Colloquium: Neutrino detectors as tools for nuclear security

For over 40 years, physicists have considered possible uses for neutrino detectors in nuclear nonproliferation, arms control, and fissile materials security. Neutrinos are an attractive fission signature because they readily pass through matter. The same property makes neutrinos challenging to detect in systems that would be practical for nuclear security applications. This Colloquium presents a broad overview of several potential neutrino applications, including the near-field monitoring of known reactors, far-field monitoring of known or discovery of undeclared reactors, detection of reactor waste streams, and detection of nuclear explosions. Recent detector advances have made near-field monitoring feasible, whereas farther-field reactor detection and waste stream detection monitoring may be possible in some cases with further research and development. Very long-range reactor monitoring and nuclear explosion detection do not appear feasible for the foreseeable future due to considerable physical and/or practical constraints.

Figure 1

An example of beta decay chain of fission fragments resulting in the emission of eight neutrinos. Figure courtesy of R. Carr.

Figure 2

Left: Fission fragment yields from the four major nuclides in fission sources according to JEFF 3.3 ([101]). Right: The detection cross section per fission for neutrinos from each of the four fissile isotopes, which is obtained as the product of IBD cross section and the neutrino flux.

Figure 3

(Top) Measurement of fuel burnup at Rovno. The detected reactor antineutrino rate decreases over an observation period of 300 days as production and burning of ${}^{239}\mathrm{Pu}$ reduces the emitted antineutrino flux. From [76]. (Bottom) Observation of reactor start-up at SONGS. The correlated event rate tracks the change in reactor power at start-up, where the events measured at zero power are due to the background. From [31].

Figure 4

On-surface measurement of the reactor operational state by PROSPECT with less than 1 mwe of overburden. Green (gray) shaded periods correspond to full power operation of the host reactor, with the correlated event excess relative to reactor off periods being due to detection of reactor antineutrinos. From [16].

Figure 5

The $1\sigma$ accuracy for the determination of the plutonium content of the reactor as a function of time in the reactor cycle. The data-taking period is 90 days each. Dashed error bars indicate the accuracy from a fit to the plutonium fission rate $f_{\mathrm{Pu}}$, whereas the solid error bars show the result of a fit constrained by a burnup model. The blue (dark) line indicates operation without refueling and the orange (light) line indicates operation with a refueling after 270 days. From [45].