Does backreaction enforce the averaged null energy condition in semiclassical gravity?

TL;DR

This paper investigates whether the averaged null energy condition (ANEC) holds in semiclassical gravity, finding that quantum effects generally enforce ANEC except in flat spacetime, implying traversable wormholes are likely only at Planck scales.

Contribution

The study demonstrates that quantum backreaction tends to enforce ANEC in semiclassical gravity, limiting traversable wormholes to Planck-scale sizes, and clarifies issues related to the semiclassical Einstein equation.

Findings

ANEC can be violated locally but is enforced on average with Planck-scale smearing.

Traversable wormholes, if they exist, are likely only at Planck scales.

Quantum backreaction supports the validity of ANEC in semiclassical gravity.

Abstract

The expected stress-energy tensor <T_{ab}> of quantum fields generically violates the local positive energy conditions of general relativity. However, <T_{ab}> may satisfy some nonlocal conditions such as the averaged null energy condition (ANEC), which would rule out traversable wormholes. Although ANEC holds in Minkowski spacetime, it can be violated in curved spacetimes if one is allowed to choose the spacetime and quantum state arbitrarily, without imposition of the semiclassical Einstein equation G_{ab} = 8 \pi <T_{ab}>. In this paper we investigate whether ANEC holds for solutions to this equation, by studying a free, massless scalar field with arbitrary curvature coupling in perturbation theory to second order about the flat spacetime/vacuum solution. We "reduce the order" of the perturbation equations to eliminate spurious solutions, and consider the limit in which the lengthscales determined by the incoming state are much larger than the Planck length. We also need to assume that incoming classical gravitational radiation does not dominate the first order metric perturbation. We find that although the ANEC integral can be negative, if we average the ANEC integral transverse to the geodesic with a suitable Planck scale smearing function, then a strictly positive result is obtained in all cases except for the flat spacetime/vacuum solution. This result suggests --- in agreement with conclusions drawn by Ford and Roman from entirely independent arguments --- that if traversable wormholes do exist as solutions to the semiclassical equations, they cannot be macroscopic but must be ``Planck scale''. A large portion of our paper is devoted to the analysis of general issues concerning the nature of the semiclassical Einstein equation and of prescriptions for extracting physically relevant solutions.