Pith. sign in

REVIEW 4 major objections 5 minor 61 references

Topology induced modifications in the critical behavior of the Yaldram Khan catalytic reaction model

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 The paper claims that the spatial structure of randomness on a catalytic surface controls whether the Yaldram–Khan model's discontinuous phase transition survives: random geometric graphs convert it to continuous at low connectivity…

desk verdict YK model on random networks is new and the ERN results are solid, but the RGG order-conversion claim is visually inferred, not demonstrated. read the letter →

arxiv 2506.04485 v1 pith:YVAIEFYL submitted 2025-06-04 cond-mat.stat-mech

Paulo F. Gomes , Henrique A. Fernandes , Roberto da Silva This is my paper
classification cond-mat.stat-mech MSC 82C2682C2005C80
keywords Yaldram–KhanmodelcatalyticsurfacereactionsphasetransitionorderrandomgeometricgraphErdős–RényinetworknonequilibriumsteadystateMonteCarlosimulationreactivewindow
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 asks whether the kind of randomness used to model a real catalyst surface can change the phase diagram of the Yaldram–Khan (YK) reaction, a standard Monte Carlo model of NO + CO conversion. On a regular hexagonal lattice the model has a reactive window bracketed by one continuous and one discontinuous phase transition. Simulating the same reaction on two random network surfaces, the authors report that an Erdős–Rényi network preserves both transitions for every average degree studied, while a random geometric graph converts the discontinuous transition into a continuous one at small average degree, recovering the jump only near K ≈ 9. This matters because the order of the transition controls whether the catalytic window appears abruptly or gradually on disordered surfaces.

What carries the argument

The central objects are two random network ensembles used as catalytic surfaces: the Erdős–Rényi network, in which every pair of sites is connected with fixed probability, giving long-range randomness; and the random geometric graph, in which sites are placed uniformly in a square and connected when closer than a radius, giving spatially constrained randomness. The order parameter is the steady-state density of vacant sites ρV as a function of the CO adsorption rate y, and the reactive window y2 − y1 between the two transition points is used to quantify the active phase. The paper reads the order of each transition off the shape of ρV(y): a sharp drop marks a discontinuous transition, a smooth bend a continuous one.

What would settle it

Compute Binder cumulants of the order parameter or the probability histogram of ρV for the RGG at K ≈ 4.5 across several system sizes, say N = $64^{2}$, $128^{2}$, and $256^{2}$: if the histogram is bimodal or the cumulant minimum deepens with N, the upper transition is discontinuous, contradicting the paper's claim; if the cumulant curves cross consistently, the second-order assignment holds.

Watch

Extended reading notes

Core claim

The paper's central claim is that how randomness is placed on the catalytic surface determines whether the reaction's phase transitions keep their character. On an Erdős–Rényi network, where edges connect sites without regard to spatial distance, the YK model retains one continuous and one discontinuous transition for all average degrees studied (μ from about 2 to 9), with the reactive window beginning near μ ≈ 2.1. On a random geometric graph, where edges only join sites within a fixed radius, the upper (originally discontinuous) transition appears to become continuous for small average degree (reactive window starting between K ≈ 3.8 and 4.5), and the discontinuous character returns only for K ≈ 9.0. The same steady-state Monte Carlo protocol, with NO dissociation rate rNO = 1 and density of vacant sites as the order parameter, is used throughout, so the contrast is attributed to topology.

Load-bearing premise

The assignment of transition order is based on visually judging whether the density curves are smooth or sharp at a single system size (N = $128^{2}$), not on finite-size scaling, Binder cumulants, or order-parameter histograms.

Editorial extensions

If this is right

  • For an Erdős–Rényi surface, the YK model's two transitions remain first and second order for every average degree studied, so long-range randomness alone does not change the nature of the transitions.
  • For a random geometric surface at small average degree (K below about 9), the upper transition is claimed to become continuous, so the catalyst would enter and leave the reactive state without a discontinuous jump.
  • The reactive window widens with average connectivity on both networks, and for the RGG it begins to open between K ≈ 3.8 and 4.5.
  • If the two continuous transitions persist for rNO < 1, the model would exhibit two lines of second-order transitions in the (rNO, y) plane, a rare phase-diagram structure.
  • The contrast between the two networks implies that spatially localized randomness, rather than randomness in general, is what alters the order of the phase transition in this model.

Reading between the lines

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

  • Inference: The claim of a converted transition order would be much stronger with finite-size scaling; a Binder-cumulant or order-parameter-histogram study across N = 64^2, 128^2, and 256^2 at fixed K would confirm whether the smooth ρV curve is a true second-order transition or finite-size rounding.
  • Inference: Because the small-K RGG is not fully percolated (sc ≈ 0.05 at K ≈ 3.8), the effective catalytic surface is fragmented; the observed continuous upper transition may be an effect of disconnected reaction clusters rather than of local geometric randomness per se, and a study restricting dynamics to the giant component could separate the two.
  • Inference: Interpolating between the RGG and the ERN by gradually rewiring edges while keeping the average degree fixed could isolate whether it is spatial locality or network fragmentation that suppresses the discontinuous transition.
  • Inference: A direct experimental analogue would be a catalytic pellet whose pore connectivity is locally constrained rather than globally random; the model predicts the former should show smoother poisoning and ignition curves at low connectivity.
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 paper studies the Yaldram–Khan (YK) catalytic reaction model on three surfaces: the hexagonal lattice, Erdős–Rényi networks (ERN), and random geometric graphs (RGG), using steady-state Monte Carlo simulations at NO dissociation rate rNO=1. The authors report that the ERN preserves the qualitative phase diagram of the hexagonal lattice—one continuous and one discontinuous transition—for all average degrees considered, while the RGG appears to convert the upper, first-order transition into a second-order one for small average degree K, recovering a discontinuous transition only near K≈9. The paper also presents the reactive-window width as a function of average degree for both networks and discusses percolation of the RGG substrates.

Significance. If the central claim is correct, the RGG result is a genuinely interesting finding: local, spatially constrained randomness in the catalytic surface would change the order of a nonequilibrium phase transition, whereas long-range ERN randomness would not. The manuscript provides a useful dataset of phase boundaries and reactive-window widths for the YK model on two network families, and the authors are careful to report simulation parameters (N, τ, S) and to compare against known hexagonal-lattice transition points. However, the central order-of-transition claim is not yet established: it is inferred from the visual smoothness of density curves at a single system size, with no finite-size scaling, histogram analysis, or quantitative order-parameter diagnostic. The paper therefore needs substantial additional analysis before the central conclusion can be accepted.

major comments (4)
  1. [Sec. III.D, Figs. 4(c), 8(a)] The central claim that the RGG converts the upper YK transition from first to second order for small K rests on the visual smoothness of ρV(y) at a single system size N=128^2. In a finite system a first-order transition is rounded, so the smooth approach to y2 for K≈4.5–6.0 could be finite-size rounding rather than a genuine continuous transition. No finite-size scaling, Binder cumulant, order-parameter histogram, or hysteresis analysis is provided for the RGG. I ask for multi-N simulations at representative K values (e.g., K≈4.5, 6.06, and 9.1) with a quantitative order-parameter criterion—such as bimodality of the density distribution or crossing of a cumulant—before the order conversion is claimed.
  2. [Sec. IV vs Sec. III.D] The Conclusions state without qualification that 'the first-order phase transition is converted into second one for small values of the average degree K' and that the discontinuity is recovered near K≈9.0. This conflicts with the hedged language in Sec. III.D ('seems', 'suggests', 'need for further investigation'). The conclusions should be explicitly conditioned on the finite-size and order-parameter analysis requested above; as written, the abstract and conclusions assert what the results only suggest.
  3. [Appendix A and Figs. 5(b), 8(b)] The extraction of y1 and y2 is not defined, and the transition points in Figs. 5(b) and 8(b) are reported without error bars, even though the density curves in Fig. 8(a) carry standard errors from g=21 runs. The manuscript should state the criterion used to locate a transition on the ρV(y) curves and provide errors on y1 and y2 (e.g., from a threshold rule or an extrapolation procedure). Without this, the reported reactive-window widths are not quantitatively reproducible.
  4. [Appendix A, Sec. III.D] The averaging procedure over disorder is unclear. The appendix says one network instance is generated per y value and that the quoted K for the RGG is an average over all y, while Fig. 8(a) reports standard errors from g=21 runs. It is not stated whether the 21 runs at fixed y use the same graph or independently generated graphs, so the error bars may reflect only time/seed fluctuations and not quenched topological disorder. This should be clarified and, ideally, observables should be averaged over several RGG realizations at each (y,K). The issue is particularly relevant at small K, where at K≈3.8 the largest component has sc≈0.05457 and the network is not percolated.
minor comments (5)
  1. [Fig. 1 caption] The caption labels two panels as '(b)'; the RGG panel should be labeled '(c)'.
  2. [Fig. 2 caption] The caption writes 'N = 322, = 642 and 1282'; these should be typeset as 32^2, 64^2, and 128^2 for clarity.
  3. [Sec. II.B and Sec. III.B] The notation alternates between m, µ, and K for the average degree; a short explicit statement that m is the measured value, µ the control parameter for the ERN, and K the reported final average for both networks would avoid confusion.
  4. [Eq. (8) and surrounding text] In the definition of ρξ, the list ξ = V, O, CO, N2, CO2 mixes surface species with gas-phase products; this is not incorrect, but a sentence clarifying that N2 and CO2 are counted only as produced molecules, not as surface densities, would improve readability.
  5. [Sec. III.C, Fig. 7] The snapshots present configurations for µ=2.0, 3.0, and 9.0, but the text says the active window begins at µ≈2.1; it would be helpful to state explicitly that the µ=2.0 case is an absorbing state, as the reader must infer this from ρV=0.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central claim is a direct Monte Carlo measurement with no fitted parameters, no prediction-from-fit step, and no load-bearing self-citation chain.

full rationale

The paper's central content is a steady-state Monte Carlo study of the Yaldram-Khan model on two random networks. The main observable, the density of vacant sites rho_V as a function of y, is measured directly from simulation, and the phase-transition points y1 and y2 are read from those curves rather than produced by a fitting procedure. No parameter is fitted to a subset of data and then renamed as a prediction. The average degree K is a measured network property, not an adjustable parameter used to force the order of the transition; indeed, the paper explicitly reports K obtained from the average of all generated instances. Self-citations by the authors (e.g., Refs. [30], [40], [46]) provide context from prior ZGB-network studies and scaling conventions, but the order-of-transition claim for the RGG is not derived from those citations. The comparison values for the hexagonal lattice come from an independent literature source, Ref. [47]. The paper invokes no uniqueness theorem and imports no ansatz via self-citation. The main weakness is statistical and methodological: the assignment of transition order relies on visual smoothness of density curves at a single system size, without Binder cumulants, histograms, or finite-size scaling. That is a validity concern about the strength of the evidence, not a circularity in which the conclusion is equivalent to the input by definition. The paper itself acknowledges this uncertainty in Sec. III.D, stating that the y2 transitions 'seem to change' and that 'further investigation' is needed, and in Sec. III.D it also reports the non-percolated largest-component densities for small K. These limitations are explicit in the manuscript and weigh against the strength of the conclusion, but they do not make the derivation circular. The central claim is a direct numerical measurement, so no self-definitional, fitted-input, or self-citation-based circularity is present.

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

The paper contributes a Monte Carlo measurement and introduces no new free parameters or invented entities. The main burden is not circularity but the unsupported identification of transition order at a single system size.

assumptions (3)
  • domain assumption The Yaldram-Khan reaction steps, Eqs. (2) through (7), with rNO=1 are the correct microscopic dynamics for the NO-CO reaction on the studied surfaces.
    Imported from Ref. [16] and used without modification in Section II.A; all results are conditional on this model.
  • domain assumption The system reaches a unique steady state after tau = 10^6 Monte Carlo steps for all networks and parameter values.
    Calibration for the hexagonal lattice in Section III.A is assumed to transfer to the random networks; no independent convergence check for the ERN or RGG is reported.
  • domain assumption Averaging over one random network instance per y value represents the quenched-disorder ensemble for the phase diagram.
    Stated in Section III.B and Appendix A; fluctuations of the phase boundaries between network instances are not quantified.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Topology induced modifications in the critical behavior of the Yaldram Khan catalytic reaction model." pith.science (2026). https://pith.science/paper/YVAIEFYL

@misc{pith2026250604485,
  author       = {Pith},
  title        = {Pith review of: Topology induced modifications in the critical behavior of the Yaldram Khan catalytic reaction model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YVAIEFYL}},
  note         = {Machine review of arXiv:2506.04485}
}
read the original abstract

In this work, we investigated how the use of complex networks as catalytic surfaces can affect the phase diagram of the Yaldram-Khan model, as well as how the order of the phase transitions present in the seminal work behaves when the randomness is added to the model. The study was conducted by taking into consideration two well-known random networks, the Erdos-Renyi network (ERN), with its long-range randomness, and the random geometric graph (RGG), with its spatially constrained randomness. We perform extensive steady-state Monte Carlo simulations assuming the NO dissociation rate is equal to 1 and show the behavior of the reactive window as function of the average degree of the networks. Our results also show that, different from the ERN, which preserves the nature of the phase transitions of the original model for all considered average degrees, the RGG seems to have two second-order phase transitions for small values of average degree.

Figures

Figures reproduced from arXiv: 2506.04485 by the authors.

Figure 1
Figure 1. FIG. 1. Graphical representation of networks. (a) Regular hexagonal lattice with [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. shows the density of vacant sites, ρV , as a function of the CO adsorption rate, y, for the hexagonal lattice, for N = 1024 = 322 , 4096 = 642 and 16384 = 1282 . In both FIG. 2. Density ρV as a function of y for N = 322 , = 642 and 1282 with τ = 5 × 106 and S = 1 for the HN. cases, the densities were calculated after τ = 5 × 106 MC steps and no average sampling (S = 1). The idea is to identify the minimum size N tha… view at source ↗
Figure 3
Figure 3. FIG. 3. Analysis of [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Densities [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Steady reactive window for the ERN with different values for the average degree [PITH_FULL_IMAGE:figures/full_fig_p011_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6. Phase diagram in the ( [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Snapshot of the Erd¨os-R´enyi network with different values of [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. Steady-state reactive window for the RGG with different values for the average degree [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. Snapshot of the RGG with different average degrees [PITH_FULL_IMAGE:figures/full_fig_p015_9.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

61 extracted references · 58 canonical work pages

  1. [1]

    The incident molecule is sorted: CO with chance y and NO with probability 1 − y

    work page

  2. [2]

    If it is occupied, nothing happens and a new analysis is initiated

    A random site s1 is selected. If it is occupied, nothing happens and a new analysis is initiated. The following steps happen if the site is vacant

    work page

  3. [3]

    Furthermore, if there is a neighboring oxygen-occupied site, there is a reaction creating CO2 (Eq

    If a CO molecule is selected, it occupies the selected site s1. Furthermore, if there is a neighboring oxygen-occupied site, there is a reaction creating CO2 (Eq. 7), which leaves the network. The two sites become vacant. The analysis ends and a new one is initiated

    work page

  4. [4]

    If the selected site s1 has neighbors, one of them is randomly selected and called s2

    If a NO molecule is selected, it is dissociated in N and O (as we use d = 1 .0). If the selected site s1 has neighbors, one of them is randomly selected and called s2. If s2 is empty, N and O enter the network. There are two possibilities with equal probabilities: (a) N occupies s1 and O occupies s2 or (b) vice versa

    work page

  5. [5]

    If it finds one, reaction 6 occurs: N2(g) is produced and the two sites become empty

    Considering the first possibility (a), once the nitrogen atom is adsorbed and before the oxygen one is adsorbed, nitrogen searches for nitrogen-occupied neighbors other than s2. If it finds one, reaction 6 occurs: N2(g) is produced and the two sites become empty. However, if it finds a NO-occupied site, reaction 5 occurs: N2(g) is produced and oxygen 17 i...

    work page

  6. [6]

    If this search is successful, reaction 7 occurs producing CO2 and two involved sites become empty

    Considering the oxygen absorbed on site s2, it looks for an CO-occupied neighbor site. If this search is successful, reaction 7 occurs producing CO2 and two involved sites become empty

    work page

  7. [7]

    The overall algorithm is the following:

    The case of the second possibility (b) is equivalent. The overall algorithm is the following:

    work page

  8. [8]

    Input parameters: N , S, τ , y and µ or r

    work page

Show all 61 references
  1. [9]

    The average degree K is calculated

    An instance of the network is generated on a random basis. The average degree K is calculated

  2. [10]

    The states are reset to the initial states: all sites are empty

  3. [11]

    The states are updated

    Steady state stage: time evaluation over τ MCS. The states are updated

  4. [12]

    The averages of ρξ are calculated

    Sampling stage over S MCS. The averages of ρξ are calculated. Considering the RGG, the used value of K in the results is the average considering all the networks of all y values

  5. [13]

    K Nørskov, F

    J. K Nørskov, F. Abild-Pedersen, F. Studt, T. Bligaard PNAS, 108, 937–943 (2008)

    2008

  6. [14]

    Hammer, J

    B. Hammer, J. K. Nørskov, Adv. Catal. 45, 71-129 (2000)

    2000

  7. [15]

    J. W. Niemantsverdriet, Spectroscopy in Catalysis: An Introduction, Wiley-VCH (2007)

    2007

  8. [16]

    Hinrichsen, Non-equilibrium critical phenomena and phase transitions into absorbing states, Adv

    H. Hinrichsen, Non-equilibrium critical phenomena and phase transitions into absorbing states, Adv. Phys. 49, 815 (2000)

    2000

  9. [17]

    Henkel, M

    M. Henkel, M. Pleimling, Non-equilibrium Phase Transitions, Vol. 2: Ageing and Dynamical Scaling far from Equilibrium, Springer, Dordrecht (2010)

    2010

  10. [18]

    T. Tome, M. J. de Oliveira, J. Phys. A 55, 275602 (2022)

    2022

  11. [19]

    da Silva

    R. da Silva. H. A. Fernandes, J. Stat. Mech. P06011 (2015)

    2015

  12. [20]

    da Silva, M

    R. da Silva, M. J. de Oliveira, T. Tome, J. R. Drugowich de Fel ´ ıcio, Phys. Rev. E 101, 012130 (2020)

    2020

  13. [21]

    R.M. Ziff, E. Gulari, and Y. Barshad, Phys. Rev. Lett. 56, 2553 (1986). 18

    1986

  14. [22]

    Marro and R

    J. Marro and R. Dickman, Nonequilibrium Phase Transitions in Lattice Models (Cambridge Uni- versity Press, Cambridge, U.K., 1999)

    1999

  15. [23]

    Tom´ e and R

    T. Tom´ e and R. Dickman, Phys. Rev. E47, 948 (1993)

    1993

  16. [24]

    da Silva, H

    R. da Silva, H. A. Fernandes, Comp. Phys. Commun. 230, 1-9 (2018)

    2018

  17. [25]

    Fernandes, R

    H.A. Fernandes, R. da Silva, E.D. Santos, P.F. Gomes, and E. Arashiro, Phys. Rev. E 94, 022129 (2016)

    2016

  18. [26]

    H. A. Fernandes, R. da Silva, A. B. Bernardi, Phys. Rev. E 98, p. 1-9, (2018)

    2018

  19. [27]

    H. A. Fernandes, R. da Silva, Phys. Rev. E. 111, 044101 (2025)

    2025

  20. [28]

    Yaldram and M

    K. Yaldram and M. A. Khan, Journal of Catalysis 131, 369-377 (1991)

    1991

  21. [29]

    K. M. Khan and W. Ahmad, J. Phys. A: Math. Gen. 35, 2713 (2002)

    2002

  22. [30]

    Loscar and E

    E. Loscar and E. V. Albano, Rep. Prog. Phys. 66, 1343 (2003)

    2003

  23. [31]

    L. A. Avalos, V. Bustos, R. U˜ nac, F. Zaera, and G. Zgrablich, J. Phys. Chem. B 110, 24964 (2006)

    2006

  24. [32]

    Hui-Yun and W

    P. Hui-Yun and W. H. Jun, Physica A 227, 234 (1996)

    1996

  25. [33]

    A. G. Dickman, B. C. S. Grandi, W. Figueiredo, and R. Dickman, Phys. Rev. E 59, 6361 (1999)

    1999

  26. [34]

    M. A. Khan, K. Yaldram, G. K. Khalil, and K. M. Khan, Phys. Rev. E 50, 2156 (1994)

    1994

  27. [35]

    T. Aida, D. Na-Ragong, R. Kobayashi, and H. Niiyama, Chem. Eng. Sci. 54, 4449 (1999)

    1999

  28. [36]

    J. J. Luque, A. G´ omez, and A. C´ ordoba, Physica A331, 505 (2004)

    2004

  29. [37]

    C. D. Lorenz, R. Haghgooie, C. Kennebrew, R. M. Ziff, Surf. Sci. 517, 75 (2002)

    2002

  30. [38]

    Ahmad and M.K

    W. Ahmad and M.K. Baloch, App. Surf. Sci. 253, 8447 (2007)

    2007

  31. [39]

    Meng, H Weinberg, and J

    B. Meng, H Weinberg, and J. W. Evans, J. Chem. Phys. 101, 3234–3242 (1994)

    1994

  32. [40]

    J. J. D ´ ıaz and G. M. Buend ´ ıa, Physica A491, 13 (2018)

    2018

  33. [41]

    E. J. Hern´ andez and G. M. Buend ´ ıa, Int. J. Mod. Phys. B36, 2240002 (2022)

    2022

  34. [42]

    Vilela, H.A

    E.B. Vilela, H.A. Fernandes, F.L.P. Costa e P.F. Gomes, Journal of Computational Chemistry 41, 1964 (2020)

    2020

  35. [43]

    M. E. J. Newman, Networks: an introduction. Oxford University Press, Oxford (2010). https: //doi.org/10.1093/oso/9780198805090.001.0001

    2010

  36. [44]

    Albert, A-L

    R. Albert, A-L. Barab´ asi, Rev. Mod. Phys. 74, 47 (2002)

    2002

  37. [45]

    Gilbert, Random plane networks, Journal of the Society for Industrial and Applied Mathe- matics, 9, 533 (1961)

    E. Gilbert, Random plane networks, Journal of the Society for Industrial and Applied Mathe- matics, 9, 533 (1961). https://doi.org/10.1137/0109045

    1961 doi

  38. [46]

    Seshadhri, T.G

    C. Seshadhri, T.G. Kolda, and Ali Pinar, Community structure and scale-free collections of Erd¨ os-R´ enyi graphs, Phys. Rev. E,85(5), 056109 (2012). https://link.aps.org/doi/10.1103/ PhysRevE.85.056109

    2012

  39. [47]

    T. W. Moore, P. D. Finley, B. J. Apelberg, B. K. Ambrose, N. S. Brodsky, T. J. Brown, C. Husten, 19 R. J. Glass, An opinion-driven behavioral dynamics model for addictive behaviors. European Physical Journal B 88, 95 (2015). https://doi.org/10.1140/epjb/e2015-40462-y

    2015 doi

  40. [48]

    Saeedian, M

    M. Saeedian, M. Miguel, R. Toral, Absorbing phase transition in the coupled dynamics of node and link states in random networks. Scientific Reports 9, 9726 (2019). https://doi.org/10. 1038/s41598-019-45937-y

    2019

  41. [49]

    Newman, D.J

    M.E.J. Newman, D.J. Watts, and S.H. Strogatz, Random graph models of social networks, Proc. Natl. Acad. Sci. USA 99, 2566 (2002). https://doi.org/10.1073/pnas.012582999

    2002 doi

  42. [50]

    Reia, P.F

    S.M. Reia, P.F. Gomes e J.F. Fontanari, The European Physical Journal B 92, 205 (2019)

    2019

  43. [51]

    Reia, P.F

    S.M. Reia, P.F. Gomes e J.F. Fontanari, Journal of Statistical Mechanics 033402 (2020)

    2020

  44. [52]

    Gomes, H.A

    P.F. Gomes, H.A. Fernandes e A.A. Costa, Physica A 597, 127269 (2022)

    2022

  45. [53]

    F. F. Franco, and P. F. Gomes, Eur. Phys. J. B 97, 190 (2024). https://doi.org/10.1140/ epjb/s10051-024-00828-8

    2024

  46. [54]

    Solomonoff, A

    R. Solomonoff, A. Rapoport, Connectivity of random networks. Bulletin of Mathematical Biology 13, 107-117 (1951). https://doi.org/10.1016/S0092-8240(76)80054-7

    1951 doi

  47. [55]

    Erd¨ os, A

    P. Erd¨ os, A. R´ enyi, On random graphs. Publicationes Mathematicae6, 290-297 (1959)

    1959

  48. [56]

    A. L. Barab´ asi, Network Science, Cambridge University Press, Glasgow (2016). https:// networksciencebook.com/

    2016

  49. [57]

    Dall e M

    J. Dall e M. Christensen, Physical Review E 66, 016121 (2002)

    2002

  50. [58]

    Gomes, S.M

    P.F. Gomes, S.M. Reia, F.A. Rodrigues e J.F. Fontanari, Physical Review E 99, 032301 (2019)

    2019

  51. [59]

    B. J. Brosilow, and R. M. Ziff, J. Catal. 136, 275 (1992)

    1992

  52. [60]

    Bagnoli, N

    F. Bagnoli, N. Boccara, and R. Rechtman, Phys. Rev. E, 63, 046116 (2001). https://doi.org/ 10.1103/PhysRevE.63.046116

    2001 doi

  53. [61]

    Bagnoli, F

    F. Bagnoli, F. Franci, and R. Rechtman, Phys. Rev. E, 71, 046108 (2005). https://doi.org/ 10.1103/PhysRevE.71.046108

    2005 doi

Pith tools

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