Bubble Networks: Framed Discrete Geometry for Quantum Gravity

TL;DR

The paper introduces bubble networks as an extended, framed discretization of 3d geometry for canonical quantum gravity, augmenting twisted geometries with a discretized 2d boundary metric encoded in an $\mathrm{SU}(2)\times \mathrm{SL}(2,{\mathbb R})$ phase space. A Casimir balance equation ${\mathcal{D}}=\det D=\ell_{-}\ell_{+}-\ell_{0}^{2}=\vec J^{2}$ ties the intrinsic 2d boundary data to extrinsic curvature, and SL(2,R) matching constraints implement area-preserving diffeomorphisms at interfaces, ensuring consistent boundary geometry across bubbles. The framework shows that, upon symplectic reduction, bubble networks reproduce the $T^*\mathrm{SO}(3)$ phase space of twisted geometries, with twist angles arising from relative plane data between bubble faces; in conformal gauge the construction reduces to the standard spinor (twisted geometry) parametrization. This approach offers a concrete route to quantize extended boundary degrees of freedom, interprets edge states as maximally entangled across interfaces, and paves the way for extended spin networks and holographic explorations in loop quantum gravity.

Abstract

In the context of canonical quantum gravity in 3+1 dimensions, we introduce a new notion of bubble network that represents discrete 3d space geometries. These are natural extensions of twisted geometries, which represent the geometrical data underlying loop quantum geometry and are defined as networks of SU(2) holonomies. In addition to the SU(2) representations encoding the geometrical flux, the bubble network links carry a compatible SL(2,R) representation encoding the discretized frame field which composes the flux. In contrast with twisted geometries, this extra data allows to reconstruct the frame compatible with the flux unambiguously. At the classical level this data represents a network of 3d geometrical cells glued together. The SL(2,R) data contains information about the discretized 2d metrics of the interfaces between 3d cells and SL(2,R) local transformations are understood as the group of area-preserving diffeomorphisms. We further show that the natural gluing condition with respect to this extended group structure ensures that the intrinsic 2d geometry of a boundary surface is the same from the viewpoint of the two cells sharing it. At the quantum level this gluing corresponds to a maximal entanglement along the network edges. We emphasize that the nature of this extension of twisted geometries is compatible with the general analysis of gauge theories that predicts edge mode degrees of freedom at the interface of subsystems.

Figures (3)

• Figure 1: Example of a bubble network, experimentally realized with soap bubbles: bubbles fill up the 3d space and are glued along surface patches.
• Figure 2: Gluing of two bubbles imposing the matching of the 2d geometry of the corresponding surface patches through the symplectic constraints $D_{e}^s=D_{e}^t$, equating the norms $|\vec{e}_{1}^{\,s}|=|\vec{e}_{1}^{\,t}|$, $|\vec{e}_{2}^{\,s}|=|\vec{e}_{2}^{\,t}|$ and the scalar product $\vec{e}_{1}^{\,s}\cdot \vec{e}_{2}^{\,s}=\vec{e}_{1}^{\,t}\cdot \vec{e}_{2}^{\,t}$, resulting in the existence of a unique $\mathrm{SO}(3)$ transport between the two bubbles given by the group element $h_{e}$.
• Figure 3: In the bubble network framework, a bubble with discretized 2d geometry has the topology of a 3-ball with a 2-sphere boundary and is represented as a (not necessarily convex) polyhedron. From the perspective of twisted geometries, a bubble is the blown-up version of a graph node. Then the graph edges (here in blue) link each surface patch on the bubble's surface to a neighboring bubble. These edges carry the symplectic matching constraints for the 2d geometry and the $\mathrm{SO}(3)$ transport from bubbles to bubbles.

Theorems & Definitions (4)

• Proposition 2.1
• proof
• Lemma A.1
• proof