Noise-driven pseudovorticity multipoles in self-focusing beams with quintic saturation

Chengbo Zhang; Xiaohui Gao

REVIEW 3 major objections 4 minor 22 references

Noise-driven pseudovorticity multipoles in self-focusing beams with quintic saturation

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read The paper shows that in a cubic-quintic nonlinear medium, noise-induced asymmetries are trapped inside a breathing soliton rather than radiated away, producing sustained oscillating pseudovorticity multipoles.

desk verdict New numerical observation of noise-driven pseudovorticity multipoles in a cubic-quintic beam, with a plausible but unproven trapping mechanism; worth a referee but needs convergence and boundary checks. read the letter →

arxiv 2608.06045 v1 pith:RZ2DOOQW submitted 2026-08-06 physics.optics nlin.PS

Chengbo Zhang , Xiaohui Gao This is my paper · ORCID

classification physics.opticsnlin.PS

keywords pseudovorticitycubic-quinticnonlinearSchrödingerequationself-focusingcollapsearrestopticalsolitontorquenoiseengineeringmultipoledecomposition

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 what happens to the rotational structure of an optical beam when the beam is noisy and the medium has a saturating nonlinearity. It establishes that in the cubic-quintic nonlinear Schrödinger equation, the quintic term arrests self-focusing collapse and traps amplitude and phase noise inside the resulting breathing soliton, instead of allowing the noise to be expelled as it is in the pure cubic case. As a result, the pseudovorticity — the curl of the optical momentum flux, which measures local rotational flow without phase singularities — develops a persistent multipolar structure that oscillates at half the soliton breathing period. The quadrupole component generally dominates over the dipole except near oscillation troughs, and this behavior is reproducible across random noise seeds and correlation lengths. If correct, the result suggests that input noise statistics can be engineered to control local optical torque in saturable media.

What carries the argument

The load-bearing object is the pseudovorticity ω = ∇⊥ × j = ∇⊥I × ∇⊥φ, the curl of the optical momentum density j = I∇⊥φ. It is zero for a cylindrically symmetric field and becomes nonzero whenever intensity and phase gradients are non-collinear, so it acts as a direct diagnostic of local rotational flow in the absence of phase singularities. The argument is carried by the cubic-quintic nonlinear Schrödinger equation, where the quintic term (−ε|ψ|^4ψ with ε=$10^{-2}$) saturates the Kerr nonlinearity; this arrest of collapse converts the noise-seeded asymmetries into bound internal modes of the soliton. The angular Fourier decomposition of intensity, phase factor, and pseudovorticity into m=1 and m=2 harmonics provides the quantitative measure that distinguishes dipole and quadrupole dynamics.

What would settle it

Repeat the same simulation at higher resolution, for example a 2048×2048 grid with Δz=0.0005, and compare the time series of the maximum pseudovorticity and the modal coefficients I₂, Φ₂, and Ω₂; if the oscillations' period or amplitude changes materially, or if the multipoles decay instead of persisting, the central claim fails. A complementary experiment would measure pseudovorticity in a photorefractive or doped-glass sample with controlled noise and look for the predicted quadrupolar oscillations at half the breathing period.

Watch

Extended reading notes

Core claim

The central claim is that noise does not merely perturb a self-focusing beam; it seeds an internal rotational structure that the quintic nonlinearity preserves. In the pure cubic model, collapse squeezes the core and the noise asymmetries are radiated away, so the pseudovorticity spike should decay through self-cleaning, although the numerics cannot resolve that decay. With the quintic saturation term, collapse is arrested near z=0.13 and the beam enters a persistent focusing-refocusing cycle; the noise-induced asymmetric intensity and phase components remain trapped, and their interplay produces a pseudovorticity field localized where intensity and phase gradients are non-collinear. Angular Fourier decomposition shows the quadrupole (m=2) mode dominates the dipole (m=1), with the intensity quadrupole real and the pseudovorticity quadrupole imaginary, giving oscillations shifted by π/2. Four independent noise seeds and four correlation lengths reproduce the qualitative picture, and phase noise produces larger pseudovorticity with a more prominent dipole channel, especially for long correlation lengths.

Load-bearing premise

The persistence of the trapped pseudovorticity multipoles rests on the numerical resolution of the split-step Fourier scheme (1024×1024 grid, Xmax=20, Δz=0.001); the paper reports no convergence study, and its own cubic comparison cannot resolve the late collapse decay, so an under-resolved quintic run could in principle produce the observed oscillations as grid artifacts.

Editorial extensions

If this is right

In the cubic-quintic model, collapse is arrested and the beam survives as a breathing soliton; noise-induced pseudovorticity persists rather than being radiated away.
The pseudovorticity multipoles oscillate with a period equal to half the focusing-refocusing period, with the quadrupole dominating the dipole except near oscillation troughs.
The qualitative behavior is generic: it appears for both amplitude and phase noise, across correlation lengths from 10Δx to 40Δx, and across independent noise seeds.
Because local optical torque on a small particle is proportional to integrated pseudovorticity, sustained multipole oscillations imply a time-varying, controllable local torque in saturable media.
Long-correlated phase noise preferentially couples into the dipole channel, linking common optical aberrations such as coma and astigmatism to dipole-dominated pseudovorticity.

Reading between the lines

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

Editorial extension: if the trapping mechanism is generic, one testable extension is to seed the beam with controlled low-order aberrations instead of random noise; the framework predicts the corresponding pseudovorticity multipoles should persist and oscillate at half the breathing period.
Editorial extension: the same mechanism may apply beyond optics, for example to Bose-Einstein condensates with saturable three-body losses, where noise-trapped current vorticity could appear in the absence of vortices.
Editorial extension: the paper's cubic-case decay claim rests on self-cleaning that the numerics cannot resolve at late collapse, so a high-resolution or experimental comparison between cubic and cubic-quintic media would directly test the trapping narrative.
Editorial extension: a practical consequence the authors leave implicit is that tuning the noise correlation length could select the dominant multipole, providing a control knob for the direction and frequency of optical torque on trapped particles.

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.

Desk Editor's Note

read the letter

Honest take: this paper has one genuinely new result and one unresolved numerical worry. The new result is that amplitude and phase noise in a self-focusing Gaussian beam with quintic saturation produce a persistent, mostly quadrupolar pseudovorticity pattern that oscillates at half the breathing period. That is not in the papers they cite—pseudovorticity in solitons [18], self-cleaning [10-13], filamentation [14-16]—and the angular harmonic decomposition (Eqs. 9-13) is a clean way to quantify it. Multiple noise seeds and correlation lengths show the qualitative behavior is generic. Credit where due: this is a legitimate, incremental numerical contribution, and the paper is clearly written.

The soft spot is the persistence claim. The evidence is entirely from one split-step Fourier configuration: 1024^2 grid, Xmax=20, dz=0.001, periodic boundary conditions. The paper reports no convergence test (grid, step, window), no larger-window or absorbing-boundary run, and no radiation diagnostic. With a 40-unit-wide periodic domain, high-transverse-k radiation shed during collapse arrest can wrap around and re-enter the core on z-scales comparable to the reported ~0.2 oscillation period. So 'noise asymmetries are not radiated away but persist as internal modes' is not actually distinguished from boundary recirculation. The paper's own caveat that the pure-cubic comparison cannot resolve late-collapse decay (Sec. III) makes the contrast weaker than it appears.

Also minor: the modal integration radius Rmax=2 is not justified; the conclusion says 'quantitative links' when all they show is qualitative trends; no code or data is shipped. None of this is fatal—the observation could be right—but as it stands the central claim is conditional.

Who is this for? People working on saturable nonlinear media, optical self-cleaning, and torque control will want to read it. I would send it to a serious referee, with instructions to ask for a convergence study and a boundary test before acceptance. I wouldn't cite it yet as evidence for trapped internal modes, but I might cite it as the first numerical report of noise-induced pseudovorticity multipoles. Bring it to reading group if you want a discussion of periodic-boundary artifacts in split-step simulations.

Referee Report

3 major / 4 minor

Summary. This manuscript studies the generation and persistence of pseudovorticity in a (2+1)-dimensional cubic-quintic nonlinear Schrödinger equation. It defines pseudovorticity as the curl of the optical momentum flux, launches a noisy Gaussian beam with amplitude and phase noise, and integrates the equation with a split-step Fourier method. The authors report that quintic saturation arrests collapse and that noise-induced asymmetries produce angularly structured pseudovorticity, with quadrupole modes dominating and oscillating at half the beam breathing period; they contrast this with the pure cubic case, in which they argue noise is radiated away. The central claim is that noise-induced asymmetries are not radiated away but persist as internal modes, giving rise to sustained periodic oscillations of the pseudovorticity multipoles.

Significance. If the persistence claim is correct, the paper would provide a simple and potentially falsifiable pathway from input noise statistics to controllable local optical torque in saturable media, and it would sharpen the distinction between self-cleaning and noise trapping in collapsing beams. The numerical exploration is commendably broad: four independent noise seeds, four correlation lengths for both amplitude and phase noise, and a clean angular harmonic decomposition are presented. The paper does not rely on fitted parameters to produce the central effect, and the pseudovorticity signal is an output rather than an input. The main weakness is that the key persistence claim currently rests on a single numerical configuration with no convergence study and with boundary conditions that cannot distinguish trapped internal modes from recycled radiation.

major comments (3)

[Sec. II and Sec. IV] The central assertion that 'noise-induced asymmetries are not radiated away but persist as internal modes' is not yet established because the split-step Fourier implementation with a 1024×1024 grid, Xmax=20, and dz=0.001 imposes periodic boundary conditions and is used without a convergence study. High-transverse-k components generated during collapse arrest can traverse the 40-unit-wide periodic domain and re-enter the core on z-scales comparable to the reported ~0.2 oscillation period, so the sustained oscillations of max|ω| and of I1, I2, Ω1, Ω2 in Figs. 4–7 could be boundary recirculation rather than physically trapped modes. I ask for (i) a grid/step convergence study, (ii) a larger-window run or an absorbing boundary layer, and (iii) a diagnostic that separates radiation from bound modes, such as the flux of the momentum density through a fixed radius or the time-resolved high-k spectral content.
[Sec. III, Eqs. (11)–(13)] The modal integrals are truncated at Rmax=2, but the low-intensity background expands monotonically with propagation (Fig. 2(a)) on a domain with Xmax=20. This choice may exclude precisely the radiated component and thereby bias the conclusion that the multipoles persist inside the soliton. Please report the dependence of Im, Φm, and Ωm on Rmax (for example Rmax=2, 4, 10, 20) and justify the chosen value, or present radial profiles of the modal amplitudes showing that the integrands have decayed by Rmax=2.
[Sec. III, Fig. 2(c)] The paper states that the pure-cubic case cannot resolve the late-stage collapse decay, yet the claimed contrast between 'cubic radiates the noise away' and 'quintic traps the noise' is based on that comparison. Without a resolved cubic run (e.g., adaptive mesh refinement or an analytically known decay law) or at least a clear statement that the cubic comparison is only qualitative and not resolved, the interpretation goes beyond what the simulation supports. This does not invalidate the quintic-saturation result, but the contrast should be presented as unresolved rather than as a demonstrated mechanism.

minor comments (4)

[Sec. II, Eq. (6)] Equation (6) is not equivalent to Eq. (4). For ψ=A(x)e^{iky}, Eq. (4) gives ω=2k A dA/dx, whereas Eq. (6) gives k A dA/dx. If Eq. (6) is used in the code, the pseudovorticity magnitude is off by a factor of two; if Eq. (4) is used, please correct or delete Eq. (6).
[Sec. III, Figs. 5–7] The four-seed and correlation-length scans are shown as individual curves without ensemble statistics. Adding mean ± standard deviation over the four seeds would make statements such as 'the quadrupole generally dominates the dipole' quantitative and would strengthen the robustness claim.
[Sec. III, Eq. (12)] The quantity Φm is a coefficient of e^{iφ}, not of the phase φ itself; this should be stated explicitly in the text next to Eq. (12) rather than only implicitly, to avoid confusion with a Fourier coefficient of the unwrapped phase.
[General] There are minor language issues, including 'The field are rescaled' in Sec. II and the awkward phrase 'Here representative examples ... are displayed'; a brief editing pass is needed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the pseudovorticity multipoles are computed from the simulated fields rather than built into the model or fitted.

full rationale

The paper's central claim, that quintic saturation arrests collapse and traps noise-induced asymmetries as persisting internal modes, is an output of the split-step Fourier simulation initialized with Gaussian-plus-noise fields (Eqs. (1), (7), (8)); no parameter is fitted to reproduce the pseudovorticity multipoles, and the multipole coefficients in Eqs. (11)-(13) are post-processing diagnostics of the simulated intensity, phase, and pseudovorticity. The same-author citation [10] is used only to frame two known noise regimes (self-cleaning versus filamentation) and is not the evidence for the quintic trapping result, which is demonstrated across four noise seeds (Fig. 5) and several correlation lengths (Figs. 6-7). The admitted inability to resolve pure-cubic late-collapse decay (Sec. III) is a numerical-resolution limitation, not a circularity: the cubic case is a benchmark, not an input from which the quintic result is derived. The absence of convergence tests and the periodic-boundary nature of the Fourier method are correctness risks, not circularity. Therefore no step reduces, by the paper's own equations or by self-citation, to its inputs.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The central numerical result rests on the chosen cubic-quintic model, the pseudovorticity-torque correspondence, the thermal noise model, and an unverified numerical resolution. No new entities are introduced.

free parameters (4)

quintic saturation coefficient epsilon = 0.01
Chosen by hand to give a soliton of moderate transverse extent and good spatial resolution (Section II). Collapse arrest, breathing period, and trapped-mode dynamics depend on this value, and no scan over epsilon is reported.
initial beam amplitude A0 = 6 (input power approximately 3.3 Pcr)
Sets the beam power relative to the critical power for self-focusing. The focusing dynamics and the existence of a soliton depend on this choice (Section II).
noise RMS levels and correlation length = amplitude RMS 0.02, phase RMS 0.1 rad, sigma = 10 to 40 grid steps
Noise levels and correlation lengths are chosen as representative laboratory values (Sections II and III). The reported oscillation amplitudes and mode dominance change with these inputs, although the qualitative behavior is said to be robust.
modal integration radius Rmax = 2 (in units of initial 1/e field radius)
Global multipole coefficients are integrated only out to Rmax=2 (Section III). The truncation could bias comparisons between dipole and quadrupole amplitudes, and its effect is not tested.

assumptions (4)

domain assumption The (2+1)D cubic-quintic NLSE with epsilon=1e-2 describes relevant saturable-media self-focusing.
The equation is invoked as a phenomenological model for doped glasses and photorefractive crystals (Section II). Its validity for the claimed noise-trapping regime is assumed, not derived.
domain assumption Within the paraxial scalar approximation, local optical torque on a small absorbing particle is proportional to integrated pseudovorticity.
This links the pseudovorticity diagnostic to the claimed application. It is asserted in Section II and relies on cited work [19] without derivation.
domain assumption Gaussian-filtered white noise with specified RMS and correlation length is a faithful model of physical amplitude and phase noise.
The noise model (Eqs. 7 and 8) is chosen as representative; no comparison against measured noise spectra is provided (Section II).
ad hoc to paper The split-step Fourier discretization (1024x1024, dz=0.001, Xmax=20) resolves collapse arrest and soliton breathing with negligible numerical error.
No convergence study is reported, and the paper explicitly admits resolution limits for the cubic collapse case (Section III), leaving the cubic-quintic case unverified.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Noise-driven pseudovorticity multipoles in self-focusing beams with quintic saturation." pith.science (2026). https://pith.science/paper/RZ2DOOQW

@misc{pith2026260806045,
  author       = {Pith},
  title        = {Pith review of: Noise-driven pseudovorticity multipoles in self-focusing beams with quintic saturation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RZ2DOOQW}},
  note         = {Machine review of arXiv:2608.06045}
}

read the original abstract

We investigate pseudovorticity generation in Gaussian beams undergoing self-focusing under amplitude and phase noise, using the cubic-quintic nonlinear Schr\"odinger equation. Pseudovorticity, defined as the curl of the optical momentum flux, characterizes local rotational flow in the absence of phase singularities. Our numerical simulations show that thermal amplitude and phase noise induce a multipolar pseudovorticity pattern. Unlike the pure cubic case, where noise asymmetries are radiated away during collapse, the quintic saturation arrests collapse and traps the noise in the resulting soliton. Hence, pseudovorticity multipoles persist, oscillating at the focusing-refocusing period. These results suggest a potential pathway for controlling local optical torque through noise engineering.

Figures

Figures reproduced from arXiv: 2608.06045 by the authors.

**Figure 3.** shows spatial snapshots at three characteristic propagation distances: the first focus (z = 0.13), a middle position (z = 0.283), and the second focus (z = 0.37). The intensity and phase are decomposed into their symmetric (angularly averaged) and asymmetric parts via FIG. 3. Snapshots at focus (z = 0.13), mid-focus (z = 0.283), and second focus (z = 0.37) (rows). Columns: asymmetric intensity (left), asymmetric p… view at source ↗

**Figure 2.** FIG. 2. Amplitude noise results ( [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Figures from the paper (3 more)

**Figure 4.** Figure 4: FIG. 4. (a) Moduli of dipole ( [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

**Figure 6.** Figure 6: FIG. 6. Dipole, quadrupole, and peak pseudovorticity ver [PITH_FULL_IMAGE:figures/full_fig_p004_6.png]

**Figure 7.** Figure 7: FIG. 7. Dipole, quadrupole, and peak pseudovorticity versus [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

22 extracted references · 21 canonical work pages

[18]

Pseudovorticity of 2+1D optical solitons

L. Falsi, G. Agostino, A. Villois, F. Coppini, P. M. Santini, M. Onorato, A. J. Agranat, S. Trillo, and E. DelRe, Pseudovorticity of 2+1D optical solitons (2026), arXiv:2605.03543 [physics.optics]

work page Pith review arXiv 2026
[1]

J. H. Marburger, Self-focusing: Theory, Prog. Quantum Electron.4, 35 (1975)

work page 1975
[2]

Couairon and A

A. Couairon and A. Mysyrowicz, Femtosecond filamen- tation in transparent media, Phys. Rep.441, 47 (2007)

work page 2007
[3]

Fibich,The Nonlinear Schr¨ odinger Equation: Singu- lar Solutions and Optical Collapse, Applied Mathemati- cal Sciences, Vol

G. Fibich,The Nonlinear Schr¨ odinger Equation: Singu- lar Solutions and Optical Collapse, Applied Mathemati- cal Sciences, Vol. 192 (Springer International Publishing, Cham, 2015)

work page 2015
[4]

K. D. Moll, A. L. Gaeta, and G. Fibich, Self-similar op- tical wave collapse: Observation of the townes profile, Phys. Rev. Lett.90, 203902 (2003)

work page 2003
[5]

Fibich and A

G. Fibich and A. L. Gaeta, Critical power for self- focusing in bulk media and in hollow waveguides, Opt. Lett.25, 335 (2000)

work page 2000
[6]

T. D. Grow, A. A. Ishaaya, L. T. Vuong, A. L. Gaeta, N. Gavish, and G. Fibich, Collapse dynamics of super- Gaussian beams, Opt. Express14, 5468 (2006)

work page 2006
[7]

Mihalache, D

D. Mihalache, D. Mazilu, L.-C. Crasovan, B. A. Mal- omed, F. Lederer, and L. Torner, Robust soliton clusters in media with competing cubic and quintic nonlinearities, Phys. Rev. E68, 046612 (2003)

work page 2003

Show all 22 references

[8]

F. K. Abdullaev and J. Garnier, Dynamical stabilization of solitons in cubic-quintic nonlinear schr¨ odinger model, Phys. Rev. E72, 035603(R) (2005)

2005
[9]

Pasquazi, M

A. Pasquazi, M. Peccianti, M. Clerici, C. Conti, and R. Morandotti, Collapse arrest in instantaneous kerr me- dia via parametric interactions, Phys. Rev. Lett.113, 133901 (2014)

2014
[10]

Zhang and X

C. Zhang and X. Gao, Transition from self-cleaning to modulation instability in noisy single filament collapse of Gaussian beams, Acta Optica Sinica (Online)2, 2107001 (2025)

2025
[11]

Prade, M

B. Prade, M. Franco, A. Mysyrowicz, A. Couairon, H. Buersing, B. Eberle, M. Krenz, D. Seiffer, and O. Vasseur, Spatial mode cleaning by femtosecond fil- amentation in air, Opt. Lett.31, 2601 (2006)

2006
[12]

Liu and S

W. Liu and S. L. Chin, Abnormal wavelength dependence of the self-cleaning phenomenon during femtosecond- 6 laser-pulse filamentation, Phys. Rev. A76, 013826 (2007)

2007
[13]

Sagiv, A

A. Sagiv, A. Ditkowski, R. H. Goodman, and G. Fibich, Loss of physical reversibility in reversible systems, Phys- ica D410, 132515 (2020)

2020
[14]

Fibich and B

G. Fibich and B. Ilan, Vectorial and random effects in self-focusing and in multiple filamentation, Physica D 157, 112 (2001)

2001
[15]

Garnier, Statistical analysis of noise-induced multiple filamentation, Phys

J. Garnier, Statistical analysis of noise-induced multiple filamentation, Phys. Rev. E73, 046611 (2006)

2006
[16]

Choudhary, A

S. Choudhary, A. N. Black, A. Antikainen, and R. W. Boyd, Controlling nonlinear rogue-wave formation using the coherence length of phase noise, Phys. Rev. Res.6, 013174 (2024)

2024
[17]

M. V. Berry, Optical currents, J. Opt. A: Pure Appl. Opt. 11, 094001 (2009)

2009
[19]

Roichman, B

Y. Roichman, B. Sun, Y. Roichman, J. Amato-Grill, and D. G. Grier, Optical forces arising from phase gradients, Phys. Rev. Lett.100, 013602 (2008)

2008
[20]

Senthilnathan, Q

K. Senthilnathan, Q. Li, K. Nakkeeran, and P. K. A. Wai, Robust pedestal-free pulse compression in cubic-quintic nonlinear media, Phys. Rev. A78, 033835 (2008)

2008
[21]

Jeng, M.-F

C.-C. Jeng, M.-F. Shih, K. Motzek, and Y. Kivshar, Par- tially incoherent optical vortices in self-focusing nonlin- ear media, Phys. Rev. Lett.92, 043904 (2004)

2004
[22]

DelRe and M

E. DelRe and M. Segev, Self-Focusing and Solitons in Photorefractive Media, inSelf-focusing: Past and Present, Vol. 114, edited by R. W. Boyd, S. G. Lukishova, and Y. Shen (Springer New York, New York, NY, 2009) pp. 547–572, series Title: Topics in Applied Physics

2009

Pith tools

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

[1] [18]

Pseudovorticity of 2+1D optical solitons

L. Falsi, G. Agostino, A. Villois, F. Coppini, P. M. Santini, M. Onorato, A. J. Agranat, S. Trillo, and E. DelRe, Pseudovorticity of 2+1D optical solitons (2026), arXiv:2605.03543 [physics.optics]

work page Pith review arXiv 2026

[2] [1]

J. H. Marburger, Self-focusing: Theory, Prog. Quantum Electron.4, 35 (1975)

work page 1975

[3] [2]

Couairon and A

A. Couairon and A. Mysyrowicz, Femtosecond filamen- tation in transparent media, Phys. Rep.441, 47 (2007)

work page 2007

[4] [3]

Fibich,The Nonlinear Schr¨ odinger Equation: Singu- lar Solutions and Optical Collapse, Applied Mathemati- cal Sciences, Vol

G. Fibich,The Nonlinear Schr¨ odinger Equation: Singu- lar Solutions and Optical Collapse, Applied Mathemati- cal Sciences, Vol. 192 (Springer International Publishing, Cham, 2015)

work page 2015

[5] [4]

K. D. Moll, A. L. Gaeta, and G. Fibich, Self-similar op- tical wave collapse: Observation of the townes profile, Phys. Rev. Lett.90, 203902 (2003)

work page 2003

[6] [5]

Fibich and A

G. Fibich and A. L. Gaeta, Critical power for self- focusing in bulk media and in hollow waveguides, Opt. Lett.25, 335 (2000)

work page 2000

[7] [6]

T. D. Grow, A. A. Ishaaya, L. T. Vuong, A. L. Gaeta, N. Gavish, and G. Fibich, Collapse dynamics of super- Gaussian beams, Opt. Express14, 5468 (2006)

work page 2006

[8] [7]

Mihalache, D

D. Mihalache, D. Mazilu, L.-C. Crasovan, B. A. Mal- omed, F. Lederer, and L. Torner, Robust soliton clusters in media with competing cubic and quintic nonlinearities, Phys. Rev. E68, 046612 (2003)

work page 2003