In the 1960s, researchers at the science lab of the Ford Motor Company developed the superconducting quantum interference device, also known as a “SQUID.” It was the first usable sensor to take advantage of a quantum mechanical property—in this case, superconductivity.
That made the SQUID one of the first generation of quantum sensors: devices that use a quantum system, quantum properties or quantum phenomena to make a physical measurement. Physicists took the idea and ran with it, coming up with new types of sensors they continue to use and improve today.
SQUIDs have played a key role in the development of ultrasensitive electric and magnetic measurement systems and are still in use. For example, they amplify the detector signals for the Super Cryogenic Dark Matter Search. “As particle physicists, we’ve been using quantum sensing techniques for decades,” says SuperCDMS physicist Lauren Hsu of the US Department of Energy’s Fermi National Accelerator Laboratory.
But SQUIDs are no longer the only quantum sensors around. One important recent development in quantum sensing is known as quantum squeezing—a way to circumvent quantum limitations that even quantum sensors have faced in the past.
“The only way to do better is to start beating quantum mechanics.”
The first quantum sensors
Ford’s SQUIDs, which needed to be cooled to a few degrees above absolute zero, used superconducting loops to measure minuscule magnetic fields.
SQUIDs didn’t turn out to be of much use in an automobile. But not all Ford researchers were beholden to expectations that their creations would wind up in a car. “This shows you how different the world was back in the 1960s,” says Kent Irwin, a physicist at Stanford University and SLAC National Accelerator Laboratory. “These days Ford is not doing basic physics.”
A few decades later, while in graduate school, Irwin built on the idea of the Ford Company’s SQUID to develop a new quantum sensor: the first practical superconducting transition-edge sensor.
Irwin took advantage of the fact that superconducting material loses its superconductivity when it heats up, regaining its resistance at a precise temperature. By keeping a superconducting material as close as possible to this temperature limit, he could create a sensor that would undergo a significant change at the introduction of even a small amount of energy. Just a single photon hitting one of Irwin’s transition-edge sensors would cause it to shift to a different state.
The transition-edge sensor is well-known and has been adopted widely in X-ray astronomy, dark matter detection, and measurements of the cosmic microwave background radiation. “It’s very much old-school quantum 1.0,” Irwin says.
Quantum sensing for gravitational waves
A new generation of quantum sensors goes beyond quantum 1.0. Some of today’s sensors make use of more than just superconductivity: They’ve managed to use the Heisenberg uncertainty principle—usually thought of as a limitation to how well physicists can make measurements—to their advantage.
The Heisenberg uncertainty principle puts a cap on how accurately you can measure a pair of related properties. For example, the more you know about the position of a particle, the less you can know about its momentum.
Quantum squeezing takes advantage of these relationships by purposefully tipping the balance: moving all the uncertainty of a measurement to one side or the other.
Gravitational-wave detectors, such as LIGO in the US, and Virgo and GEO in Europe, have used quantum squeezing to great effect. In 2015, LIGO—the Laser-Interferometer Gravitational-wave Observatory—detected the first gravitational waves, undulations of spacetime first predicted by Albert Einstein. Once it got going, it was picking up new signs of gravitational-wave events every month.
LIGO detects gravitational waves using an interferometer, an L-shaped device in which two beams of light are set up to bounce off identical mirrors and return. Under normal conditions, the beams will arrive at the same time and cancel one another out. No signal will hit the detector.
But if a subtle outside force knocks them out of sync with one another, they won’t cancel each other out, and photons will hit the detector. If a gravitational wave passes through the two beams, it will hit one and then the other, interrupting their pattern.
LIGO’s measurements are limited by the quantum properties of the photons that make up their beams of light. At the quantum level, photons are affected by fluctuations, virtual particles popping in and out of existence in the vacuum. Those fluctuations could cause a false signal in the detector. How could LIGO researchers tell the difference?
“LIGO is using the most powerful lasers they can build, and the best mirrors they can build, and their back is against the wall,” Irwin says. “The only way to do better is to start beating quantum mechanics.”
Scientists at LIGO and other gravitational-wave detectors looked to quantum squeezing to help them with their virtual photon problem.
To generate squeezed light, researchers used a technology called an optical parametric oscillator, within which an input wave of laser light is converted to two output waves with smaller frequencies. This process entangles pairs of photons, and the resultant correlations of their properties serve to reduce uncertainty in one aspect of the arriving photons, allowing LIGO scientists to better measure another aspect, helping them sort the signal from the noise.
Since April 2019, when LIGO began running with the quantum squeezers, the observatory has been able to detect new gravitational-wave signals—signs of collisions between massive objects such as black holes and neutron stars—more frequently, going from about one detection per month to about one per week.
Quantum sensing for dark matter detection
Quantum squeezing has also recently found an application in the search for dark matter.
Dark matter has never been observed directly, but clues in cosmology point to it making up approximately 85% of the matter in the universe. There are several different theories that describe what a dark matter particle could be.
“The mass can be anywhere from a billionth the size of an electron up to a supermassive black hole,” Hsu says. “There are over 100 orders of magnitude that it can span.”
The most promising small dark matter candidates are axions. In the presence of a strong magnetic field, axions occasionally convert into photons, which can then be detected by an experiment’s sensors.
Like someone trying to find a radio station on a road trip in the middle of nowhere, they scan for a while at one frequency, to see if they detect a signal. If not, they turn the dial a little and try the next size up.
It takes time to listen to each “station” once the detector is tuned to a particular possible axion signal; the more noise there is, the longer it takes to determine whether there might be a signal at all.
The HAYSTAC experiment—for Haloscope at Yale Sensitive to Axion Cold Dark Matter—searches for axions by measuring two different components of electromagnetic field oscillations. Like LIGO, it is limited by the uncertainty principle; HAYSTAC researchers are unable to precisely measure both oscillations at once.
But they didn’t need to. Like LIGO scientists, HAYSTAC scientists realized that if they could squeeze all the accuracy into just one side of the equation, it would improve the speed of their search. In early 2021, researchers announced that at HAYSTAC, they had also succeeded at using quantum squeezing to reduce noise levels in their experiment.
Multiple groups have demonstrated promising new applications of superconducting circuit technology for axion detection.
The “RF quantum upconverter” uses devices similar to Ford’s SQUIDs to evade the Heisenberg uncertainty principle in dark-matter searches at frequencies below HAYSTAC’s searches. Another uses a technology borrowed from quantum computing—qubits—as a sensor to evade Heisenberg’s limits at frequencies higher than HAYSTAC. Although neither technology has been used in dark matter searches yet, scientists believe that they could speed searches up by several orders of magnitude.
At the current rate, it will still take axion experiments thousands of years to scan through every possible axion “station.” They may get lucky and find what they’re looking for early in the search, but it’s more likely that they’ll still need to find other ways to speed up their progress, perhaps with advances in quantum sensing, says Daniel Bowring, a Fermilab physicist who is involved in another axion search, the Axion Dark Matter Experiment.
“It’s going to take a lot of people with really good imaginations,” Bowring says.