Comment on: Discontinuous codimension-two bifurcation in a Vlasov equation barre2023 .

Tarcísio N. Teles, Renato Pakter, Yan Levin · 2026

Read Comment on: Discontinuous codimension-two bifurcation in a Vlasov equation barre2023 . by Tarcísio N. Teles, Renato Pakter, Yan Levin (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-22066.txt.

Introduction

Teles and co-authors comment on a 2023 study of phase transitions in a mean field model. That study used linear stability analysis of the Vlasov equation to locate the transition. Teles and his team run molecular dynamics simulations with 100 million particles instead. They find the flagged threshold is not a ferromagnetic transition at all. The system stays paramagnetic there, its magnetization averaging to zero. The true transition comes at much stronger coupling and is discontinuous. They conclude bifurcation analysis cannot predict the long-time fate of such systems.

Catalog record

CREATORTarcísio N. Teles, Renato Pakter, Yan Levin
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE9567
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.22066

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Comment on: Discontinuous codimension-two bifurcation in a Vlasov equation barre2023 . Tarcísio N. Teles, Renato Pakter, Yan Levin Abstract pacs 05.20.-y, 05.70.Ln, 05.70.Fh In their recent work barre2023 , Yamaguchi and Barré (YB) conducted a stability analysis of a class of initially homogeneous solutions to the Vlasov equation for a generalized Hamiltonian mean field (gHMF) model – a system of unit-mass particles confined to a ring and interacting via a 2\pi -periodic pair potential, \phi(\theta)=-[K\cos(\theta)+K_{2}\cos(2\theta)] , where K and K_{2} represent the strengths of the two potential terms, and \theta=\theta_{i}-\theta_{j} denotes the angular separation between two arbitrary particles i and j . The authors investigated the stability of a family of stationary paramagnetic states (in which particles are uniformly distributed between (0,2\pi] , with the one particle distribution function given by F_{\alpha}(\theta,p)=A\exp\left(-\frac{\beta_{2}p^{2}}{2}-\frac{9p^{4}}{4}\right) , where A is the normalization constant satisfying \iint F_{\alpha}(\theta,p)\,d\theta\,dp=1 and \alpha=-\beta_{2}A . These distributions are classified as unimodal for \alpha\leq 0 and bimodal for \alpha>0 . Henceforth referred to as the momentum distribution function, as illustrated in Fig. 2 of barre2023 and reproduced in Fig. 1 below. Figure 1: Momentum distribution functions for the unimodal, flat-top, and bimodal cases, where f_{\alpha}(p)=\int F_{\alpha}(\theta,p)\,d\theta . The distribution differ in terms of their concavity at the origin. Using linear perturbation theory, the authors identified the stability threshold at which the momentum distribution becomes unstable, referring to it as the “critical point”, thus drawing an analogy with continuous equilibrium phase transitions ogawa1 . They state: “…bifurcations have a universal character and tend to provide information about the structure of the phase space, sometimes extending over a rather wide neighborhood of the critical point.” Indeed, the belief that “bifurcation” leads to a phase transition is common in the literature on long-range interacting systems, see for example bachelard . Nevertheless, a claim that the bifurcation analysis alone can provide a meaningful insight into the nature of a quasi-stationary state (qSS) to which the system will evolve, contradicts our recent findings arxiv . To explore this issue in greater depth, we conducted extensive molecular dynamics simulations, which demonstrate that the bifurcation analysis is insufficient to predict either the location or the order of the phase transition between the qSS states. The gHMF model used by YB to justify their assertions was introduced in our earlier work prl ; pre ; report and was more recently analyzed in marciano . The gHMF model combines ferromagnetic and nematic-like interactions, providing a versatile framework for exploring complex phase behaviors and non-equilibrium dynamics. Moreover, it serves as a long-range extension of the generalized XY -model, initially introduced in its short-range form in Lee1985 . While in our previous study prl ; report , the parameter K_{2} was set as K_{2}=1-K , with K varying within the range K\in[0,1] , YB, in their work barre2023 , fixed K_{2}=0.5 and used an arbitrary value of K in conjunction with the parameter \alpha to identify the instability threshold at which the uniform (paramagnetic) particle distribution becomes unstable. Furthermore, unlike our earlier work on the gHMF model, which used the water-bag distribution for initial particle position and velocity distribution, YB studied more general initial velocity distributions presented in Fig. 1. It is well established that for systems with long-range interactions, in the limit of large number of particles, MD simulations become equivalent to the Vlasov evolution on a suitably fine grid Braun:1977 . In this study, we simulate a system comprising N=10^{8} particles, each evolving under Hamilton’s equations, subject to the same interaction potential \phi(\theta) considered by YB. To maintain consistency with the notation used in barre2023 , we enforce the symmetry condition for the initial distribution, so that m_{l,y}=0 for l=1,2 . Furthermore, since K>K_{2}=0.5 , the relevant order parameter is m_{1,x} , which we simply denote as m(t)=\langle\cos[\theta(t)]\rangle , where \langle\cdot\rangle represents the average over all particles. For the unimodal case with \alpha=0 , the results align with our earlier findings for the flat-top water bag (WB) distribution, predicting a discontinuous transition between the paramagnetic and ferromagnetic states. We next focus on the bimodal ( \alpha>0 ) case studied by YB, specifically the one presented in Fig. 7, panel (c) of barre2023 . The theory developed by YB predicts that the transition at the instability threshold is continuous and occurs at K\approx 0.95 . When analyzing the time series for magnetization, we indeed observe that as the threshold is crossed the amplitude of oscillation of the order parameter m(t) starts to increase, however the time average {\overline{m}} remains zero, as is shown in Fig. 2 below. Figure 2: Magnetization before and after the bifurcation point (K_{c}=0.95) identified by YB for the bimodal distribution with \alpha=0.0311643 . In both instances, the time-averaged magnetization is zero – indicating that the system remains in a paramagnetic state above the instability threshold. Therefore, after the instability threshold, the system remains paramagnetic! The instability of the initial distribution does not necessarily imply a phase transition – only that the initial paramagnetic distribution becomes unstable and will oscillate. On the other hand YB incorrectly interpret the oscillating state as indicative of a continuous transition to a ferromagnetic qSS and attempt to use the amplitude of the oscillation as the order parameter to extract a scaling exponent, analogous to the critical exponents in equilibrium systems ogawa2 . In view of our MD simulation results, this interpretation is clearly flawed, since the time-averaged magnetization remains zero both above and below the instability threshold. The real paramagnetic-ferromagnetic transition is observed to be of first order and occurs at K significantly larger than the threshold K_{c} predicted by YB. Indeed, as we increase the parameter K further, we begin to observe a coexistence of two distinct qSS: one ferromagnetic and the other oscillating (paramagnetic), as illustrated in Fig. 3 . Figure 3: For K=1.070 , corresponding to the bimodal distribution with \alpha=0.0311643 , we observe a coexistence of two distinct qSS using particles with exactly the same initial momentum and differing only by a random number exchange in the homogeneous angular distribution. In all cases, the initial magnetization was |m_{0}|<10^{-6} , and the particles evolve using a fourth-order symplectic integrator Yoshida1990 with a time step of dt=0.02 . The energy conservation in the simulation is maintained with a precision accurate to the seventh decimal place. The system evolves to either one of these qSS, starting from initial conditions drawn from exactly the same distribution. This “coexistence” is a clear hallmark of a first order phase transition. As we increase K even further we leave the coexistence region, and observe that all the initial conditions drawn from the same distribution evolve to a ferromagnetic qSS. Figure 4: Schematic representation of the system’s behavior as a function of the parameter K for the bimodal case ( \alpha>0 ). Below the threshold K_{c} , the magnetization exhibits microscopic fluctuations around zero. Beyond the threshold, the system develops oscillations around \overline{m}=0 . For larger K we reach a coexistence region, in which initial conditions drawn from the same distribution evolve either to a paramagnetic or a ferromagnetic state. For still larger K all the initial conditions evolve to a ferromagnetic state. Notably, as \alpha\to 0 , the region characterized by the paramagnetic state with

Read the complete text