[Paper Review] The Symplectic Camel and Quantum Universal Invariants: the Angel of Geometry vs. the Demon of Algebra

Maurice A. de Gosson|arXiv (Cornell University)|Mar 23, 2012

Quantum chaos and dynamical systems9 references3 citations

TL;DR

This paper introduces a new class of geometric quantum invariants based on the volumes of orthogonal projections of quantum covariance ellipsoids onto symplectic subspaces of phase space. Using Gromov's non-squeezing theorem and symplectic capacity theory, it proves that these projected volumes are bounded from below by $ \frac{h^k}{2^k k!} $, offering a stronger, geometric formulation of the uncertainty principle that surpasses algebraic invariants in conceptual clarity and physical insight.

ABSTRACT

A positive definite symmetric matrix σ qualifies as a quantum mechanical covariance matrix if and only if σ+(1/2)i\hbarΩ\geq0 where Ω is the standard symplectic matrix. This well-known condition is a strong version of the uncertainty principle, which can be reinterpreted in terms of the topological notion of symplectic capacity, closely related to Gromov's non-squeezing theorem. We show that a recent refinement of the latter leads to a new class of geometric invariants. These are the volumes of the orthogonal projections of the covariance ellipsoid on symplectic subspaces of the phase space. We compare these geometric invariants to the algebraic "universal quantum invariants" of Dodonov and Serafini.

Motivation & Objective

• To establish a geometric formulation of the quantum uncertainty principle using symplectic topology and symplectic capacities.
• To introduce new invariants—volumes of projections of covariance ellipsoids onto symplectic subspaces—as fundamental quantum observables.
• To compare these geometric invariants with algebraic 'universal invariants' proposed by Dodonov and Serafini, arguing for the superiority of geometric approaches.
• To demonstrate that the symplectic structure of phase space imposes non-trivial constraints on quantum covariance matrices beyond standard uncertainty relations.

Proposed method

• Utilizes Gromov's non-squeezing theorem and the concept of symplectic capacity to reframe the uncertainty principle as a topological constraint on phase space volumes.
• Applies Williamson's diagonalization theorem to reduce the covariance matrix to a canonical form with symplectic eigenvalues $ \nu_j $, preserving symplectic invariants.
• Derives the volume of the orthogonal projection of the covariance ellipsoid $ W_\sigma $ onto any $ 2k $-dimensional symplectic subspace $ \mathbb{F}_{2k} $ as $ \frac{(2\pi)^k}{k!} \nu_1 \cdots \nu_k $.
• Employs the Abbondandolo–Matveyev inequality to extend the lower bound on projected volumes from linear to nonlinear symplectic transformations.
• Compares the geometric invariants (projected volumes) with algebraic invariants $ \Delta_j^n $, the principal minors of $ \Omega\sigma $, showing that the former contain richer physical information.

Experimental results

Research questions

• RQ1 Can the uncertainty principle be reformulated in terms of symplectic topology and geometric invariants of phase space projections?
• RQ2 Do the volumes of projections of quantum covariance ellipsoids onto symplectic subspaces yield stronger, more physically meaningful invariants than algebraic expressions?
• RQ3 How do geometric invariants based on symplectic capacities compare to the algebraic 'universal invariants' of Dodonov and Serafini?
• RQ4 Is the lower bound $ \frac{h^k}{2^k k!} $ on projected volumes robust under nonlinear symplectic transformations?

Key findings

• The volume of the orthogonal projection of the covariance ellipsoid $ W_\sigma $ onto any $ 2k $-dimensional symplectic subspace $ \mathbb{F}_{2k} $ satisfies $ \operatorname{Vol}(\Pi_{\mathbb{F}_{2k}}W_\sigma) \geq \frac{h^k}{2^k k!} $.
• For $ k=1 $, this yields a geometric uncertainty principle: the area of the projection onto any conjugate pair $ (x_j, p_j) $ is at least $ \frac{1}{2}h $.
• The symplectic eigenvalues $ \nu_j $ of the covariance matrix satisfy $ \nu_j \geq \frac{1}{2}\hbar $, a direct consequence of the strong uncertainty principle $ \sigma + \frac{1}{2}i\hbar\Omega \geq 0 $.
• The geometric invariants (projected volumes) contain more physical information than the algebraic invariants $ \Delta_j^n $, which are sums of products of squared symplectic eigenvalues.
• The lower bound on projected volumes is preserved under symplectic transformations, including nonlinear ones, provided the Abbondandolo–Matveyev inequality holds.
• The paper argues that geometric invariants, rooted in symplectic topology, should be preferred over algebraic invariants in foundational quantum mechanics.