Squeezing light: Gravitational wave interferometry is on a new trajectory

On 14 September 2015, the detectors of the Laser Interferometer Gravitational-Wave Observatory (LIGO) stirred with activity as a gravitational wave emitted by a black hole collision roughly one billion years ago swept past Earth. The disturbance first rattled LIGO’s detector in Hanford, Washington, and then seven milliseconds later, after transiting southeast at the speed of light, the wave jostled another LIGO detector in Livingston, Louisiana. Humanity had succeeded in making its first detection of gravitational waves, opening a new window for astronomical observation.

Gravitational wave messages are encoded in ripples of space-time emitted when two massive objects such as black holes or neutron stars collide. As Albert Lazzarini, a deputy director of LIGO, says, “These coalescences of massive objects are occurring all the time, and they take us all the way back to the very first stars that were formed, right after the Big Bang.”

Gravitational wave astronomy will help explore some of the great questions in physics: What did the universe look like right after the Big Bang? How does matter act under the extremes of temperature and pressure in neutron stars and supernovae?

Physicists are now hoping to glean new insight into these standing questions when the search for gravitational waves resumes later this year with highly anticipated improvements. Notably, these include an emerging quantum-optics technology called “squeezed light,” which provides a means for making gravitational wave detectors even more sensitive. Squeezed light works by advantageously manipulating the quantum mechanical behavior of photons.

As Lisa Barsotti, an astrophysicist at the Massachusetts Institute of Technology (MIT) and a LIGO collaborator, says, “We hope to have access to different types of sources, like supernovae, since we have never detected the gravitational wave side of them. And squeezed light is part of this.”

Later this year, LIGO will be joined by Europe’s Italy-based Virgo detector, named after the Virgo Cluster of galaxies, as well as by Japan’s Kamioka Gravitational-Wave Detector (KAGRA). LIGO and Virgo will be coming on-line with improved implementations of squeezed light, whereas KAGRA will be implementing squeezed light for the first time. Together, these detectors are ushering in a new, sensitivity-enhanced era of gravitational wave observation.

To detect gravitational waves, LIGO, Virgo, and KAGRA employ a familiar light-based measurement strategy—interferometry. In fact, their main facilities are giant interferometers.

An interferometer can be viewed as an optical junction. Using a beam splitter, light is sent down two arms, each of which contains a mirror at its end. The light bounces off these end mirrors and is sent back towards the beam splitter. Upon encountering the beam splitter for a second time, the light from the mirrors combines to create an interference signal—what an interferometer measures—and does so by routing the interference signal to a photodetector, where it is converted to an electrical signal that can be recorded and analyzed.

As Lee McCuller, a physicist at the California Institute of Technology (Caltech) and a LIGO collaborator, explains, “You can really think of the interferometer as an antenna.” It encodes gravitational waves into a beam of light. In the case of gravitational wave interferometers, this is accomplished by bouncing a laser beam off mirrors positioned at the ends of kilometers-long vacuum tubes. As a gravitational wave passes it displaces the mirrors ever so slightly, which alters the interference signal. And this is how gravitational waves are measured interferometrically. That it works is nothing short of astonishing. The displacements induced by gravitational waves are smaller than the width of an atomic nucleus.

The LIGO project, which broke ground on its gravitational- wave interferometers in 1994, is an international consortium of some 1,000 scientists managed by Caltech and MIT, with funding from the US National Science Foundation. To achieve the remarkable sensitivity of its interferometers, LIGO pushed laser power to the limit. When you make a laser very powerful, the quality of the beam can easily deteriorate. For instance, when you ramp up the power of a laser, heat becomes a big deal. To deal with this, LIGO invented a new type of laser, a zig-zag slab laser, which provided a novel pump geometry, permitting effective cooling at higher powers. These innovations have since enabled ultrafast laser applications, such as those involving the study of light-matter interactions occurring on the order of a quintillionth of a second (10^–18 s).

Increasing power also causes a loss of control over the frequency of the light. To mitigate this, LIGO proceeded to invent a means of stabilizing its high-power lasers, a technique called Pound-Drever-Hall (PDH) locking, which allows high-power lasers to have a precisely defined frequency response. Since being invented, PDH locking has become a cornerstone technology, not just in gravitational wave interferometers, but also optical communications and metrology, where controlling the frequency response of a laser can be critical.

High-power lasers allow LIGO and Virgo to amplify the extraordinarily weak signals originating in their interferometers. When you increase laser power, McCuller says, “you get a statistical improvement in sensitivity.” And in LIGO and Virgo, they get more bang for their buck by turning each arm of their interferometers into a type of optical resonator called a Fabry-Perot cavity. These cavities receive an already high-power light source and amplify it further still.

LIGO’s mirrors must be so well shielded from external vibration that the random motion of the atoms within the mirrors and their housings can be detected. To do so, they are suspended at the end of a 360 kg quadruple pendulum system. Photo credit: LIGO Laboratory

This is how LIGO and Virgo succeeded in making their first detections of gravitational waves. In the first and second observing runs, LIGO and Virgo were circulating 200 kW of laser power around their interferometers.

However, as the initial data trickled in between 2015–17, it was clear that increased laser power alone wasn’t going to be enough to increase sensitivity as much as desired. LIGO and Virgo had pushed their laser power to the maximum and were detecting roughly one event per month. At this rate, physicists would have to wait a very long time to resolve questions about what was going on in the early days of the universe.

Preventing LIGO and Virgo from recording more detections over the first two runs was the fact that the interference signals exiting their beam splitters were riddled with noise from so-called quantum vacuum fluctuations. As Barsotti explains, these fluctuations “push” the photons around in the interferometer. They mess up the timing of the photons as they arrive at the photodetector. Some photons arrive early while others arrive late, producing a bell-curve distribution in photon arrival times that manifests as noise in the recorded signal.

Luckily, LIGO and Virgo had a trick up their sleeve: squeezed light. By 2019, LIGO and Virgo succeeded in implementing squeezed light in all three of their detectors. And the improvement was astonishing. On 26 March 2020, when LIGO and Virgo concluded their 11-month-long third observing run, they had succeeded in detecting a whopping 86 gravitational wave events. And that number could have been higher. When the covid-19 pandemic made it impossible to continue operations in early 2020, the third run was suspended.

In gravitational wave interferometers, squeezed light improves sensitivity by narrowing the distribution of photon arrival times. But LIGO and Virgo didn’t just pull squeezed light out of a hat as if it were an off-the-shelf remedy. Rather, the search for gravitational waves provided the impetus that drove squeezed light into existence in the first place. And this goes back to the 1980s, when the impact of these quantum vacuum fluctuations was first understood by Caltech physicist Carlton Caves. After completing a PhD under Kip Thorne, who received the 2017 Nobel Prize in Physics for his contributions to LIGO, Caves outlined how these vacuum fluctuations enter an interferometer and, more importantly perhaps, what could be done about them.

Caves discovered that the vacuum fluctuations enter the interferometer at the beam splitter where the interference signal departs through an interferometer’s so-called unused port. This is a port that has no signal input to the interferometer. But vacuum fluctuations do enter this port and propagate in the reverse direction of the interference signal. In fact, Caves says, “all of the (quantum) noises come from the unused port.”

Understanding this led Caves, in 1981, to propose the squeezed light solution, though it was only a mathematical concept at the time. “The point is that you can do something about it, you could change what’s coming into that (unused) port, from vacuum to something else,” he says. Squeezed light is the something else that you put into the unused port instead of vacuum. It effectively acts as a noise-altering signal.

Caves continues, “Squeezing is pretty simple. You take coherent light [a laser], run it through some nonlinear optics [a crystal], and you get squeezed light.”

To be clear, however, there is no physical squeezing of the light. Rather, squeezed light is said to represent a squeezed vacuum state produced by manipulating the quantum mechanical behavior of the photons. And this is what the crystal does. It produces pairs of correlated photons. For every one photon that enters the crystal, two come out, and they do so at exactly the same time.

“And that’s exactly what they’ve currently installed in LIGO,” says Caves. LIGO took Caves’ idea and ran with it, allowing squeezed light to emerge as a quantum-optics technology.

Caves’ proposal was one of the first times a quantum technology had been linked to a real-world application. A former Caves postdoc at the University of New Mexico, Rafael Alexander, says, “Here’s a real-world benefit. It will, at some point, make economic sense, rather than bringing the losses down, or ramping up the power of the pump [laser], to instead put quantum [squeezed] light into this port. And that’s what I think made it a technology as opposed to just a mathematical curiosity.”

In-vacuum equipment being installed as part of a LIGO detector squeezed-light upgrade. Photo credit: Caltech/MIT/LIGO Lab 

But squeezing light is not without a drawback. When you reduce one type of quantum noise, you amplify another. This is how quantum mechanics works—it is squishy, in a way. Squeeze it somewhere, it spikes elsewhere.

In the case of LIGO and Virgo, squeezed light amplified a type of noise, quantum radiation pressure noise, that caused the interferometers’ mirrors to move. Barsotti says that the radiation pressure noise can be attributed to the momentum of photons being transferred onto the mirrors, inducing movement.

Fortunately, for the LIGO/Virgo third observing run, the radiation pressure noise proved to be negligible. “That’s not going to be true in the future, because in the next observing run the laser power in the interferometer will be increased,” Barsotti says.

In the case of LIGO and Virgo, advances permitting an increase in laser power are now proving to be a double-edged sword. Compared to the third run, their light sources are even more powerful. Going forth, McCuller says, the interferometers will have a whopping 800 kW of light circulating within them. This improves sensitivity. But it also means increased radiation pressure noise, since more photons are inevitably hitting their mirrors. And this leads to diminished returns associated with squeezed light.

LIGO and Virgo’s most recent upgrades resolve the conundrum by complementing their increase in laser power with an improvement in squeezed light called frequency-dependent squeezing, which filters squeezed light as a function of frequency. It constrains the number of photons hitting the interferometer mirrors at low frequencies, while maintaining a tight distribution of photon arrival times at higher frequencies. This works since the radiation pressure noise is only a factor at low frequencies.

The LIGO and Virgo teams will use a filter cavity for frequency-dependent squeezing. As McCuller explains, “The cavity is 300-m long. The [squeezed] photons bounce back and forth in this cavity 6,000 times.” In this sense, the cavity plays the role of an optical capacitor, storing the squeezed light for around 3 milliseconds. By bouncing the photons back and forth within such a cavity, the correlations inherent to squeezed light become filtered in just the right way.

The cavities installed in LIGO and Virgo, sit in each detector between the squeezed light source and interferometer. “By preparing light in such a cavity, you can have your cake and eat it too,” Lazzarini says. And LIGO and Virgo will not be the only ones attending this party. In the fourth observing run, the KAGRA interferometer will also feature an implementation of frequency-dependent squeezed light.

When it kicks off later this year, the fourth observing run will feature four squeezing-enhanced interferometers on three continents—two in LIGO, and one each in Virgo and KAGRA. “And so, we hope,” says Barsotti. She is optimistic that increases in sensitivity afforded by frequency-dependent squeezing will permit an open question in physics to be resolved—the determination of the structure and composition of neutron stars.

Since gravitational wave detection provided the driving force that brought squeezed light to bear, the technology has become a go-to in the field of quantum optics. When improvements in measurement sensitivity are sought and more photons are not an option, squeezed light is one of the first methods that scientists consider. “So, naturally when people have an application where they think they might need quantum resources, they think about squeezing first, because you can get it,” Caves says. “It’s a technology that’s developing. It’s moving out of quantum metrology into quantum information processing.”

In the realm of quantum information processing, squeezed light provides a potentially game-changing technology for the budding quantum computer industry. In this regard, Alexander would seem to have come full circle from Caves’ lab to being the head of architecture at Xanadu, a Toronto-based quantum computer company.

“Squeezing has enabled a way of optical quantum computing,” Alexander says. It serves as a resource to encode and manipulate quantum information. And this happens by virtue of weaving together squeezed light to make entangled states. These entangled states are containers of quantum information, containers that can be transported optically and, if needed, manipulated by additional interferometers. “If you start with squeezed light, all you need to do to make this entangled state is send it through an array of interferometers.”

Now, the world awaits the squeezed-light-enhanced results of the LIGO, Virgo, and KAGRA fourth observing run. Those projects have bet big on squeezed light to enable significant discovery with gravitational wave astronomy. And until another quantum optics technology comes around, the efforts of Alexander and colleagues are also going to involve squeezed light. As Alexander quips,
“Without squeezing I wouldn’t have a job.”

Matthew Ellis is a life-long tinkerer and founder of Neuroptica, a medical imaging startup.