (2026). https://pith.science/paper/YBESVPKU

@misc{pith2026250107785,
  author       = {Pith},
  title        = {Pith review of: DSMC: A Statistical Mechanics Perspective},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YBESVPKU}},
  note         = {Machine review of arXiv:2501.07785}
}
read the original abstract

This paper presents a perspective in which Direct Simulation Monte Carlo (DSMC) is viewed not in its traditional role as an algorithm for solving the Boltzmann equation but as a numerical method for statistical mechanics. First, analytical techniques such as the collision virial and Green-Kubo relations, commonly used in molecular dynamics, are used to study the numerical properties of the DSMC algorithm. The stochastic aspect of DSMC, which is often viewed as unwanted numerical noise, is shown to be a useful feature for problems in statistical physics, such as Brownian motion and thermodynamic fluctuations. Finally, it is argued that fundamental results from statistical mechanics can provide guardrails when applying machine learning to DSMC.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

49 extracted references · 49 canonical work pages

  1. [44]

    J. S. Thomsen , Logical relations among the principles of statistical me- chanics and thermodynamics , Physical Review, 91 (1953), p. 1263

    work page 1953

  2. [31]

    Garcia and C

    A. Garcia and C. Penland , Fluctuating hydrodynamics and principal os- cillation pattern analysis , Journal of statistical physics, 64 (1991), pp. 1121– 1132

    work page 1991

  3. [1]

    Alder, D

    B. Alder, D. Gass, and T. W ainwright , Studies in molecular dynam- ics. viii. the transport coefficients for a hard-sphere fluid , The Journal of Chemical Physics, 53 (1970), pp. 3813–3826

    work page 1970

  4. [2]

    B. J. Alder and T. E. W ainwright , Studies in molecular dynamics. i. general method, The Journal of Chemical Physics, 31 (1959), pp. 459–466

    work page 1959

  5. [3]

    B. J. Alder and T. E. W ainwright , Studies in molecular dynamics. ii. behavior of a small number of elastic spheres , The Journal of Chemical Physics, 33 (1960), pp. 1439–1451

    work page 1960

  6. [4]

    F. J. Alexander and A. L. Garcia , The direct simulation monte carlo method, Computers in Physics, 11 (1997), pp. 588–593

    work page 1997

  7. [5]

    F. J. Alexander, A. L. Garcia, and B. J. Alder , A consistent boltz- mann algorithm , Physical Review Letters, 74 (1995), p. 5212

    work page 1995

  8. [6]

    , The consistent boltzmann algorithm for the van der waals equ ation of state , Physica A: Statistical Mechanics and its Applications, 240 (1997), pp. 196–201

    work page 1997

Show all 49 references
  1. [7]

    1540–1542

    , Cell size dependence of transport coefficients in stochastic particle algorithms, Physics of Fluids, 10 (1998), pp. 1540–1542

    1998

  2. [8]

    N. D. Ball, J. F. MacArt, and J. Sirignano , Online optimisation of machine learning collision models to accelerate direct mol ecular simulation of rarefied gas flows , arXiv preprint arXiv:2411.13423, (2024)

    2024 arXiv

  3. [9]

    Bandak, N

    D. Bandak, N. Goldenfeld, A. A. Mailybaev, and G. Eyink , Dissipation-range fluid turbulence and thermal noise , Physical Review E, 105 (2022), p. 065113

    2022

  4. [10]

    Baras, M

    F. Baras, M. M. Mansour, A. Garcia, and M. Mareschal , Particle simulation of complex flows in dilute systems , Journal of Computational Physics, 119 (1995), pp. 94–104

    1995

  5. [11]

    J. B. Bell, A. Nonaka, A. L. Garcia, and G. Eyink , Thermal fluctu- ations in the dissipation range of homogeneous isotropic tu rbulence, Journal of Fluid Mechanics, 939 (2022), p. A12

    2022

  6. [12]

    Bird , Approach to translational equilibrium in a rigid sphere gas , Phys

    G. Bird , Approach to translational equilibrium in a rigid sphere gas , Phys. fluids, 6 (1963), pp. 1518–1519

    1963

  7. [13]

    , Molecular Gas Dynamics , Oxford engineering science series, Claren- don Press, 1976

    1976

  8. [14]

    , Molecular Gas Dynamics and the Direct Simulation of Gas Flow s, no. v. 1 in Molecular Gas Dynamics and the Direct Simulation of Gas Flows , Clarendon Press, 1994

    1994

  9. [15]

    , The DSMC Method , CreateSpace Independent Publishing Platform, 2013

    2013

  10. [16]

    , The rate of energy transfer from air as an initially stationa ry particle acquires brownian motion , Nano energy, 11 (2015), pp. 463–470

    2015

  11. [17]

    DSMC & Statistical Mechanics 9

    , Brownian relaxation of an inelastic sphere in air , Physics of Fluids, 28 (2016). DSMC & Statistical Mechanics 9

    2016

  12. [18]

    , The possibility of harvesting useful energy from the therma l motion of air molecules , NanoWorld J, 3 (2017), pp. 18–22

    2017

  13. [19]

    G. Bird, M. Gallis, J. Torczynski, and D. Rader , Accuracy and efficiency of the sophisticated direct simulation monte carl o algorithm for simulating noncontinuum gas flows , Physics of Fluids, 21 (2009)

    2009

  14. [20]

    Boyd and T

    I. Boyd and T. Schwartzentruber , Nonequilibrium Gas Dynamics and Molecular Simulation , Cambridge Aerospace Series, Cambridge University Press, 2017

    2017

  15. [21]

    Bruno, A

    D. Bruno, A. Frezzotti, and G. P. Ghiroldi , Rayleigh–brillouin scat- tering in molecular oxygen by ct-dsmc simulations , European Journal of Mechanics-B/Fluids, 64 (2017), pp. 8–16

    2017

  16. [22]

    Bruno and V

    D. Bruno and V. Giovangigli , Internal energy relaxation processes and bulk viscosities in fluids , Fluids, 7 (2022), p. 356

    2022

  17. [23]

    Cercignani and M

    C. Cercignani and M. Lampis , Kinetic models for gas-surface interac- tions, Transport theory and statistical physics, 1 (1971), pp. 101–1 14

    1971

  18. [24]

    De Groot and P

    S. De Groot and P. Mazur , Non-Equilibrium Thermodynamics , Dover Books on Physics, Dover Publications, 2013

    2013

  19. [25]

    Donev, B

    A. Donev, B. J. Alder, and A. L. Garcia , Stochastic hard-sphere dynamics for hydrodynamics of nonideal fluids , Physical review letters, 101 (2008), p. 075902

    2008

  20. [26]

    Donev, J

    A. Donev, J. B. Bell, A. de La Fuente, and A. L. Garcia , Diffu- sive transport by thermal velocity fluctuations , Physical review letters, 106 (2011), p. 204501

    2011

  21. [27]

    Frezzotti , A particle scheme for the numerical solution of the enskog equation, Physics of Fluids, 9 (1997), pp

    A. Frezzotti , A particle scheme for the numerical solution of the enskog equation, Physics of Fluids, 9 (1997), pp. 1329–1335

    1997

  22. [28]

    Gallis, N

    M. Gallis, N. Bitter, T. Koehler, J. Torczynski, S. Plimpton , and G. Papadakis, Molecular-level simulations of turbulence and its decay , Physical review letters, 118 (2017), p. 064501

    2017

  23. [29]

    Garcia , Estimating hydrodynamic quantities in the presence of micr o- scopic fluctuations , Communications in Applied Mathematics and Compu- tational Science, 1 (2007), pp

    A. Garcia , Estimating hydrodynamic quantities in the presence of micr o- scopic fluctuations , Communications in Applied Mathematics and Compu- tational Science, 1 (2007), pp. 53–78

    2007

  24. [30]

    Garcia, M

    A. Garcia, M. M. Mansour, G. Lie, and E. Clementi , Hydrodynamic fluctuations in a dilute gas under shear , Physical Review A, 36 (1987), p. 4348

    1987

  25. [32]

    Garcia and W

    A. Garcia and W. W agner , Time step truncation error in direct simu- lation monte carlo , Physics of Fluids, 12 (2000), pp. 2621–2633

    2000

  26. [33]

    N. G. Hadjiconstantinou , Analysis of discretization in the direct simu- lation monte carlo , Physics of Fluids, 12 (2000), pp. 2634–2638

    2000

  27. [34]

    N. G. Hadjiconstantinou, A. L. Garcia, and B. J. Alder , The sur- face properties of a van der waals fluid , Physica A: Statistical Mechanics and its Applications, 281 (2000), pp. 337–347

    2000

  28. [35]

    N. G. Hadjiconstantinou, A. L. Garcia, M. Z. Bazant, and G. He , Statistical error in particle simulations of hydrodynamic phenomena, Jour- nal of computational physics, 187 (2003), pp. 274–297. 10 Alejandro L. Garcia

    2003

  29. [36]

    L. D. Landau and E. M. Lifshitz , Statistical Physics: Volume 5 , vol. 5, Elsevier, 2013

    2013

  30. [37]

    M. M. Mansour, A. L. Garcia, G. C. Lie, and E. Clementi , Fluc- tuating hydrodynamics in a dilute gas , Physical review letters, 58 (1987), p. 874

    1987

  31. [38]

    R. M. McMullen, M. C. Krygier, J. R. Torczynski, and M. A. Gallis, Navier-stokes equations do not describe the smallest scale s of tur- bulence in gases , Physical Review Letters, 128 (2022), p. 114501

    2022

  32. [39]

    Metropolis, A

    N. Metropolis, A. W. Rosenbluth, M. N. Rosenbluth, A. H. Teller, and E. Teller , Equation of state calculations by fast computing machines, The journal of chemical physics, 21 (1953), pp. 1087–1092

    1953

  33. [40]

    J. M. Montanero and A. Santos , Simulation of the enskog equation a la bird, Physics of Fluids, 9 (1997), pp. 2057–2060

    1997

  34. [41]

    ¨Ottinger, Beyond Equilibrium Thermodynamics , Wiley, 2005

    H. ¨Ottinger, Beyond Equilibrium Thermodynamics , Wiley, 2005

    2005

  35. [42]

    S. J. Plimpton, S. Moore, A. Borner, A. Stagg, T. Koehler, J. Torczynski, and M. Gallis , Direct simulation monte carlo on petaflop supercomputers and beyond, Physics of Fluids, 31 (2019)

    2019

  36. [43]

    A. P. Thompson, H. M. Aktulga, R. Berger, D. S. Bolintineanu, W. M. Brown, P. S. Crozier, P. J. In’t Veld, A. Kohlmeyer, S. G. Moore, T. D. Nguyen, et al. , Lammps-a flexible simulation tool for particle-based materials modeling at the atomic, meso, and continuum scales, Comput...

    2022

  37. [45]

    M. W. Tysanner and A. L. Garcia , Non-equilibrium behaviour of equi- librium reservoirs in molecular simulations , International journal for nu- merical methods in fluids, 48 (2005), pp. 1337–1349

    2005

  38. [46]

    V ailati and M

    A. V ailati and M. Giglio , Giant fluctuations in a free diffusion process , Nature, 390 (1997), pp. 262–265

    1997

  39. [47]

    V alentini, C

    P. V alentini, C. Zhang, and T. E. Schwartzentruber , Molecular dynamics simulation of rotational relaxation in nitrogen: Implications for rotational collision number models , Physics of Fluids, 24 (2012)

    2012

  40. [48]

    W agner, A convergence proof for bird’s direct simulation monte carl o method for the boltzmann equation , Journal of Statistical Physics, 66 (1992), pp

    W. W agner, A convergence proof for bird’s direct simulation monte carl o method for the boltzmann equation , Journal of Statistical Physics, 66 (1992), pp. 1011–1044

    1992

  41. [49]

    W ainwright, Calculation of hard-sphere viscosity by means of correla- tion functions , The Journal of Chemical Physics, 40 (1964), pp

    T. W ainwright, Calculation of hard-sphere viscosity by means of correla- tion functions , The Journal of Chemical Physics, 40 (1964), pp. 2932–2937

    1964

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.