Hunting for dark matter axions

Gravity has been studied since the days of Isaac Newton and has more recently been understood to obey Einstein’s theory of general relativity. But the movements of galaxies cannot be explained by gravitational laws alone. To reconcile theory and observation, scientists have hypothesized a kind of “dark matter” as an unseen source of gravity. Though direct experimental observation of dark matter has proven elusive, scientists have used a century of astronomical observations to predict that roughly 85 percent of the universe consists of dark matter. 

Axions, a weakly interacting low mass particle, are amongst the contender subatomic particles in the search for direct evidence of dark matter. It takes finesse to know how and where to look for dark matter axions, which are thought to be roughly ten billion to a few hundred billion times lighter than an electron. 

At Pacific Northwest National Laboratory (PNNL), physicists Erik Lentz, Christian Boutan, and Noah Oblath hunt for these elusive particles. As part of the Axion Dark Matter eXperiment (ADMX), they employ an array of techniques, from radio frequency (RF) signal detection to quantum sensing  and even artificial intelligence.

“The particular way we look for axions is how they interact with photons,” said Oblath. “If axions exist, they would couple with photons and, in the presence of a magnetic field, convert into microwave wavelength photons.” Oblath specializes in RF signal detection and has applied his expertise to other projects involving tiny, elusive neutrinos. Though Oblath’s description of finding axions sounds relatively straightforward, there is one huge challenge.

“We’re looking for a very, very small signal in terms of overall power—something that's measured on the level of yoctowatts,” said Lentz. That amount of power is roughly one octillionth of the power of a standard light bulb. 

To detect these tiny signals, the multi-institutional ADMX team built an instrument called a haloscope, which consists of an enormous superconducting magnet and an insert—both of which are currently located at the Center for Experimental Nuclear Physics and Astrophysics (CENPA) at the University of Washington in Seattle—connected to an extraordinarily sensitive RF detector. A central feature of ADMX is the low operating temperature of the cavity, at the cryogenic level, which helps to reduce noise.

[ https://youtu.be/wC8FPr2NbKU ]

The idea behind this construction is that, when axions are exposed to the magnetic field, they would produce photons that RF detectors could recognize as a signal. However, the instrument that can detect such an event needs to be incredibly sensitive.

“To be a sensitive enough detector, we need to ‘tune’ the cavity to the particular frequency that we are looking for,” said Oblath. 

When the axion converts into photons, the frequency of the photons depends on the mass of the axion, which we don’t know. The axion-created photons would get amplified in power when the cavity mode overlaps with that frequency, hopefully to the point where the signal is powerful enough to observe.

Boutan added, “it’s essentially like listening for a faint radio station—you keep turning the dial until you find it. We slowly tune our experiment and listen with one of the world's most sensitive cryogenic radio receivers, hoping to hear a cosmic radio station for the first time. This find would represent the discovery of dark matter axions.”            

Finding a signal in a sea of noise

One challenge with trying to detect a very tiny signal is that even if the sought-after signal were to be detected, the noise—the background fluctuations read out by the detector—would overpower the signal. 

“There are two ways to face this challenge—we can either reduce the noise, or enhance the signal,” said Lentz.

One way the PNNL team does this is by amplifying, filtering, and digitizing the signals that come out of the cryogenic space. Leveraging their expertise in RF engineering, the PNNL team designs and constructs the RF electronics attached to the haloscope.

More recently, the researchers have used artificial intelligence to characterize the RF signals. This study is one component of Boutan’s Early Career Research project funded by the Department of Energy (DOE).

“We assemble and characterize the RF components of the instrument at ambient temperature, but we run the experiments near absolute zero to reduce noise from photons emitted by the instrument,” said Boutan. “Changing the environment makes it difficult to calibrate each component. We are exploring how machine learning can help with this.”

[  https://youtu.be/C2IgEGfLgS4 ]

Bringing quantum sensing to the search for dark matter

Typically, a principle called the ‘standard quantum limit’ sets the lower limit for how much noise a system produces. However, it may be possible to circumvent this limit with quantum technologies.

“In theory, it is possible to push our sensors beyond the standard quantum limit to something called the Heisenberg limit,” said Lentz. “One way we can do this is having many cavity haloscopes operating in unison and correlating them together using a quantum sensor network. This way, their net signal strength increases like the sum of the cavities, but the noise intensity remains at the level of one cavity. This is much better than the operating the haloscopes independently.”

Another way to reduce noise is to create a new readout mechanism using qubits to count the axion-produced photons. 

“Instead of amplifying the signal, creating more noise, and measuring yoctowatts of extra power, we’d like to build a microwave photon counter instead, that could count "clicks" like a Geiger counter,” said Boutan. "I am working on using a superconducting qubit, as the basis for this counter as we speak." 

Since axion-produced photons will only occur around a set frequency, it is possible to differentiate the signals from those photons from the noise across all frequencies. Additionally, since the power of the photons is proportional to the magnetic field squared inside the cavity, one can lower the magnetic field strength and watch the signal get smaller.

Boutan, Lentz, and Oblath will continue to incorporate advanced technologies like these into their search for dark matter particles.

Stretching imagination

Inspired by science fiction, Lenz has also used his free time to think about how to advance space transportation through warp drives.

“Out of pure fascination, I had been tracking the scientific literature of warp drives,” said Lentz. “Then when I found myself with free time during the COVID lockdown, I started to look into the literature more carefully and play with Einstein’s equations to see if I could make some progress towards proposing a more realistic warp drive.”

Lentz’s interest in warp drives resulted in a publication in Classical and Quantum Gravity. Though PNNL did not initially have a research project in this area when Lentz began working at the Laboratory, Maj. Gen. Sandy Finan (United States Air Force, retired), who is an advisor at PNNL with a background in nuclear, space, and cyber, encouraged his pursuit of this research topic.

“This work is the kind of cutting-edge research that stretches out our thinking and could lead to huge breakthroughs if successful,” said Finan.

ADMX is sponsored by the Office of Science, High Energy Physics, of DOE and is an international collaboration composed of researchers from University of Washington, University of Florida, Lawrence Livermore National Laboratory, Fermi National Laboratory, PNNL, the University of California, Berkeley, Washington University in St. Louis, Sheffield University, and the National Radio Astronomy Observatory. Fermilab is a DOE Office of Science, HEP User Facility. Boutan’s research is in part supported by the Early Career Research Program sponsored by the DOE HEP Program. Lentz’s warp drive work was sponsored by the Alexander von Humboldt Foundation through its Sofja Kovaleskaja award.