Pith. sign in

REVIEW 4 major objections 5 minor 43 references

Reaction Dynamics for the [NNO] System from State-Resolved and Coarse-Grained Models

T0 review · 4 major / 5 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 For the NO + N ↔ N2 + O network at 10,000 K, Arrhenius rates under-predict the equilibrium N2 yield by about 20 percent; state-to-state rates reach full turnover.

desk verdict The retrained STS dictionaries and the dissociation-included comparison are worth a look, but the paper's headline claim that Arrhenius rates under-predict N2 yield is likely an artifact of missing reverse flux in the STS exchange-only master equations. read the letter →

arxiv 2506.06146 v1 pith:RZLX3C7Q submitted 2025-06-06 physics.chem-ph

Juan Carlos San Vicente Veliz , Sung Min Jo , Jingchun Wang , Raymond J. Bemish , Markus Meuwly This is my paper · ORCID
classification physics.chem-ph
keywords state-to-statedynamicsNO+N↔N2Omasterequationpotentialenergysurfacesquasi-classicaltrajectoriesnon-equilibriumchemistryhypersonicre-entry
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

This paper asks whether the chemical evolution of the NO + N ↔ N2 + O network at 10,000 K depends on how reaction rates are represented. It compares coarse-grained Arrhenius rate expressions with state-to-state (STS) rate dictionaries derived from quasi-classical trajectories, using two independently built high-level potential energy surfaces: one represented by reproducing kernels, the other by permutationally invariant polynomials. The central claim is that Arrhenius rates leave NO-to-N2 conversion incomplete—the asymptotic N2 mole fraction is about $0.4$—while STS-based master equations reach complete exchange turnover at about $0.5$, and that STS concentration profiles are consistent over 14 orders of magnitude in time whether the rates come from the reproducing-kernel surface or the polynomial surface. The paper argues this gap comes from non-equilibrium rovibrational energy flow that only a state-resolved treatment captures, and that the choice of surface or of reverse-rate construction shifts ignition times but not the qualitative outcome. If right, equilibrium thermal-rate fits are not enough to predict product yields in strongly non-equilibrium high-temperature gases.

What carries the argument

The machinery is a pair of state-to-state rate dictionaries that resolve every initial and final rovibrational state of NO and N2 together with collision energy, fed into a master equation over species and internal states. One dictionary is a neural-network interpolation trained on 160,000 quasi-classical trajectories per initial condition on the reproducing-kernel surface; the other is an explicitly computed quasi-classical dictionary on the permutationally invariant polynomial surface, generated with 50,000 trajectories per initial condition. The mechanism that drives the result is the contrast between integrating these roughly $10^8$ state-resolved rates and integrating a few Arrhenius expressions coupled to a Landau-Teller vibrational relaxation model: only the state-resolved treatment lets heterogeneous exchange and dissociation act on the same non-equilibrium rovibrational populations. A control model that Boltzmann-weights a thermal rate across product states reproduces the Arrhenius result, which the paper takes as evidence that the missing ingredient is state dynamics, not mere rate dimensionality.

What would settle it

Recompute the explicit state-to-state dictionary on the polynomial surface with the same 160,000-trajectory budget used for the neural-network dictionary and re-integrate the master equation; if the asymptotic N2 mole fraction shifts from $0.5$ toward $0.4$, the apparent independence of the two surfaces was an artifact of an under-converged reference rate set.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the asymptotic composition of the reacting N/O gas depends on the resolution of the rate coefficients used in the master equation, not on which accurate potential energy surface supplies them. For the exchange network NO + N ↔ N2 + O at 10,000 K, Arrhenius rates fitted to quasi-classical trajectory thermal rates converge to $[\mathrm{N_2}] \approx 0.4$, whereas state-to-state rate dictionaries—one a neural-network interpolation of QCT data, one an explicitly computed QCT dictionary—converge to $[\mathrm{N_2}] \approx 0.5$. The paper attributes this 20 percent difference to non-equilibrium energy flow through individual rovibrational states: the Arrhenius treatment uses a two-temperature Landau-Teller relaxation model whose relaxation times are derived for inelastic and homogeneous exchange rather than the heterogeneous exchange of interest, so it cannot represent the state-resolved pathways. It also shows that using explicit reverse rates instead of microreversibility changes the product concentration by only 0.03 mole fraction units, and that adding full dissociation makes both STS models asymptotically reach the stoichiometric 2:1 $[\mathrm{N}]:[\mathrm{O}]$ ratio.

Load-bearing premise

The comparison assumes that the explicitly computed STS dictionary on the polynomial surface is accurate enough to act as an independent reference, even though the paper states that 50,000 trajectories per initial condition do not suffice to converge all state-to-state cross sections, and that the Landau-Teller relaxation times used in the Arrhenius runs are appropriate for the heterogeneous exchange reaction.

Editorial extensions

If this is right

  • Hypersonic air-chemistry models that use Arrhenius rates with Landau-Teller vibrational relaxation will under-predict N2 production by about 20 percent in the NO + N exchange network at 10,000 K, even if the rates are fitted to the same trajectory data.
  • Species concentrations from state-resolved master equations are stable to the choice of surface representation and to whether reverse rates come from microreversibility or explicit calculation; only the ignition time shifts, by up to about an order of magnitude.
  • Including dissociation channels is necessary for thermodynamic consistency: with them, both STS models asymptotically produce the stoichiometric $[\mathrm{N}]:[\mathrm{O}] = 2:1$ composition, whereas without them the diatomic mole fractions freeze at finite values.
  • The Arrhenius-versus-STS gap is not an artifact of state resolution itself, because a Boltzmann-weighted intermediate model with the same number of states as the STS model still lands at the Arrhenius limit.

Reading between the lines

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

  • If this gap survives improved convergence of the explicit dictionary, it would imply that thermal-rate fits are fundamentally insufficient for non-equilibrium high-enthalpy air chemistry, and that coarse-grained CFD models need reduced-order state-dynamics corrections rather than better-fitted Arrhenius parameters.
  • A direct test of the consistency claim would be to regenerate the explicit dictionary with the full 160,000-trajectory budget on the same surface; agreement would make the PES-insensitivity result robust, while disagreement would trace it to trajectory statistics.
  • The study is restricted to the 3A' electronic state, so an extension to the 3A'' surface and to the NO + O channels would show whether the state-resolved requirement persists for the full five-species air network.
  • Because the paper finds ignition times are similar but final yields differ, an experimentally accessible signature might be the ratio of N2 to NO downstream of a shock tube: state-resolved chemistry predicts a higher N2 fraction than Arrhenius-based fits.
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

4 major / 5 minor

Summary. The manuscript compares Arrhenius-based and state-to-state (STS) master-equation kinetics for NO(X2Π)+N(4S) ↔ N2(X1Σg+)+O(3P) on the 3A' potential energy surface at 10000 K, using two high-level PESs (RKHS-based PES B and PIP-based PES M) and two STS dictionaries (neural-network-based STS2025 and explicit-QCT-based STS-UI). It reports that ignition times are similar across the models, that Arrhenius rates yield incomplete N2 conversion (~0.4 mole fraction) whereas STS rates yield near-complete conversion (~0.5), that this difference arises from non-equilibrium state dynamics, and that including dissociation restores the 2:1 [N]:[O] asymptote, with profiles consistent over 14 orders of magnitude in time.

Significance. If the central claim were valid, the paper would be a strong demonstration that state-resolved kinetics is necessary to capture product yields in high-temperature exchange reactions, and it would support the robustness of coarse-grained predictions across different PES representations. The work contains several useful technical elements: the microreversibility versus explicit reverse-rate comparison (Fig. 4), the group-reconstructed intermediate model (Eq. 9, Fig. 5A), the inclusion of dissociation channels, and public code and data. However, as detailed below, the STS-versus-Arrhenius gap is internally inconsistent with the thermochemistry of the same PESs, and the claimed phenomenon is not established.

major comments (4)
  1. [The Influence of the Underlying PES; Table 1; Figs. 5B and 6] The 'complete turnover' asymptote violates the equilibrium constant implied by the same PESs. For the exchange-only network with initial [NO]=[N]=0.5, K_eq=[N2][O]/([NO][N])=x^2/(0.5-x)^2. Using the Table 1 Arrhenius parameters at T=10000 K gives K_eq≈13.8 (PES B) and ≈11.2 (PES M), hence x∞≈0.39-0.40, matching Fig. 5A. The STS asymptotes 0.48-0.5 (Figs. 5B and 6) imply K_eq≥576, roughly 40 times larger than the thermochemical value. Since Fig. 6A uses microreversibility with partition-function-based K^E_{i,m}, the STS dictionaries are not enforcing detailed balance in aggregate. The 20% Arrhenius-versus-STS gap is therefore not evidence for non-equilibrium state dynamics; it indicates a missing or imbalanced reverse flux in the STS dictionaries or in the microreversibility implementation.
  2. [State-to-State Information from PES B and PESM; Fig. 4B; Fig. S2] The microreversibility test in Fig. 4B compares two ways of reconstructing reverse rates from the same forward dictionary; it cannot detect a systematically missing set of reverse transitions. The Methods statement that 50000 trajectories per initial condition for STS-UI 'does not suffice to converge all state-to-state cross sections', together with the limited test-set range for the reverse STS2025 model (Fig. S2, cross sections 0-0.09 a0^2), makes missing high-v N2+O→NO+N channels a plausible source of the inflated N2 yield. The authors must show that the reverse dictionaries recover the state-resolved detailed-balance ratios over the entire populated state space and that the long-time STS asymptote is consistent with the partition-function equilibrium constant.
  3. [The Influence of the Underlying PES; Figs. 5B and 6] The claimed PES independence is not established because the two STS dictionaries differ in both the PES and the generation protocol: STS2025 is a neural-network interpolation of 160000-trajectory-per-condition QCT on PES B, while STS-UI is an explicit dictionary from 50000-trajectory QCT on PES M. Agreement between two potentially biased dictionaries is not evidence of insensitivity to the PES. A clean test would apply identical sampling and convergence criteria to both PESs, or at minimum re-run the STS-UI reverse transitions to convergence.
  4. [Discussion and Outlook; Eq. 8] The Arrhenius-versus-STS comparison is conditional on the Landau-Teller relaxation times in Eq. 8 taken from Park (1993), which the authors themselves note are derived for inelastic and homogeneous exchange rather than the heterogeneous NO+N / N2+O exchange. Because the 'incomplete conversion' of the Arrhenius model could partly be an artifact of that τVT choice, a sensitivity test with alternative relaxation times or with STS-derived relaxation is required before the difference is attributed to state-resolved non-equilibrium dynamics.
minor comments (5)
  1. [The NN-based State-to-State Models] The sentence 'the model preforms satisfactorily' contains a typo: 'preforms' should be 'performs'.
  2. [Master Equation Analysis, Eq. 7] The notation k^{E,N2}_{m→i}|_X is not defined; clarify what the subscript X denotes.
  3. [Figure 2 caption] The caption writes NO(X2Π)+N(2S); this should likely be N(4S) to match the text.
  4. [Abstract and Fig. 5A] The abstract states ignition points are 'at ~10^-6 s' for both STS and Arrhenius, but Fig. 5A shows an Arr M ignition time of 0.72×10^-7 s, nearly four times earlier than the Arr B value; the range should be stated more precisely.
  5. [Supporting Information description] The SI description mentions photodissociating trajectories and singlet PESs, which do not match the actual SI figures (S1-S4); update the description to reflect the content.

Circularity Check

1 steps flagged · score 2.0 of 10

Minor tautological consistency check in the group-reconstructed (Eq. 9) model; the central STS-vs-Arrhenius and two-PES comparisons are independent simulation outputs, not circular.

  1. self definitional [Discussion and Outlook, Eq. 9 (group-reconstructed rates) and the paragraph immediately following]
    "In other words, the rates ˜kE,NO i→m are obtained from Boltzmann-weighting the thermal kE,NO T at temperature T, and Eq. 9 provides a route to treat an Arrhenius-based model at the equivalent resolution of the STS model. Using the rates ˜kE,NO i→m in the PLATO simulations it is found (dashed black lines in Figure 5A) that the results agree with the corresponding Arrhenius-based approach."

    Equation 9 constructs the state-resolved rates by multiplying the thermal Arrhenius rate k_T by a Boltzmann weight over N2 states. A master equation built from these rates is, by construction, the same Arrhenius rate law resolved onto rovibrational states, so the PLATO output must coincide with the Arrhenius result. The paper presents this agreement as an empirical observation and as evidence that increasing dimensionality alone does not change concentration profiles, but it is a mathematical identity, not an independent test. The step is not load-bearing for the main claim because the full STS dictionaries (STS2025 and STS-UI) are generated from QCT trajectories and NN fits, not from Eq. 9; however, the Eq. 9 agreement is circular by definition.

full rationale

The central results of the paper are master-equation concentration profiles obtained from two independently generated state-to-state dictionaries (STS2025 from NN-trained QCT on PES B, STS-UI from explicit QCT on PES M) and from fitted Arrhenius rates. These are simulation outputs, not derivations that reduce to their inputs. The microreversibility-versus-explicit-reverse-rate comparison is an internal validation rather than a circular step: for PES B the explicit reverse rates come from a separately trained reverse NN, and for PES M from explicit QCT, so the agreement is informative. The use of Refs 18 and 33 (same research group) provides the PES B / STS2025 input data, but the consistency conclusion is also supported by the independent PES M / STS-UI source, so the self-citation is not load-bearing. The only circular element is the 'intermediate model' of Eq. 9: since those rates are Boltzmann-weighted forms of the thermal Arrhenius rate, their agreement with the Arrhenius simulation is guaranteed by construction; the paper uses this only as a consistency check. The acknowledged limitations—STS-UI using 50000 trajectories per condition 'does not suffice to converge all state-to-state cross sections' and the Park Landau-Teller relaxation times being for inelastic/homogeneous rather than heterogeneous exchange—are correctness and uncertainty concerns, explicitly disclosed by the authors, not circularity. Overall, the paper's derivation chain is self-contained and the circularity is minor and non-central.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

Everything the quantitative claims rest on is either a fitted rate expression or an assumed modeling choice. The four Arrhenius parameter sets are fit in prior work (Refs 18 and 12) to the same QCT data that generates the STS dictionaries, which means the Arrhenius-vs-STS comparison is a within-family reduction comparison for the Basel data. No physical entity is invented; the only new artifact is the STS2025 neural network, a regression model rather than an entity. The weakest unverified inputs are the under-converged STS-UI dictionary and the Park relaxation times.

free parameters (5)
  • Arrhenius forward parameters, NO+N to N2+O (Arr B) = A=2.47e-12 cm3 molecule-1 s-1, n=0.4, Ea=8312 K
    Table 1, from Ref 18, fitted to QCT thermal rates on PES B; drives the Arrhenius master equation and the claimed 20 percent under-conversion.
  • Arrhenius reverse parameters, N2+O to NO+N (Arr B) = A=1.39e-10 cm3 molecule-1 s-1, n=0.1, Ea=47180 K
    Table 1, from Ref 18, fitted to QCT thermal rates on PES B.
  • Arrhenius forward parameters, NO+N to N2+O (Arr M) = A=1.43e-13 cm3 molecule-1 s-1, n=0.8, Ea=6276 K
    Table 1, from Ref 12, fitted to QCT thermal rates on PES M.
  • Arrhenius reverse parameters, N2+O to NO+N (Arr M) = A=3.50e-11 cm3 molecule-1 s-1, n=0.4, Ea=48596 K
    Table 1, from Ref 12, fitted to QCT thermal rates on PES M.
  • Landau-Teller vibrational relaxation times tau_VT = values from Park (1993); not reproduced in paper
    Eq. 8: empirical relaxation times for N2+O and NO+N that set the vibrational relaxation rate in the Arrhenius treatment; directly shapes the Arrhenius-vs-STS comparison.
assumptions (6)
  • domain assumption PES B (RKHS) and PES M (PIP) faithfully represent the true 3A' [NNO] potential energy surface.
    Invoked throughout; both surfaces are MRCI+Q-quality fits from Refs 18 and 19, but no experimental scattering data validate them in this work.
  • domain assumption Quasi-classical trajectories with Gaussian binning yield accurate state-to-state cross sections.
    Methods, 'QCT simulations'. Standard semiclassical approach, but the paper notes 50000 trajectories per condition do not converge all STS cross sections for PES M.
  • domain assumption The STS2025 neural network interpolates unsampled (v, j, collision energy) states accurately.
    Methods, 'The NN-based State-to-State Models'. Validated on a held-out test set (Figure 3B, Figure S2) but not for every state the master equation uses.
  • standard math Detailed balance with partition-function equilibrium constants supplies valid reverse rates.
    Methods, 'Master Equation Analysis', Eq. 7. The paper explicitly tests microreversibility against explicit reverse rates, which supports this axiom empirically.
  • domain assumption An isothermal, isobaric heat-bath reactor at 10000 K with 300 K initial internal temperature represents the system of interest.
    Methods, 'Master Equation Analysis'. All equilibrium and ignition claims are conditional on this reactor model, not on a coupled flow simulation.
  • domain assumption The Landau-Teller form with Park's relaxation times describes vibrational relaxation in the Arrhenius treatment.
    Eq. 8. The authors note tau_VT from Park (1993) is derived for inelastic and homogeneous exchange, not necessarily heterogeneous exchange, so the Arrhenius-vs-STS gap is conditional on this approximation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reaction Dynamics for the [NNO] System from State-Resolved and Coarse-Grained Models." pith.science (2026). https://pith.science/paper/RZLX3C7Q

@misc{pith2026250606146,
  author       = {Pith},
  title        = {Pith review of: Reaction Dynamics for the [NNO] System from State-Resolved and Coarse-Grained Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RZLX3C7Q}},
  note         = {Machine review of arXiv:2506.06146}
}
abstract

The dynamics for the NO($X^2 \Pi$) + N($^4$S) $\leftrightarrow$ N$_{2}(X^{1}\Sigma_{g}^{+}$) + O($^{3}$P) reaction was followed in the $^3$A' electronic state using state-to-state (STS) and Arrhenius-based rates from two different high-level potential energy surfaces represented as a reproducing kernel (RKHS) and permutationally invariant polynomials (PIPs). Despite the different number of bound states supported by the RKHS- and PIP-PESs the ignition points from STS and Arrhenius rates are at $\sim 10^{-6}$ s whether or not reverse rates are from assuming microreversibility or explicitly given. Conversion from NO to N$_2$ is incomplete if Arrhenius-rates are used but complete turnover is observed if STS-information is used. This is due to non-equilibrium energy flow and state dynamics which requires a state-based description. Including full dissociation leads asymptotically to the correct 2:1 [N]:[O] concentration with little differences for the species' dynamics depending on the PES used for the STS-information. In conclusion, concentration profiles from coarse-grained simulations are consistent over 14 orders of magnitude in time using STS-information based on two different high-level PESs.

Figures

Figures reproduced from arXiv: 2506.06146 by the authors.

Figure 1
Figure 1. Both PESs are drawn at identical values of the isocontours with the zero of energy [PITH_FULL_IMAGE:figures/full_fig_p013_1.png] view at source ↗
Figure 1
Figure 1. Representations of V (R, θ) for rNO = 2.30 a0 in Jacobi coordinates for the N2O RKHS-PES (panel A), and the PIP-PES (panel B). The energies (kcal/mol) are with respect to the global minimum of each PES. Energy-wise, TS1 and TS2 are at approximately 61.5 / 64.0 kcal/mol and 47.7 / 47.7 kcal/mol relative to MIN in panels A and B, respectively. For both PESs isocontours are drawn at the same energies. The NN-based Stat… view at source ↗
Figure 2
Figure 2. Energy level diagram. From right to left: NO(X [PITH_FULL_IMAGE:figures/full_fig_p015_2.png] view at source ↗
Figures from the paper (8 more)
Figure 3
Figure 3. Figure 3: Panel A: The NO(v, j) states from which QCT simulations were run for training STS2019 (black circles, vNO ∈ [0, 34] and jNO ∈ [0, 210]) and additions for retraining STS2025 for the forward (downhill) reaction NO(X2Π) + N(4S) → N2 (X 1Σ + g ) + O(3P). Black: NO(v, j) gr…
Figure 4
Figure 4. Figure 4: Concentration of N2 (red) and NO (green) species as a function time for generating N2 (NO + N→N2 + O, forward). All PLATO simulations use the 3A′ PESB and the initial populations were [NO](t = 0) = 0.5 and [N2](t = 0) = 0. Panel A: ArrB parameters fitted to QCT simulat…
Figure 5
Figure 5. Figure 5: Effect of using reference data from two different PESs and MR-reverse rates The initial populations in the PLATO simulations are [NO](t = 0) = 0.5 and [N2](t = 0) = 0. Population of N2 (red) and NO (green) as a function of time. Panel A assuming MR for the reverse reac…
Figure 6
Figure 6. Figure 6: Effect of using reference data from two different PESs using STS rates Temporal evolution of state-to-state derived mole fraction concentration at T = 10000 K. Panel A: Reverse rates from assuming microreversibility using [k STS2025 f , kSTS2025,MR r ] (solid) and [k S…
Figure 7
Figure 7. Figure 7: Effect of including the dissociation channel: Time evolution of state-to-state-derived [PITH_FULL_IMAGE:figures/full_fig_p023_7.png]
Figure 8
Figure 8. Figure 8: Population analysis at T = 10000 K with STS-information using STS2025 (blue) and STS-UI (red). Panel A: NO population at 30% mole fraction. Panel B: Quasi-steady state (QSS) population of NO. Panel C: QSS population of N2. In panel A the features around 6 eV are due to…
Figure 9
Figure 9. Figure 9: Distributions of the forward rates for NO( [PITH_FULL_IMAGE:figures/full_fig_p026_9.png]
Figure 10
Figure 10. Figure 10: Distributions of the dissociation rates for [PITH_FULL_IMAGE:figures/full_fig_p027_10.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

43 extracted references · 41 canonical work pages

  1. [1]

    Anderson, J. D. Hypersonic and High Temperature Gas Dynamics, 2nd ed.; American Institute of Aeronautics and Astronautics: Reston, VA, 2006

    work page 2006

  2. [2]

    H \"o rst, S.; Yelle, R.; Buch, A.; Carrasco, N.; Cernogora, G.; Dutuit, O.; Quirico, E.; Sciamma-O'Brien, E.; Smith, M.; Somogyi, A. et al. Formation of amino acids and nucleotide bases in a Titan atmosphere simulation experiment. Astrobiology 2012, 12, 809--817

    work page 2012

  3. [3]

    N.; Qualls, G

    Desai, P. N.; Qualls, G. D. Stardust Entry Reconstruction. J. Spacecraft Rock. 2010, 47, 736--740

    work page 2010

  4. [4]

    Computation of atmospheric entry flow about a Leonid meteoroid

    Boyd, I. Computation of atmospheric entry flow about a Leonid meteoroid. Earth Moon Plan. 1998, 82-3, 93--108

    work page 1998

  5. [5]

    Boltzmann rovibrational collisional coarse-grained model for internal energy excitation and dissociation in hypersonic flows

    Munaf \`o , A.; Panesi, M.; Magin, T. Boltzmann rovibrational collisional coarse-grained model for internal energy excitation and dissociation in hypersonic flows. Phys.\ Rev. E 2014, 89, 023001

    work page 2014

  6. [6]

    G.; Jo, S

    Kim, J. G.; Jo, S. M. Modification of chemical-kinetic parameters for 11-air species in re-entry flows. Int. J. Heat Mass Transf. 2021, 169, 120950

    work page 2021

  7. [7]

    Construction of a coarse-grain quasi-classical trajectory method

    Macdonald, R.; Jaffe, R.; Schwenke, D.; Panesi, M. Construction of a coarse-grain quasi-classical trajectory method. I. Theory and application to N _2 --N _2 system. J. Chem.\ Phys. 2018, 148

    work page 2018

  8. [8]

    S.; Boyd, I

    Chaudhry, R. S.; Boyd, I. D.; Torres, E.; Schwartzentruber, T. E.; Candler, G. V. Implementation of a chemical kinetics model for hypersonic flows in air for high-performance CFD. AIAA Scitech 2020 Forum. 2020; p 2191

    work page 2020

Show all 43 references
  1. [9]

    Adaptive coarse graining method for energy transfer and dissociation kinetics of polyatomic species

    Sahai, A.; Lopez, B.; Johnston, C.; Panesi, M. Adaptive coarse graining method for energy transfer and dissociation kinetics of polyatomic species. J. Chem.\ Phys. 2017, 147

    2017

  2. [10]

    General multi-group macroscopic modeling for thermo-chemical non-equilibrium gas mixtures

    Liu, Y.; Panesi, M.; Sahai, A.; Vinokur, M. General multi-group macroscopic modeling for thermo-chemical non-equilibrium gas mixtures. J. Chem.\ Phys. 2015, 142

    2015

  3. [11]

    P.; Lopez, B.; Panesi, M

    Venturi, S.; Sharma, M. P.; Lopez, B.; Panesi, M. Data-Inspired and Physics-Driven Model Reduction for Dissociation: Application to the O _2 + O System. J. Phys.\ Chem. A 2020, 124, 8359--8372

    2020

  4. [12]

    M.; Venturi, S.; Sharma, M

    Jo, S. M.; Venturi, S.; Sharma, M. P.; Munafò, A.; Panesi, M. Rovibrational-Specific QCT and Master Equation Study on N _2 (X ^1 _ g ^ + ) + O( ^3 P) and NO(X ^2 ) + N( ^4 S) Systems in High-Energy Collisions. J. Phys.\ Chem. A 2022, 126, 3273--3290

    2022

  5. [13]

    Rodriguez, L.; Macdonald, R. L. Rovibrational-Specific Kinetics Database and Master Equations Study on NO(X ^2 ) + O ( ^3 P) and O _2(^3 _ g ^- )+ N( ^4 S) Systems. AIAA SCITECH 2025 Forum. 2025; p 2341

    2025

  6. [14]

    C.; Schwartzentruber, T

    Torres, E.; Geistfeld, E. C.; Schwartzentruber, T. E. High-temperature nonequilibrium air chemistry from first principles. J. Thermophys. Heat Transf. 2024, 38, 260--291

    2024

  7. [15]

    Bose, D.; Candler, G. V. Kinetics of the N _2 +O yields NO+N reaction under thermodynamic nonequilibrium. J. Thermophys. Heat Transf. 1996, 10, 148--154

    1996

  8. [16]

    Ab initio state-specific N _2 +O dissociation and exchange modeling for molecular simulations

    Luo, H.; Kulakhmetov, M.; Alexeenko, A. Ab initio state-specific N _2 +O dissociation and exchange modeling for molecular simulations. J. Chem.\ Phys. 2017, 146

    2017

  9. [17]

    S.; Bisek, N

    Valentini, P.; Grover, M. S.; Bisek, N. J. Toward ab initio molecular simulation of reacting air: Mach 15 air flow over a blunt wedge. J. Fluid Mech. 2024, 997, A27

    2024

  10. [18]

    C.; Bemish, R

    Koner, D.; San Vicente Veliz, J. C.; Bemish, R. J.; Meuwly, M. Accurate reproducing kernel-based potential energy surfaces for the triplet ground states of N _2 O and dynamics for the N + NO O + N _2 and N2 + O 2N + O reactions. Phys.\ Chem.\ Chem.\ Phys. 2020, 22, 18488--18498

    2020

  11. [19]

    Lin, W.; Varga, Z.; Song, G.; Paukku, Y.; Truhlar, D. G. Global triplet potential energy surfaces for the N _2 (X ^1 ) + O( ^3 P) → NO(X ^2 ) + N( ^4 S) reaction. J. Chem.\ Phys. 2016, 144, 024309

    2016

  12. [20]

    B.; Panesi, M

    Munafò, A.; Alberti, A.; Pantano, C.; Freund, J. B.; Panesi, M. A computational model for nanosecond pulse laser-plasma interactions. J. Comp.\ Phys. 2020, 406, 109190

    2020

  13. [21]

    Goswami, S.; Veliz, J. C. S. V.; Upadhyay, M.; Bemish, R. J.; Meuwly, M. Quantum and quasi-classical dynamics of the C( ^3 P)+ O _2 ( ^3 ^-_g ) CO ( ^1 ^+ ) + O( ^1 D) reaction on its electronic ground state. Phys.\ Chem.\ Chem.\ Phys. 2022, 24, 23309--23322

    2022

  14. [22]

    Reaction Dynamics of the H + HeH ^+ He+ H _2^+ System

    Upadhyay, M.; K \"a ser, S.; Sahoo, J.; Scribano, Y.; Meuwly, M. Reaction Dynamics of the H + HeH ^+ He+ H _2^+ System. Prec. Chem. 2025, in print

    2025

  15. [23]

    C.; Torres, E.; Schwartzentruber, T

    Geistfeld, E. C.; Torres, E.; Schwartzentruber, T. Quasi-classical trajectory analysis of three-body collision induced recombination in neutral nitrogen and oxygen. J. Chem.\ Phys. 2023, 159

    2023

  16. [24]

    Gradient-based methods for uncertainty quantification in hypersonic flows

    Lockwood, B.; Mavriplis, D. Gradient-based methods for uncertainty quantification in hypersonic flows. Computers & Fluids 2013, 85, 27--38

    2013

  17. [25]

    K.; Hosder, S

    West IV, T. K.; Hosder, S. Uncertainty quantification of hypersonic reentry flows with sparse sampling and stochastic expansions. J. Spacecraft Rock. 2015, 52, 120--133

    2015

  18. [26]

    E.; Boyd, I

    Holloway, M. E.; Boyd, I. D. Sensitivity analysis of thermochemical kinetics modeling for hypersonic air flows. J. Thermophys. Heat Transf. 2022, 36, 584--593

    2022

  19. [27]

    A general method for constructing multidimensional molecular potential energy surfaces from ab initio calculations

    Ho, T.-S.; Rabitz, H. A general method for constructing multidimensional molecular potential energy surfaces from ab initio calculations . J. Chem.\ Phys. 1996, 104, 2584

    1996

  20. [28]

    T.; Meuwly, M

    Unke, O. T.; Meuwly, M. Toolkit for the Construction of Reproducing Kernel-based Representations of Data: Application to Multidimensional Potential Energy Surfaces. J.\ Chem. Inf. Model. 2017, 57, 1923--1931

    2017

  21. [29]

    Permutationally Invariant, Reproducing Kernel-based Potential Energy Surfaces for Polyatomic Molecules: From Formaldehyde to Acetone

    Koner, D.; Meuwly, M. Permutationally Invariant, Reproducing Kernel-based Potential Energy Surfaces for Polyatomic Molecules: From Formaldehyde to Acetone. J. Chem.\ Theor.\ Comp. 2020, 16, 5474--5484

    2020

  22. [30]

    Veliz, J. C. S. V.; Koner, D.; Schwilk, M.; Bemish, R. J.; Meuwly, M. The C( ^3 P)+ O _2 ( ^3 _ g ) CO _2 CO( ^1 ^+ ) + O( ^1 D)/O( ^3 P) Reaction: Thermal and Vibrational Relaxation Rates from 15 K to 20000 K. Phys.\ Chem.\ Chem.\ Phys. 2021, 23, 11251--11263

    2021

  23. [31]

    Bernstein, R. B., Ed. Atom - Molecule Collision Theory; Springer New York, NY, 1979

    1979

  24. [32]

    Scattering studies of proton transfer reactions between rare gas atoms

    Koner, D. Scattering studies of proton transfer reactions between rare gas atoms. Ph.D.\ thesis, 2016

    2016

  25. [33]

    T.; Boe, K.; Bemish, R

    Koner, D.; Unke, O. T.; Boe, K.; Bemish, R. J.; Meuwly, M. Exhaustive state-to-state cross sections for reactive molecular collisions from importance sampling simulation and a neural network representation. J. Chem.\ Phys. 2019, 150, 211101

    2019

  26. [34]

    Gaussian weighting in the quasiclassical trajectory method

    Bonnet, L.; Rayez, J.-C. Gaussian weighting in the quasiclassical trajectory method. Chem. Phys. Lett. 2004, 397, 106--109

    2004

  27. [35]

    L.; Panesi, M

    Venturi, S.; Jaffe, R. L.; Panesi, M. Bayesian Machine Learning Approach to the Quantification of Uncertainties on Ab Initio Potential Energy Surfaces. J. Phys.\ Chem. A 2020, 124, 5129--5146

    2020

  28. [36]

    Schwenke, D. W. Calculations of rate constants for three-body recombination of H _2 in the presence of H _2 . J. Chem.\ Phys. 1988, 89, 2076--2091

    1988

  29. [37]

    Numerical treatment of ordinary differential equations by extrapolation methods

    Bulirsch, R.; Stoer, J. Numerical treatment of ordinary differential equations by extrapolation methods. Numerische Mathematik 1966, 8, 1--13

    1966

  30. [38]

    Review of chemical-kinetic problems of future NASA missions

    Park, C. Review of chemical-kinetic problems of future NASA missions. I-Earth entries. J. Thermophys. Heat Transf. 1993, 7, 385--398

    1993

  31. [39]

    C.; White, D

    Millikan, R. C.; White, D. R. Systematics of vibrational relaxation. J. Chem.\ Phys. 1963, 39, 3209--3213

    1963

  32. [40]

    Nonequilibrium hypersonic aerothermodynamics

    Park, C. Nonequilibrium hypersonic aerothermodynamics. 1989,

    1989

  33. [41]

    C.; Arnold, J.; Bemish, R

    San Vicente Veliz, J. C.; Arnold, J.; Bemish, R. J.; Meuwly, M. Combining machine learning and spectroscopy to model reactive atom+ diatom collisions. J. Phys.\ Chem. A 2022, 126, 7971--7980

    2022

  34. [42]

    L.; Schwenke, D

    Panesi, M.; Jaffe, R. L.; Schwenke, D. W.; Magin, T. E. Rovibrational internal energy transfer and dissociation of N _2 (^1 _ g ^+) --N( ^4 S _ u ) system in hypersonic flows. J. Chem.\ Phys. 2013, 138

    2013

  35. [43]

    intermediate model

    Grover, M. S.; Valentini, P.; Bisek, N. J.; Jo, S. M.; Davuluri, R.; Martin, A.; Notey, A.; Panesi, M.; Andrienko, D.; Hanquist, K. M. et al. Comparative Analysis of Thermophysics Models for the Simulations of Nonequilibrium Flow. AIAA SCITECH 2025 Forum. 2025; p 0641 mcitethe...

    2025

Pith tools

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