Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Rational disk pages for the radial Reeb flow on a symmetric convex boundary

Open
BirkhoffGlobalSection.reeb_elliptic_disk_page_of_dynamical_convexity

by caleb · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

celestial-mechanicscontact-geometryhamiltonian-dynamics

Let B⊂R4B \subset \mathbb{R}^4B⊂R4 be a compact convex body with 000 in its interior and centrally symmetric boundary S=∂BS = \partial BS=∂B. Let HHH be smooth near SSS, with H=0H = 0H=0, dH≠0dH \ne 0dH=0 and positive tangential Hessian on SSS. Assume the radial contact form λy(v)=−12 ω(y,v)\lambda_y(v) = -\tfrac12\,\omega(y,v)λy​(v)=−21​ω(y,v) satisfies λ(XH)>0\lambda(X_H) > 0λ(XH​)>0 on SSS, and that the Hamiltonian flow of HHH on SSS is dynamically convex (IsDynamicallyConvexOn).

Let ψ\psiψ be a complete flow on SSS with velocity

R=XHλ(XH),R = \frac{X_H}{\lambda(X_H)},R=λ(XH​)XH​​,

the Reeb vector field of λ∣S\lambda|_Sλ∣S​. Let Γ\GammaΓ be a simple periodic orbit of ψ\psiψ with trace KKK, and assume the radial normalization of KKK bounds a positive elliptic rational disk on the round sphere. Then there is a map PPP from the closed unit disk to SSS with the following properties:

  • PPP is smooth and an injective immersion on the closed disk;
  • the interior of PPP is transverse to RRR;
  • P(∂D)=KP(\partial\mathbb{D}) = KP(∂D)=K;
  • P(u)=−P(v)P(u) = -P(v)P(u)=−P(v) exactly when u,vu, vu,v are opposite boundary points, v=−uv = -uv=−u;
  • every ψ\psiψ-trajectory off KKK meets P(int⁡D)∪−P(int⁡D)P(\operatorname{int}\mathbb{D}) \cup -P(\operatorname{int}\mathbb{D})P(intD)∪−P(intD) at arbitrarily large positive and arbitrarily large negative times.

This is the Hryniewicz–Salomão rational open book theorem (Corollary 1.8 with p=2p = 2p=2) in the lifted picture.

  • SSS and λ\lambdaλ are invariant under y↦−yy \mapsto -yy↦−y, so R(−y)=−R(y)R(-y) = -R(y)R(−y)=−R(y) even when HHH is not even. Hence ψ\psiψ descends to the Reeb flow of a contact form on S/{±1}≅RP3S/\{\pm1\} \cong \mathbb{RP}^3S/{±1}≅RP3. Radial projection identifies (S,ker⁡λ)(S, \ker\lambda)(S,kerλ) with the standard tight sphere, conformally.
  • Closed orbits on SSS are the lifts of contractible orbits of the quotient, and the Conley–Zehnder index does not depend on the time parametrization. So the hypothesis is dynamical convexity of the quotient.
  • The disk hypothesis projects to a 222-disk for K/{±1}K/\{\pm1\}K/{±1} with embedded interior. Its characteristic foliation has a single positive elliptic singularity, so K/{±1}K/\{\pm1\}K/{±1} is a 222-unknot with self-linking number −1/2-1/2−1/2.
  • The theorem then gives a rational open book with binding K/{±1}K/\{\pm1\}K/{±1} whose disk-like pages are global surfaces of section. A page lifts to an embedded disk in SSS with boundary KKK; its antipodal image is the other lift. Normalizing the boundary parametrization makes antipodal boundary points have opposite parameters.

Formalization Note. The flow is the Reeb flow of λ∣S\lambda|_Sλ∣S​, as in the source; HHH need not be even. Dynamical convexity and λ(XH)>0\lambda(X_H) > 0λ(XH​)>0 are kept as explicit hypotheses. Both follow from strict convexity: the first by Hofer–Wysocki–Zehnder, the second from convexity with positive tangential Hessian.

Preamble
import Definitions.Def_BirkhoffGlobalSection_TransverseHopf
import Definitions.Def_BirkhoffGlobalSection_DynamicalConvexity
import Definitions.Def_BirkhoffGlobalSection_EllipticRationalDisk
Formal statement
open scoped ContDiff

namespace BirkhoffGlobalSection

theorem reeb_elliptic_disk_page_of_dynamical_convexity
    (B : Set Phase) (hBcompact : IsCompact B) (hBconvex : Convex ℝ B)
    (hB0 : (0 : Phase) ∈ interior B)
    (hsymmetric : ∀ y ∈ frontier B, -y ∈ frontier B)
    (H : Phase → ℝ) (hHsmooth : ∀ y ∈ frontier B, ContDiffAt ℝ ∞ H y)
    (hHzero : ∀ y ∈ frontier B, H y = 0)
    (hHregular : ∀ y ∈ frontier B, fderiv ℝ H y ≠ 0)
    (hHessian : HasPositiveTangentialHessianOn H (frontier B))
    (hcontact : ∀ y ∈ frontier B,
      0 < phaseRadialContact y (hamiltonianVectorField H y))
    (hdc : IsDynamicallyConvexOn H (frontier B))
    (ψ : Flow ℝ {y : Phase // y ∈ frontier B})
    (hψ : ∀ s t, HasDerivAt (fun v => ((ψ v s : {y : Phase // y ∈ frontier B}) : Phase))
      ((phaseRadialContact (ψ t s) (hamiltonianVectorField H (ψ t s)))⁻¹ •
        hamiltonianVectorField H (ψ t s)) t)
    (Γ : PeriodicOrbit ψ) (hsimple : IsSimplePeriodicOrbit ψ Γ)
    (disk : PositiveEllipticRationalDisk (radialNormalize ''
      Set.range (fun t : ℝ => ((ψ t Γ.point : {y : Phase // y ∈ frontier B}) : Phase)))) :
    ∃ page : Plane → Phase,
      (∀ u ∈ closedUnitDisk, page u ∈ frontier B) ∧
      (∀ u ∈ closedUnitDisk, ContDiffAt ℝ ∞ page u) ∧
      Set.InjOn page closedUnitDisk ∧
      (∀ u ∈ closedUnitDisk, Function.Injective (fderiv ℝ page u)) ∧
      (∀ u ∈ openUnitDisk,
        (phaseRadialContact (page u) (hamiltonianVectorField H (page u)))⁻¹ •
            hamiltonianVectorField H (page u) ∉ Set.range (fderiv ℝ page u)) ∧
      page '' unitCircle =
        Set.range (fun t : ℝ => ((ψ t Γ.point : {y : Phase // y ∈ frontier B}) : Phase)) ∧
      (∀ u ∈ closedUnitDisk, ∀ v ∈ closedUnitDisk,
        (page u = -page v ↔ u ∈ unitCircle ∧ v ∈ unitCircle ∧ v = -u)) ∧
      (∀ y : {y : Phase // y ∈ frontier B},
        (y : Phase) ∉ Set.range
          (fun t : ℝ => ((ψ t Γ.point : {y : Phase // y ∈ frontier B}) : Phase)) →
        (∀ R : ℝ, ∃ t : ℝ, R < t ∧ ∃ u ∈ openUnitDisk,
          ((ψ t y : {y : Phase // y ∈ frontier B}) : Phase) = page u ∨
          ((ψ t y : {y : Phase // y ∈ frontier B}) : Phase) = -page u) ∧
        (∀ R : ℝ, ∃ t : ℝ, t < R ∧ ∃ u ∈ openUnitDisk,
          ((ψ t y : {y : Phase // y ∈ frontier B}) : Phase) = page u ∨
          ((ψ t y : {y : Phase // y ∈ frontier B}) : Phase) = -page u)) := by sorry

end BirkhoffGlobalSection
Source
Hryniewicz--Salomao, Elliptic bindings for dynamically convex Reeb flows on the real projective three-space, Calc. Var. PDE 55 (2016), https://arxiv.org/html/1505.02713v3, Corollary 1.8 (via Theorem 1.7) with p = 2, and Section 4; author correction https://www.mathga.rwth-aachen.de/~hryniewicz/files/elliptic_errata.pdf (results unaffected; explains how Corollary 1.8 follows from Theorem 1.7). Lifted to the centrally symmetric hypersurface as in Liu--Salomao, https://arxiv.org/html/2506.17867v2, Section 10.

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me