Scalability of gadolinium-doped-water Cherenkov detectors for nuclear nonproliferation

TL;DR

This paper evaluates the potential of large gadolinium-doped-water Cherenkov detectors for detecting small nuclear reactors at long distances, considering background noise and technical limitations.

Contribution

It provides a detailed analysis of the reactor detection capabilities of large Gd-doped-water Cherenkov detectors, including realistic background models and technical constraints.

Findings

Detection time increases with distance and detector size.

PMT dark rate and reconstruction algorithms limit detector scalability.

50-kiloton detectors are feasible with current technology.

Abstract

Antineutrinos are an unavoidable byproduct of the fission process. The kiloton-scale KamLAND experiment has demonstrated the capability to detect reactor antineutrinos at few-hundred-km range. But to detect or rule out the existence of a single small reactor over many km requires a large detector. So large in fact that the optical opacity of the detection medium itself becomes an important factor. If the detector is so large that photons cannot traverse across the detector medium to an optical detector, then it becomes impractical. For this reason, gadolinium-doped-water Cherenkov detectors have been proposed for large volumes, due to their appealing light-attenuation properties. Even though Cherenkov emission does not produce many photons and the energy resolution is poor, there may be a place for Gd-doped-water detectors in the far-field nuclear reactor monitoring. In this paper, we focus on the reactor discovery potential of large-volume Gd-doped-water Cherenkov detectors for nuclear nonproliferation applications. Realistic background models for the worldwide reactor flux, geo-neutrinos, cosmogenic fast neutrons, and detector-associated backgrounds are included. We calculate the detector run time required to detect a small 50-MWt reactor at a variety of stand-off distances as a function of detector size. We highlight that at present, PMT dark rate and event reconstruction algorithms are the limiting factors to extending beyond ~50-kt fiducial mass.