Pith. sign in

REVIEW 2 major objections 6 minor 34 references

Dynamics of interacting cavity solitons

T0 review · 2 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read This paper derives asymptotically exact equations for the drift velocity of two Kerr cavity solitons in the Lugiato-Lefever model, showing that tail shape alone decides whether the pair repels or forms a molecule.

desk verdict A sound, parameter-free asymptotic derivation of Kerr soliton pair drift, with clean classification and credible numerics; the main gaps are reproducibility and asserted orderings, not correctness. read the letter →

arxiv 2507.07851 v1 pith:2H2B5NGY submitted 2025-07-10 nlin.PS physics.optics

Amir Leshem , Sanzida Akter , Logan Courtright , Pradyoth Shandilya , Curtis R. Menyuk , Omri Gat This is my paper · ORCID
classification nlin.PSphysics.optics MSC 35Q5135Q5537K4078A60 PACS 42.65.Tg05.45.Yv
keywords cavitysolitonsLugiato-LefeverequationsolitoninteractionsdriftvelocitymoleculestranslationalzeromodetailasymptoticsKerrmicroresonators
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 derives equations of motion for the positions of two bright solitons in a pumped Kerr cavity, described by the Lugiato-Lefever model, that become asymptotically exact as the soliton separation grows. The interaction arises because the exponential tails of the two solitons overlap, and this overlap couples to the global position variables associated with broken translation symmetry. For monotone tails the drift velocity is $d\xi/dt = b e^{-2\sigma \xi}$ and the interaction is purely repulsive; for oscillatory tails it is $d\xi/dt = B e^{-2\sigma_r \xi}\sin(2\sigma_i \xi - \beta)$, so attraction and repulsion alternate and stable fixed points correspond to soliton molecules. In both cases the coefficients are computed from single-soliton data alone. The paper checks these predictions against trajectories extracted from direct numerical solutions of the wave equation.

What carries the argument

The machinery is the tail-overlap projection method. A two-soliton waveform is written as two translated single solitons plus a constant pedestal plus a residual field $\Phi$, with the decomposition fixed by requiring that $\Phi$ be orthogonal to the two adjoint translational zero modes (gauge conditions (11)). The overlap parameter $\varepsilon = e^{-\sigma_r |\xi_2-\xi_1|}$ is the small parameter: the residual field is order $\varepsilon$ and its time derivative is order $\varepsilon^2$, so projecting the wave equation onto the adjoint zero modes leaves a single overlap integral (17). Replacing the far soliton by its tail asymptotics (3) or (4) converts that integral into the explicit drift laws (18) and (24), whose coefficients $b$, $B$, and $\beta$ depend only on the single-soliton waveform.

What would settle it

Run a direct numerical simulation of the Lugiato-Lefever equation for a parameter set with oscillatory tails, extract the two soliton positions over time for several large initial separations, and compare the measured drift velocity divided by $e^{-2\sigma_r \xi}$ with the predicted sinusoid $B\sin(2\sigma_i \xi - \beta)$; a systematic deviation that does not vanish as the separation grows would show that the leading-order projection misses a coupling.

Watch

Extended reading notes

Core claim

The central claim is that the slow drift of well-separated Kerr cavity solitons is governed by a closed, asymptotically exact effective equation for the soliton separation. Projecting the Lugiato-Lefever nonlinearity onto the adjoint zero mode of the translation symmetry, and using the known single-soliton tail asymptotics to simplify the resulting overlap integrals, the authors obtain Eq. (18) for monotone tails and Eq. (24) for oscillatory tails. The sign of the prefactor $b$ in the monotone case is positive in all studied examples, so the interaction there is purely repulsive; in the oscillatory case the prefactor $B$ multiplies a sine that changes sign periodically, producing two interlaced lattices of stable and unstable fixed points whose spacing is set by the imaginary part of the tail decay rate. Direct numerical simulation of the wave equation confirms the predicted velocity curves in the large-separation regime.

Load-bearing premise

The result stands on the assumption that the residual field left after subtracting two moving single solitons is as small as the tail overlap and changes slowly, so that its direct coupling to the soliton positions can be neglected at leading order.

Editorial extensions

If this is right

  • Solitons with monotone tails always repel one another at large separation, so no bound two-soliton state can form in that parameter regime.
  • Solitons with oscillatory tails experience attraction and repulsion in alternating, equal-length windows of separation, with stable fixed points that correspond to soliton molecules.
  • Because the drift velocity falls off exponentially, the separation of a repelling pair grows only logarithmically in time.
  • Higher-order bound states are so weakly bound that in practice only tightly bound molecules form within experimentally accessible time scales.
  • For multi-soliton waveforms, the interaction is dominated by nearest-neighbor overlaps, so the two-body law governs the approach to uniformly spaced soliton crystals.

Reading between the lines

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

  • A direct extension the authors hint at but do not develop: in waveforms with more than two solitons, nearest-neighbor dominance means the two-body drift law can be summed to predict the spacing statistics of soliton crystals, giving a testable prediction for lattice uniformity.
  • Because the binding force decays exponentially with separation, the stable fixed points in the oscillatory-tail regime are shallow potential wells; an estimate of thermal or pump-noise-induced escape rates from these wells would tell whether the predicted high-order molecules are observable at all.
  • The formula implies a clean experimental signature: measured molecule separations should be spaced by approximately $\pi/\sigma_i$, so microresonator experiments with tunable detuning could verify the oscillatory-tail parameter directly.
  • One could extend the projection calculation to include weak high-order dispersion or pump modulation, following the same zero-mode projection, and check whether the drift law retains the same exponential-sinusoid form with renormalized coefficients.
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

2 major / 6 minor

Summary. The paper derives effective equations of motion for the positions of two well-separated Kerr solitons in the Lugiato-Lefever equation. Using a two-soliton ansatz, a projection onto the adjoint translation zero mode, and the exponential tail asymptotics of the single soliton, it obtains explicit drift formulas: dξ/dt = b e^{-2σξ} for monotone tails and dξ/dt = B e^{-2σr ξ} sin(2σi ξ − β) for oscillatory tails, with b, B, β expressed as single-soliton integrals. The formulas are compared with drift velocities extracted from direct numerical solutions of Eq. (1) for several parameter sets.

Significance. If the result holds, the paper reduces the interaction of a Kerr soliton pair to a one-dimensional ODE with no fitted parameters, and it provides a concrete explanation of the repulsive/oscillatory interaction dichotomy and of soliton-molecule formation timescales. The main strengths are the absence of free parameters, the fact that all coefficients are computed from single-soliton data, the external validation against direct PDE simulation, and the falsifiable predictions about molecule formation. The method itself is not new, but its application to Kerr solitons with both monotone and oscillatory tails is a useful and timely contribution to the microresonator literature.

major comments (2)
  1. [Section II.B, Eqs. (10)-(15)] The derivation rests on the ordering assumptions Φ=O(ε), ∂Φ/∂t=O(ε^2), and dξ_i/dt=O(ε), but these are asserted rather than proved. In particular, the paper does not justify that the residual field Φ remains small and localized on the drift time scale, nor that the projection onto the adjoint zero mode captures all leading-order coupling. Since Eq. (15) and the cancellation of the linear-in-Φ terms in the class-2 expansion depend on these assumptions, the claim that the equation of motion is 'asymptotically exact' should either be accompanied by a precise set of hypotheses (with a reference to a rigorous pulse-interaction result) or be softened to 'formally asymptotically exact under the stated ordering assumptions.' This is a load-bearing gap in the presentation, though not a demonstrated error.
  2. [Abstract and Section III.A, Eqs. (18)-(19)] The abstract states that the interaction is 'either purely repulsive, or alternates between attraction and repulsion, according to whether the decay of soliton tails is monotone or oscillatory.' However, Eq. (18) with b defined by Eq. (19) permits b<0, which would give pure attraction for monotone tails. The authors only report that b is positive 'in all the cases that we studied' (Section III.A) and later state that they 'have not been able to rule out this possibility' (Section IV). The abstract therefore overstates the proven classification; it should be qualified to the parameter regimes studied, or the paper should supply an argument that b>0 for all monotone-tail solitons. This does not invalidate the drift formulas, but it is a central qualitative claim and should be corrected.
minor comments (6)
  1. [Section II.A] Equations (7) and (8) are identical; one of the two displays should be deleted.
  2. [Section II.B, Eq. (16)] The first line of Eq. (16) uses Ψs2 without a tilde while the second line uses ψ̃s2; since the ansatz is expressed in terms of subtracted solitons, the notation should be made consistent.
  3. [Section III, Eq. (17)] The sentence before Eq. (17) reads 'assumeξ1 = ξ = −ξ2'; with this literal choice the argument of ψ̃s in Eq. (17) should be x+ξ, not x+2ξ. The formulas are consistent if the origin is shifted to the first soliton, so that the second soliton is at −2ξ, and the text should state this shift explicitly.
  4. [Section III, Figs. 1-2] The numerical validation does not state the discretization, the time-integration scheme, or the procedure used to extract soliton trajectories from the PDE solutions. These details should be provided, at least in an appendix or supplement, for reproducibility.
  5. [References] Reference [16] has a formatting error: 'J. Carr and R. L. Pego, , Communications...' should include the title of the work.
  6. [Section III.B, Eq. (24)] The fixed-point stability statement after Eq. (24) relies on taking B as a positive modulus in the polar representation; the text should state this explicitly, since the assignment of stability to even/odd n would reverse if B were defined with the opposite sign.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the drift laws are derived from the single-soliton waveform and benchmarked against direct PDE simulation.

full rationale

The central claim, Eqs. (18) and (24), is not an input in disguise. The coefficients b, b+, b-, and β are defined as projection integrals over the single-soliton waveform (Eqs. 19, 22, 23), and the exponential/sinusoidal forms follow from replacing the neighbor tail by its known asymptotic forms (3) or (4), which are results about the linearized tail of the same soliton, not about two-soliton drift. The comparison in Figures 1 and 2 is an independent numerical benchmark: the drift velocities are extracted from direct numerical solution of the PDE (1) and matched against the formula, with no fitting parameters. The only weak spots are assumptions in Section II.B—|Φ| = O(ε), ∂Φ/∂t = O(ε²), and O(ε²) terms are negligible—which are asserted rather than rigorously proved; this is an omitted-proof gap in the asymptotic expansion, not a circularity, and the claimed large-separation agreement with numerics is an external consistency check. Self-citations [22,23] are used for the alternative external-force formulation and for the chosen normalization of the Lugiato-Lefever equation; they do not supply the pair-interaction result. Hence there is no load-bearing self-citation chain and no reduction of the prediction to its inputs.

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

The ledger is lean: no free parameters are fitted, because the coefficients b, b±, B, and β are defined as integrals over the single-soliton waveform rather than tuned to match the two-soliton simulation. No new physical entities are introduced; the only dynamical variables are the soliton positions, inherited from translation symmetry. The assumptions listed are the standard modeling and asymptotic premises that the derivation explicitly depends on.

assumptions (5)
  • domain assumption The basic Lugiato-Lefever equation (1) is the correct model for pumped Kerr cavity solitons.
    The entire theory is derived from this model, so any model error propagates into the effective dynamics (Section II.A).
  • domain assumption A stable, even single-soliton solution ψs exists, unique up to translations, with a simple zero eigenvalue of the linearization L.
    Required for the projection onto the adjoint zero mode; stability is assumed rather than proven in this paper (Section II.A).
  • domain assumption Soliton tails obey the exponential asymptotics (3)/(4) with exponent σ from Eq. (5), distinguishing monotone and oscillatory decay.
    The classification of interactions into repulsive vs oscillatory rests entirely on this linearized tail analysis (Section II.A).
  • ad hoc to paper The two-soliton ansatz (10) with gauge conditions (11) is valid with Φ=O(ε), ∂Φ/∂t=O(ε²) and dξ_i/dt=O(ε).
    This is the central asymptotic scaling assumption; if the residual Φ is not small or not localized the leading-order projection (17) is incomplete (Section II.B).
  • standard math Terms linear in Φ projected onto the adjoint zero mode vanish and O(ε²) terms can be neglected.
    Uses the adjoint zero-mode property ⟨Ψ̄ξ, LΦ⟩=0 and order counting; the cancellation is asserted rather than fully exhibited (Section II.B, item 2).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Dynamics of interacting cavity solitons." pith.science (2026). https://pith.science/paper/2H2B5NGY

@misc{pith2026250707851,
  author       = {Pith},
  title        = {Pith review of: Dynamics of interacting cavity solitons},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2H2B5NGY}},
  note         = {Machine review of arXiv:2507.07851}
}
read the original abstract

We derive the equations governing the motion of Kerr solitons in pair waveforms. Recent experiments in microresonators have studied a variety of interaction effects in multisoliton waveforms, including collisions and formation of soliton molecules and crystals. Here we analyze the effective interaction that arises from the coupling of soliton-tail overlap nonlinearity with global soliton variables associated with the breaking of translation symmetry. The interaction is either purely repulsive, or alternates between attraction and repulsion, according to whether the decay of soliton tails is monotone or oscillatory. In the latter case, stable fixed points of the effective dynamical system signify stable soliton molecule configuration, but the exponential weakening of the interaction with increasing inter-soliton separation may prevent the molecule from forming in experimentally accessible time scales. Our theory becomes asymptotically exact in the large-separation limit, and we verify the theoretical calculations using soliton trajectories extracted from direct numerical solutions of the wave equation.

Figures

Figures reproduced from arXiv: 2507.07851 by the authors.

Figure 1
Figure 1. FIG. 1. Kerr soliton drift velocity [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. shows two examples of the soliton drift veloc￾ity in a pair waveform with oscillatory tails, comparing the asymptotic theory of (24) with calculations based on numerical solutions of the wave equation (1). In one of the examples the solitons converge toward a stable stationary-molecule steady state, and in the second ex￾ample the leading fixed point of the effective dynamics is unstable. We note that in the red-detu… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 33 canonical work pages

  1. [29]

    N. N. Akhmediev, A. Ankiewicz, and J. M. Soto-Crespo, Multisoliton Solutions of the Complex Ginzburg-Landau Equation, Physical Review Letters 79, 4047 (1997)

    work page 1997

  2. [1]

    O(1) terms Ψ2 s1Ψ∗ s1, Ψ2 s2Ψ∗ s2, and Ψ2 cΨ∗ c , that cancel in combination with other O(1) terms from N [Ψ] because N [Ψs] = N [Ψc] = 0

    work page

  3. [2]

    These terms are of order ε but cancel in (15) because ⟨ ¯Ψξ1, L1Φ⟩ = ⟨L† 1 ¯Ψξ1, Φ⟩ = 0

    Terms that are linear in Φ and involve no factors of ˜Ψs2 combine with other terms of N [Ψ] that are lin- ear in Φ to form L1Φ where L1 is the linearization of N at Ψs1. These terms are of order ε but cancel in (15) because ⟨ ¯Ψξ1, L1Φ⟩ = ⟨L† 1 ¯Ψξ1, Φ⟩ = 0

    work page

  4. [3]

    Terms that are linear in Φ and involve factors of ˜Ψs2 are also of O(ε) but are suppressed by an O(ε) factor by the inner product with ¯Ψξ1 and are there- fore negligible

    work page

  5. [4]

    Terms that are quadratic or cubic in Φ are of O(ε2) and therefore also negligible. Hence, the only nonnegligible terms that remain in the expansion of N in the right-hand side of (15) are cross- terms that involve at least one factor each of ˜Ψs1 and ˜Ψs2 and are independent of Φ. Of these, only the ones which are linear in ˜Ψs2 are of leading order in ε ...

    work page

  6. [5]

    T. Herr, V. Brasch, J. D. Jost, C. Y. Wang, N. M. Kon- dratiev, M. L. Gorodetsky, and T. J. Kippenberg, Tempo- ral solitons in optical microresonators, Nature Photonics 8, 145 (2014)

    2014

  7. [6]

    H. Guo, M. Karpov, E. Lucas, A. Kordts, M. H. P. Pfeiffer, V. Brasch, G. Lihachev, V. E. Lobanov, M. L. Gorodetsky, and T. J. Kippenberg, Universal dynamics and deterministic switching of dissipative Kerr solitons in optical microresonators, Nature Physics 13, 94 (2017)

    work page 2017

  8. [7]

    Yi, Q.-F

    X. Yi, Q.-F. Yang, K. Y. Yang, and K. Vahala, Imaging soliton dynamics in optical microcavities, Nature Com- munications 9, 3565 (2018)

    work page 2018

Show all 34 references
  1. [8]

    Y. Wang, F. Leo, J. Fatome, M. Erkintalo, S. G. Mur- doch, and S. Coen, Universal mechanism for the binding of temporal cavity solitons, Optica 4, 855 (2017)

    2017

  2. [9]

    W. Weng, R. Bouchand, E. Lucas, E. Obrzud, T. Herr, and T. J. Kippenberg, Heteronuclear soliton molecules in optical microresonators, Nature Communications 11, 2402 (2020)

    2020

  3. [10]

    D. C. Cole, E. S. Lamb, P. Del’Haye, S. A. Diddams, and S. B. Papp, Soliton crystals in Kerr resonators, Nature Photonics 11, 671 (2017)

    2017

  4. [11]

    Karpov, M

    M. Karpov, M. H. P. Pfeiffer, H. Guo, W. Weng, J. Liu, and T. J. Kippenberg, Dynamics of soliton crystals in optical microresonators, Nature Physics 15, 1071 (2019)

    2019

  5. [12]

    Y. He, J. Ling, M. Li, and Q. Lin, Perfect Soliton Crys- tals on Demand, Laser & Photonics Reviews 14, 1900339 (2020)

    2020

  6. [13]

    Lu, H.-J

    Z. Lu, H.-J. Chen, W. Wang, L. Yao, Y. Wang, Y. Yu, B. E. Little, S. T. Chu, Q. Gong, W. Zhao, X. Yi, Y.-F. Xiao, and W. Zhang, Synthesized soliton crystals, Nature Communications 12, 3179 (2021)

    2021

  7. [14]

    F. Hu, A. K. Vinod, W. Wang, H.-H. Chin, J. F. McMil- lan, Z. Zhan, Y. Meng, M. Gong, and C. W. Wong, Spatio-temporal breather dynamics in microcomb soliton crystals, Light: Science & Applications 13, 251 (2024)

    2024

  8. [15]

    Shandilya, S.-C

    P. Shandilya, S.-C. Ou, J. Stone, C. Menyuk, M. Erkin- talo, K. Srinivasan, and G. Moille, All-optical azimuthal trapping of dissipative Kerr multi-solitons for relative noise suppression, APL Photonics 10, 016104 (2025)

    2025

  9. [16]

    V. I. Karpman, A soliton system subject to perturbation. Oscillatory shock waves, Soviet Journal of Experimental and Theoretical Physics 50, 58 (1979)

    1979

  10. [17]

    V. I. Karpman and V. V. Solov’ev, A perturbational ap- proach to the two-soliton systems, Physica D: Nonlinear Phenomena 3, 487 (1981)

    1981

  11. [18]

    Gorshkov and L

    K. Gorshkov and L. Ostrovsky, Interactions of solitons in nonintegrable systems: Direct perturbation method and applications, Physica D: Nonlinear Phenomena 3, 428 (1981)

    1981

  12. [19]

    Kawasaki and T

    K. Kawasaki and T. Ohta, Kink dynamics in one- dimensional nonlinear systems, Physica A: Statistical Mechanics and its Applications 116, 573 (1982)

    1982

  13. [20]

    Carr and R

    J. Carr and R. L. Pego, , Communications on Pure and Applied Mathematics 42, 523 (1989)

    1989

  14. [21]

    Aranson, K

    I. Aranson, K. Gorshkov, A. Lomov, and M. Rabinovich, Stable particle-like solutions of multidimensional non- linear fields, Physica D: Nonlinear Phenomena 43, 435 (1990)

    1990

  15. [22]

    Ei and T

    S.-I. Ei and T. Ohta, Equation of motion for interacting pulses, Physical Review E 50, 4672 (1994)

    1994

  16. [23]

    Shin-Ichiro Ei, The Motion of Weakly Interacting Pulses in Reaction-Diffusion Systems, Journal of Dynamics and Differential Equations 14, 85 (2002)

    2002

  17. [24]

    A. G. Vladimirov, G. V. Khodova, and N. N. Rosanov, Stable bound states of one-dimensional autosolitons in a bistable laser, Physical Review E 63, 056607 (2001)

    2001

  18. [25]

    Turaev, A

    D. Turaev, A. G. Vladimirov, and S. Zelik, Chaotic bound state of localized structures in the complex Ginzburg-Landau equation, Physical Review E 75, 045601 (2007)

    2007

  19. [26]

    J. P. Mizrahi, D. Zilberg, and O. Gat, Universal dynam- ics of spatiotemporal entrainment with phase symmetry, Physical Review E 108, 014120 (2023)

    2023

  20. [27]

    J. P. Mizrahi, L. Courtright, P. Shandilya, C. R. Menyuk, and O. Gat, Soliton synchronization in microresonators with a modulated pump, Physical Review E 109, 064204 (2024)

    2024

  21. [28]

    B. A. Malomed, Bound solitons in the nonlinear Schr¨ odinger–Ginzburg-Landau equation, Physical Re- view A 44, 6954 (1991)

    1991

  22. [30]

    Y. K. Chembo and C. R. Menyuk, Spatiotemporal Lugiato-Lefever formalism for Kerr-comb generation in whispering-gallery-mode resonators, Physical Review A 87, 053852 (2013)

    2013

  23. [31]

    I. V. Barashenkov and Yu. S. Smirnov, Existence and stability chart for the ac-driven, damped nonlinear Schrodinger solitons, Physical Review E 54, 5707 (1996)

    1996

  24. [32]

    Parra-Rivas, D

    P. Parra-Rivas, D. Gomila, P. Colet, and L. Gelens, In- teraction of solitons and the formation of bound states in the generalized Lugiato-Lefever equation, The European Physical Journal D 71, 198 (2017)

    2017

  25. [33]

    A. G. Vladimirov, M. Tlidi, and M. Taki, Dissipative soli- ton interaction in Kerr resonators with high-order disper- sion, Physical Review A 103, 063505 (2021)

    2021

  26. [34]

    Parra-Rivas, E

    P. Parra-Rivas, E. Knobloch, L. Gelens, and D. Gomila, Origin, bifurcation structure and stability of localized states in Kerr dispersive optical cavities, IMA Journal of Applied Mathematics 86, 856 (2021)

    2021

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.