Linear phase space deformations with angular momentum symmetry

Motivated by the work of Leznov–Mostovoy, we classify the linear deformations of standard $2n$-dimensional phase space that preserve the obvious symplectic $\mathfrak{o}(n)$-symmetry. As a consequence, we describe standard phase space, as well as $T^{}S^{n}$ and $T^{}\mathbb{H}^{n}$ with their standard symplectic forms, as degenerations of a 3-dimensional family of coadjoint orbits, which in a generic regime are identified with the Grassmannian of oriented 2-planes in $\mathbb{R}^{n+2}$.

💡 Research Summary

The paper addresses the problem of classifying all linear deformations of the standard 2n‑dimensional phase space that preserve the obvious orthogonal symmetry O(n), often interpreted as the conservation of angular momentum in classical mechanics. The authors begin by observing that the standard symplectic vector space (ℝⁿ⊕ℝⁿ, ω=∑dxᵢ∧dpᵢ) can be identified with a connected component of a coadjoint orbit of the semidirect product group Gₙ = O(n)⋉Hₙ, where Hₙ is the (2n+1)-dimensional Heisenberg group. This identification allows the deformation problem to be reformulated as a problem about deformations of the Lie algebra
 gₙ = 𝔬(n)⋉ℎₙ,
where ℎₙ is the Heisenberg algebra.

To study infinitesimal deformations the authors compute the second Lie‑algebra cohomology H²(gₙ,gₙ) with coefficients in the adjoint module. Using the Hochschild‑Serre spectral sequence for the semidirect product 𝔬(n)⋉ℂⁿ (the complexification of gₙ) they reduce the computation to the 𝔬(n)-invariant part of H²(ℂⁿ,gℂₙ). The vector representation ℂⁿ splits into two invariant subspaces ℂⁿ₁ and ℂⁿ₂, and the space of invariant 2‑cocycles decomposes accordingly. Three explicit, linearly independent invariant 2‑cocycles are exhibited:

• f₁(eᵢ,eⱼ)=ℓ_{ij} (supported on ℂⁿ₁∧ℂⁿ₁),
• f₂(e_{n+i},e_{n+j})=ℓ_{ij} (supported on ℂⁿ₂∧ℂⁿ₂),
• f₃(eᵢ,e_{n+j})=ℓ_{ij} (supported on ℂⁿ₁⊗ℂⁿ₂),

where ℓ_{ij} are the standard generators of 𝔬(n). The authors prove that any 𝔬(n)-invariant 2‑cocycle is a linear combination ε₁f₁+ε₂f₂+ε₃f₃, and that there are no further independent invariant cocycles. Consequently

 dim H²(gₙ,gₙ)=3

for all n≥3 (the cases n=1,2 are excluded from the main proof but the same dimension holds).

Integrating these infinitesimal deformations yields a three‑parameter family of Lie algebras gₙ(ε₁,ε₂,ε₃). In terms of the dual coordinates (xᵢ,pᵢ) on ℝⁿ⊕ℝⁿ and the central element I of the Heisenberg algebra, the deformed Lie–Poisson brackets become

{ xᵢ, xⱼ } = ε₁ ℓ_{ij}, { pᵢ, pⱼ } = ε₂ ℓ_{ij},
{ xᵢ, pⱼ } = δ_{ij} I + ε₃ ℓ_{ij},
{ xᵢ, I } = ε₃ xᵢ − ε₁ pᵢ, { pᵢ, I } = ε₂ xᵢ − ε₃ pᵢ.

These brackets satisfy the Jacobi identity for any (ε₁,ε₂,ε₃)∈ℂ³, so they define genuine Lie algebras. Scaling the generators shows that triples differing by a non‑zero scalar give isomorphic algebras; thus the space of distinct isomorphism classes is naturally stratified by the projective plane ℙ²(ε).

Special values of the parameters reproduce well‑known geometries:

• (ε₁,ε₂,ε₃) = (1,1,0) yields gₙ≅𝔬(n+2,ℂ). Real forms give 𝔬(n+1) (positive curvature) and 𝔬(n,1) (negative curvature), which correspond respectively to the cotangent bundles T* Sⁿ and T* ℍⁿ equipped with their standard symplectic forms.
• (1,0,0) or (0,1,0) give the Euclidean algebra 𝔬(n+1)⋉ℝ^{n+1}, i.e. the phase space of a particle moving on a sphere or hyperbolic space with zero curvature deformation.
• (0,0,1) recovers the original flat phase space (no deformation of the Poisson brackets).

Geometrically, the authors interpret the deformation parameters as a deformation of a degenerate quadratic form Q₀ on ℝ^{n+2} of signature (n,0) and isotropy index 2. Choosing a totally isotropic plane W=Span{v_{n+1},v_{n+2}} and modifying the values (v_{n+1},v_{n+1})=ε₁, (v_{n+2},v_{n+2})=ε₂, (v_{n+1},v_{n+2})=ε₃ produces exactly the Lie algebra gₙ(ε). Thus the space H²(gₙ,gₙ) is identified with the space of equivalence classes of such quadratic‑form deformations. Moreover, the coadjoint orbits of gₙ(ε) can be identified with the oriented Grassmannian Gr⁺₂(ℝ^{n+2}) of 2‑planes in ℝ^{n+2}; the generic orbit (when ε₁ε₂≠0 and ε₃²≠ε₁ε₂) is diffeomorphic to this Grassmannian and carries a natural symplectic structure induced from the Lie–Poisson bracket. In the degenerate limits the orbits collapse to the cotangent bundles of the sphere or hyperbolic space, providing a unified picture of these classical phase spaces as degenerations of a single three‑parameter family.

The paper concludes by emphasizing that the classification furnishes a complete description of all linear phase‑space deformations preserving angular‑momentum symmetry. This has several implications: it clarifies the role of curvature as a deformation parameter in integrable systems, offers a systematic way to construct new integrable models on curved configuration spaces, and suggests a pathway to quantization where the underlying symmetry algebra deforms from 𝔬(n) to 𝔬(n+2). The authors indicate future work on extending the analysis to non‑linear deformations, exploring quantum analogues, and applying the framework to concrete mechanical systems.