Skip Menu

Return to Skip Menu

Main Content

The CHANDLER Detector Project

Reactor Neutron Detection

Nuclear reactors are a very intense source of electron antineutrinos which come from the β-decay of the neutron-rich fission fragments. Neutrinos are extraordinarily weakly interacting such that they can pass all the way through the earth unhindered. The neutrinos produced in the core arrive at the detector just as they were produced and as such they carry information about their point of origin and parent nucleus. Most of these neutrinos pass through the detector without leaving a trace, but a tiny fraction will interact in the detector, which give us the possibility to learn directly about the inner workings of the core. For each gigawatt of thermal power a reactor produces about 2×1020 electron antineutrinos per second. A one-ton detector placed 15 meters from the core of a typical commercial reactor should see about 5000 neutrino interactions per day. That's enough to measure the daily average power to better that 1.5%. The energy distribution of the neutrinos is the sum of several thousand β-decay branches. There are sufficient differences between the branch activities for the different fissionable isotopes to make it possible to track their evolution in the core. Finally, the direction that the neutrinos are traveling and the rate of neutrinos detected along a given line tell us about the spatial distribution of fission in the core.

Reactor neutrino are detected through a process known as inverse β-decay, in which an electron antineutrino interacts with a proton (a hydrogen nucleus in the plastic scintillator) to produce a positron (or positive electron) and a neutron.

In this process, the positron acquires the energy of the neutrino, minus the mass difference between the proton and the neutron plus positron. Within the detector the positron energy is deposit promptly in the scintillator, producing light proportional to that energy deposition. In this way the energy of each neutrino can measured on an event-by-event basis. The neutron, which carries just a tiny fraction of the available energy, bounces around for a while (about 100 μs) and eventually captures on a nucleus. In the capture process the excess binding energy is release into the scintillator creating a delayed light pulse.

The CHANDLER Technology

The CHANDLER detector technology is comprised of cubes of wavelength shifting plastic scintillator cubes and thin sheets of lithium-6 (6Li) loaded zinc sulfide (ZnS) scintillator. The 6 cm cubes are arraigned in layers of up to 16×16 cubes which are separated by the 6Li-loaded ZnS sheets. The cubes and sheets are well suited for detecting electron antineutrinos from nuclear reactors, which produce a positron and a neutron when they interact in the plastic cubes. The positron produces a prompt flash of light in the cube, while the neutron bounces around for a while before capturing on the 6Li in the sheet producing a delayed flash of light. The correlation between these two distinct events provides a clean indication of a neutrino interaction. The light from both the sheets and cubes is transported by total-internal-reflection along the rows and columns of cubes to the surface of the detector where it is read out by light detectors known as photomultiplier tubes (PMTs). This unique method for reading out the light, known as a Raghavan optical lattice, was invented by the late Center for Neutrino Physics professor Raju Raghavan. It provides precise spatial information for the neutrino interaction and neutron capture, which can be used to separate the true neutrino events, which must be close together in both time and space, from the random correlation of unassociated neutron captures and positron-like events that would otherwise form fake neutrino events.

The following brief video illustrates how this technology works:

The CHANDLER R&D Program

Stages in the R&D Program: Cube String, MicroCHANDLER and MiniCHANDLER

In the early stages of the program we tested the optical properties of the cubes in strings of up to 10 cubes. This allowed us to create an optical model for the light transportation in the cubes and to determine the energy response of the scintillator. With a combination of measurements and simulation we have shown that the full-scale detector should measure the positron energy with a resolution of better than 6.5% at 1 MeV. Next we constructed a prototype detector of 3×3×3 cubes and 4 sheets. This detector, known as MicroCHANDLER has been used to study the neutron response and to measure the background rate from false coincidences. With MicroCHANDLER we obtained a neutron event purity of better than 98%. We have recently completed construction of the next-level prototype, known as MiniCHANDLER, which is constructed of 320 cubes (8×8×5) and 6 sheets. MiniCHANDLER was designed to be large enough to see neutrinos from a nuclear reactor. Once we finish commissioning MiniCHANDLER in the lab it will be transported in our Mobile Neutrino Lab to the North Anna Nuclear Power Plant near Richmond, Virginia for a full system demonstration.

The MiniCHANDLER Detector at different stages in its assembly

Neutron Identification in CHANDLER

In CHANDLER the scintillation light produced in the 6Li-loaded ZnS is released much more slowly than the scintillation light in the plastic cubes (200 ns compared to 10 ns). This means that the signal pulse observed in the PMT that comes from neutron capture is much wider than from positron-like events in formed in the cubes. This fact can be used to form a simple neutron identification (or "Neutron ID") variable by taking the ration of the signal pulse integral to the signal pulse amplitude. Due to their longer light release time, neutron capture pulses produced in the ZnS have larger values of this ratio than signals formed in the plastic scintillating cubes.

The above figure shows this Neutron ID parameter for signals readout simultaneously on the two sides of the detector (the x and y views). There are two clear populations of events: those with high Neutron ID values in both views, these are neutron-like events, and those with low Neutron ID values in both views, which are positron-like events.