Deformation Quantisation of Constrained Systems

TL;DR

This paper explores deformation quantisation of constrained Hamiltonian systems, demonstrating how to convert second class constraints into first class, analyzing anomalies, and applying the method to models like Yang-Mills and gravity.

Contribution

It introduces a systematic approach to deformation quantisation of constrained systems, including anomaly analysis and applications to gauge theories and gravity.

Findings

Deformation quantisation can turn second class constraints into first class constraints.

Yang-Mills theory's quantisation is straightforward, gravity exhibits anomalies.

Ashtekar variables simplify the quantum constraint equations for gravity.

Abstract

We study the deformation quantisation (Moyal quantisation) of general constrained Hamiltonian systems. It is shown how second class constraints can be turned into first class quantum constraints. This is illustrated by the O(N) non-linear $\sigma$-model. Some new light is also shed on the Dirac bracket. Furthermore, it is shown how classical constraints not in involution with the classical Hamiltonian, can be turned into quantum constraints {\em in} involution with respect to the Hamiltonian. Conditions on the existence of anomalies are also derived, and it is shown how some kinds of anomalies can be removed. The equations defining the set of physical states are also given. It turns out that the deformation quantisation of pure Yang-Mills theory is straightforward whereas gravity is anomalous. A formal solution to the Yang-Mills quantum constraints is found. In the \small{ADM} formalism of gravity the anomaly is very complicated and the equations picking out physical states become infinite order functional differential equations, whereas the Ashtekar variables remedy both of these problems -- the anomaly becoming simply a central extension (Schwinger term) and the equations for physical states become finite order. We finally elaborate on the underlying geometrical structure and show the method to be compatible with BRST methods.