On Earth, tens of billions of the sun’s neutrinos pass through an area the size of a thumbtack every second. But most of these particles zip straight through Earth without a single interaction with another bit of matter.
Physicists have been working to make detectors small enough to be installed inside nuclear power plants. If their prototypes are proved, such detectors could continuously monitor nuclear reactors and provide a new way to safeguard against nuclear proliferation.
We show that with a reasonable choice of detector parameters, the test can detect replacement of 73 kg of plutonium in 90 days with 95% probability, while controlling the false positive rate at 5%. We show that some improvement on this level of sensitivity may be expected by various means, including use of the method in conjunction with existing reactor safeguards methods. We also identify a necessary and sufficient daily antineutrino count rate to achieve the quoted sensitivity, and list examples of detectors in which such rates have been attained.
A joint group of physicists based in California at Lawrence Livermore National Laboratory and Sandia National Laboratories, along with researchers at Atomic Energy of Canada Limited’s Chalk River Laboratories, aims to capture reactor-born antineutrinos with a detector they plan to install next year at the Point Lepreau Generating Station, a CANDU-type nuclear reactor in New Brunswick, Canada.
To catch antineutrinos, the detector employs hundreds of liters of organic solvent mixed with gadolinium atoms. Incoming antineutrinos occasionally collide with protons in the mixture, creating a neutron and a positron, the antimatter partner of the electron. The positron creates a flash of light when it meets with and annihilates an electron, and the neutron releases light when it is captured by a gadolinium atom. The detectors are studded with photosensitive tubes that convert these flashes of light into electronic signals.
The two precisely timed flashes are unique to antineutrinos, but physicists must still contend with confounding signals, particularly those created by charged particles that originate in Earth’s atmosphere. They can insulate detectors from this background noise by placing them underground.
The Point Lepreau detector, which will measure about 3 meters on each side, is the fourth in a series developed by the Livermore-Sandia team. The previous three detectors were installed over the course of the past decade at a 1.1-GW pressurized water reactor at California’s San Onofre Nuclear Generating Station.
The team’s first demonstration detector at San Onofre picked up about 400 antineutrinos per day during a 600-day test period. That was good enough to provide near real-time data on the state of the reactor. The detector could “tell when the reactor has been turned off within a few hours, which is important, because if you wanted to remove any fissile material, you would have to turn the reactor off,” says team member Timothy Classen, of Lawrence Livermore.
But the team hopes it can mine antineutrino data to give even more information about what’s going on inside the reactor. An operator that runs a reactor at a higher power or uses fuel with more uranium-238 can boost the production rate of plutonium that could be used for nuclear weapons. Uranium releases more detectable antineutrinos than plutonium does, so monitoring the rate of antineutrino emission could potentially indicate whether a reactor is being run as intended or if material for weapons has been removed.
The International Atomic Energy Agency (IAEA) deems just 8 kg of plutonium a “significant quantity,” because it is enough to make a nuclear explosive device. On their own, antineutrino detectors may never be able to sense whether such a small amount goes missing from the entire reactor. But the data they collect, combined with reactor simulations, might be able to show whether each grouping of fuel rods contains as much plutonium as expected to within a “significant quantity” at the end of a fuel cycle
A french team led by Thierry Lasserre and David Lhuillier has developed a detector called Nucifer, set to be installed at the 70-megawatt Osiris research reactor within the next several months. Due to some design improvements, Nucifer is expected to detect half of the antineutrinos that interact with the detector—a fivefold increase over the Livermore-Sandia team’s original San Onofre detector. If Nucifer were installed at a similar plant at the same distance from the core, the team estimates it would register several thousand antineutrinos each day—enough to detect a change in plutonium content.
Bernstein estimates the cost of the Livermore-Sandia detectors at US $250 000 but expects that could be driven down to $100 000.