Deformations of Polyhedra and Polygons by the Unitary Group

TL;DR

This paper explores the geometric and quantum properties of convex polyhedra and polygons using unitary group actions, providing new mathematical frameworks for their deformation and quantization relevant to quantum gravity.

Contribution

It introduces a U(N) action on the phase space of framed polyhedra, linking geometric deformations to matrix models and extending to quantum intertwiners, with potential applications in quantum gravity.

Findings

U(N) acts transitively on the space of framed polyhedra with fixed total area.

A polynomial integral representation of geometrical averages over polyhedra is established.

Quantum states of polyhedra correspond to irreducible U(N) representations, with coherent states linking classical and quantum geometries.

Abstract

We introduce the set of framed convex polyhedra with N faces as the symplectic quotient C^2N//SU(2). A framed polyhedron is then parametrized by N spinors living in C^2 satisfying suitable closure constraints and defines a usual convex polyhedron plus a phase for each face. We show that there is an action of the unitary group U(N) on this phase space, which changes the shape of faces and allows to map any polyhedron onto any other with the same total area. This realizes the isomorphism of the space of framed polyhedra with the Grassmannian space U(N)/SU(2)*U(N-2). We show how to write averages and correlations of geometrical observables over the ensemble of polyhedra as polynomial integrals over U(N) and we use the Itzykson-Zuber formula from matrix models as the generating function for them. In the quantum case, a canonical quantization of the framed polyhedron phase space leads to the Hilbert space of SU(2) intertwiners. The individual face areas are quantized as half-integers (spins) and the Hilbert spaces for fixed total area form irreducible representations of U(N). We define coherent intertwiner states peaked on classical framed polyhedra and transforming consistently under U(N) transformations. The U(N) character formula can be considered as an extension of the Itzykson-Zuber to the quantum level and generates the traces of all polynomial observables over the Hilbert space of intertwiners. We finally show that classical (convex) polygons can be described in a similar fashion trading U(N) for O(N). We conclude with a discussion of the possible (deformation) dynamics that one can define on the space of polygons or polyhedra. This work is a priori useful in the context of discrete geometry but it should also be relevant to (loop) quantum gravity in 2+1 and 3+1 dimensions when the quantum geometry is defined in terms of gluing of (quantized) polygons and polyhedra.