Gauge-invariant coefficients in perturbative quantum gravity

TL;DR

This paper recalculates heat kernel coefficients in perturbative quantum gravity, confirms their gauge invariance on-shell, and identifies necessary corrections in the worldline approach for arbitrary dimensions.

Contribution

It provides a benchmark for gauge-invariant coefficients and proposes amendments to the worldline quantization method in arbitrary dimensions.

Findings

Recomputed heat kernel coefficients match literature in D=4.

Discovered mismatch in coefficients at D≠4 in the worldline approach.

Proposed fixing the worldline path integral to ensure consistency across dimensions.

Abstract

Heat kernel methods are useful for studying properties of quantum gravity. We recompute here the first three heat kernel coefficients in perturbative quantum gravity with cosmological constant to ascertain which ones are correctly reported in the literature. They correspond to the counterterms needed to renormalize the one-loop effective action in four dimensions. They may be evaluated at arbitrary dimensions $D$, in which case they identify only a subset of the divergences appearing in the effective action for $D\geq 6$. Generically, these coefficients depend on the gauge-fixing choice adopted in quantizing the Einstein-Hilbert action. However, they become gauge-invariant once evaluated on-shell, i.e. using Einstein's field equations with cosmological constant. We identify them and use them as a benchmark for checking alternative approaches to perturbative quantum gravity. One such approach describes the graviton in first-quantization through the use of the action of the ${\cal N}=4$ spinning particle, characterized by four supersymmetries on the worldline and a set of worldline gauge invariances. This description has been used for computing the gauge-invariant coefficients as well. We verify their correctness at $D=4$, but find a mismatch at arbitrary $D$ when comparing with the benchmark fixed earlier. We interpret this result as signaling that the path integral quantization of the ${\cal N}=4$ spinning particle should be amended. We perform this task by fixing the correct counterterm that must be used in the worldline path integral quantization of the ${\cal N}=4$ spinning particle to make it consistent in arbitrary dimensions.