Constraining axion-like dark matter with a radio-frequency atomic magnetometer

Using a broadband radio-frequency $^{87}\mathrm{Rb}$ atomic magnetometer, researchers conducted a search for axion-like dark matter in the mass range of $2.40\times10^{-10}$ to $2.11\times10^{-9}\,\mathrm{eV}/c^{2}$, finding no significant signals but establishing improved upper limits on ALP-proton couplings and complementary constraints on ALP-neutron and ALP-electron interactions.

The Gist

The Big Picture: Hunting for "Ghost" Particles

Imagine the universe is filled with a thick, invisible fog. Scientists call this Dark Matter. We know it's there because it has gravity (it holds galaxies together), but we can't see it, touch it, or smell it.

For decades, scientists have been trying to figure out what this fog is made of. One popular theory suggests it's made of ultra-light particles called Axions (or Axion-Like Particles). These aren't heavy like a rock; they are so light that they behave more like a wave rippling through the entire universe.

The Goal: This paper describes an experiment where a team of scientists built a super-sensitive "radar" to try and catch a ripple from this axion fog.

---

The Detective Tool: The Atomic Magnetometer

To catch these invisible waves, the scientists used a device called a Radio-Frequency Atomic Magnetometer.

The Analogy: Think of the device as a giant, high-tech compass made of atoms.

• Inside a glass bulb, they heated up a cloud of Rubidium atoms (a type of metal) until they were a vapor.
• They used lasers to line up all these atoms like tiny soldiers standing at attention.
• Normally, these atoms just spin in place. But if a "ghost wave" (an axion) passes through, it should give them a tiny, rhythmic nudge, making them wobble in a specific way.

The scientists tuned this "atomic compass" to listen for wobbles at different speeds (frequencies). They scanned a wide range of speeds, looking for a specific "hum" that would indicate an axion was passing by.

The Search: Listening for a Whisper

The scientists scanned a massive range of frequencies (from 58 kHz to 510 kHz). This is like tuning a radio across the entire dial, listening for a specific station that no one else has found yet.

• The Challenge: The signal they are looking for is incredibly faint. It's like trying to hear a single person whispering in a stadium full of cheering fans.
• The Noise: The "fans" in this case are tiny electrical jitters, heat from the equipment, and the Earth's own magnetic field.

The Results: A Quiet Night

After running the experiment for many hours and analyzing the data with powerful computers, the scientists found no axions.

The Metaphor: Imagine you are waiting for a specific bird to sing in your garden. You set up a super-sensitive microphone, wait for days, and record everything. You hear the wind, the cars, and the crickets, but you never hear that specific bird.

Does this mean the bird doesn't exist? No. It just means:

1. The bird might not be in this part of the garden.
2. The bird might sing at a different time or pitch than you were listening for.
3. The bird might be quieter than your microphone can hear.

What Did They Actually Achieve?

Even though they didn't find the axion, this is a huge success for science. Here is why:

1. Ruling Out a Neighborhood: They didn't just say "we didn't find it." They said, "We looked in this specific range of frequencies (masses), and if axions existed there, we would have seen them. Since we didn't, axions probably don't exist in this specific range."
2. Improving the Map: Previous experiments had looked at "heavier" axions. This team looked at a "lighter" range that hadn't been explored much before. They effectively drew a new "No Entry" sign on the map of the universe.
3. Better than Before: For the interaction between axions and protons (parts of the atom's nucleus), they set a stricter limit than any previous lab experiment. They proved that if axions exist, they are even more elusive than we thought.

Why Does This Matter?

Think of this like searching for a lost key.

• Astrophysicists look at stars and say, "If the key were this heavy, the stars would burn differently. So the key isn't that heavy."
• This Experiment is like looking in your own house with a flashlight. They are saying, "We looked right here, in this specific spot, and the key isn't here."

By combining these different approaches, scientists are slowly narrowing down the search area. Every time they say "It's not here," they get one step closer to finding where it is.

The Takeaway

The scientists built a super-sensitive atomic compass, scanned a wide range of frequencies for a "ghost" particle, and found nothing. While this might sound disappointing, it's actually a victory. They have successfully eliminated a large chunk of the universe where these particles could have been hiding, forcing us to look in new places or build even better detectors.

In short: They didn't find the ghost, but they proved the ghost isn't hiding in the living room. Now, we know exactly where not to look, which brings us closer to finding it.

Technical Summary

1. Problem Statement

The nature of Dark Matter (DM) remains unknown, though it constitutes the majority of the Universe's mass. A leading candidate is Ultralight Dark Matter (UDM), specifically Axion-Like Particles (ALPs) with masses in the range of $10^{-22}$ to $10$ eV/$c^2$. If ALPs constitute the local DM halo, they form a classical field that oscillates at a frequency determined by their mass (Compton frequency).

This field interacts with Standard Model fermions (electrons, protons, neutrons) via a gradient coupling, which generates an effective, oscillating pseudomagnetic field.

• The Challenge: Most existing spin-based searches are sensitive to lower mass ranges (lower frequencies). There is a significant gap in experimental constraints for ALP masses between $10^{-10}$ and $10^{-9}$ eV/$c^2$ (corresponding to radio frequencies of tens to hundreds of kHz).
• The Goal: To perform a broadband search for ALP-induced pseudomagnetic fields in the mass range $2.40 \times 10^{-10}$ eV/$c^2$ to $2.11 \times 10^{-9}$ eV/$c^2$ using a high-sensitivity atomic magnetometer, and to derive constraints on ALP-fermion couplings ($g_{\alpha pp}, g_{\alpha nn}, g_{\alpha ee}$).

2. Methodology

Experimental Apparatus

The authors employed a radio-frequency (RF) atomic magnetometer utilizing a vapor of spin-polarized $^{87}$Rb (Rubidium-87) atoms.

• Cell: A spherical glass cell (17.5 mm diameter) containing $^{87}$Rb vapor heated to 130°C, mixed with 4He (600 torr) and N$_2$ (50 torr) to suppress spin decoherence and improve polarization.
• Optical Setup:
  • Pump: A circularly polarized laser (780.2 nm) optically pumps atoms along the $\hat{z}$-axis, creating spin polarization.
  • Probe: A linearly polarized laser (detuned ~100 GHz below resonance) propagates along $\hat{x}$. It measures the transverse spin component via optical Faraday rotation.
• Magnetic Environment:
  • The cell is housed in an aluminum enclosure (20 mm thick) rather than a high-permeability ($\mu$-metal) shield. This avoids shielding the exotic ALP field while providing RF shielding in the 58–510 kHz range.
  • Ambient DC fields are nulled using Helmholtz coils.
  • Active Stabilization: A fluxgate magnetometer monitors the field along $\hat{z}$ to actively stabilize the Larmor frequency ($\nu_L$) against ambient drifts.

Detection Principle

1. Resonance: The magnetometer is tuned to a specific Larmor frequency $\nu_L$ by adjusting the bias field $B_0$.
2. Signal: If an ALP field exists, its gradient creates an oscillating pseudomagnetic field $B_\alpha(t)$ at the ALP frequency $\nu_a$.
3. Mechanism: When $\nu_a \approx \nu_L$, the ALP field coherently drives the atomic spins, generating a resonantly enhanced transverse spin component.
4. Readout: The Faraday rotation of the probe beam is detected via a balanced photodetector and lock-in amplifier. The system scans $B_0$ to sweep $\nu_L$ across the target frequency range (58–510 kHz).

Data Analysis

• Spectral Acquisition: Data was collected in three overlapping frequency intervals (58–110 kHz, 90–270 kHz, 250–510 kHz).
• Noise Normalization: To handle technical noise scaling as $1/\nu$, the power spectra were normalized using a two-step filtering process (Savitzky-Golay filter followed by a polynomial fit) to establish a local mean noise level.
• Matched Filtering: A matched filter with a squared-Lorentzian kernel (matching the magnetometer's 4.5 kHz linewidth) was applied to the normalized spectra to search for peaks.
• Statistical Threshold: A detection threshold ($X_{thr} = 1.183$) was established at the 95% confidence level using $10^5$ Monte Carlo realizations of synthetic noise, accounting for the "look-elsewhere effect" across the full bandwidth.

3. Key Contributions

1. Broadband Search in an Unexplored Regime: The experiment successfully covered the mass range $2.40 \times 10^{-10}$ to $2.11 \times 10^{-9}$ eV/$c^2$ (58–510 kHz), a region largely unexplored by previous spin-based probes which typically focus on lower masses.
2. Derivation of Conversion Factors: The authors rigorously derived the conversion factors ($\chi_f$) linking the measured atomic spin observable to the fundamental ALP-fermion couplings for electrons, protons, and neutrons.
   • For electrons: $\chi_e = 1/(2I+1)$.
   • For nucleons: A complex dependence on spin temperature and nuclear structure was calculated, yielding $\chi_p \approx 0.499$ and $\chi_n \approx 0.164$ for $^{87}$Rb.
3. Systematic Error Management: The paper details the handling of systematic uncertainties, including the lack of $\mu$-metal shielding (relying on active stabilization) and the calibration of RF fields, estimating an overall $\sim 10\%$ uncertainty in the magnetic field response.

4. Results

• No Detection: No statistically significant signatures of an oscillating magnetic field were observed. A single outlier peak at ~407 kHz was identified but ruled out as an instrumental artifact (intermittent laser noise) after verification with auxiliary datasets.
• Sensitivity: The experiment achieved a magnetic field sensitivity of approximately 15 fT/$\sqrt{\text{Hz}}$ at the high-frequency end of the range.
• Exclusion Limits: The authors derived 95% confidence level upper limits on the ALP-fermion couplings:
  • $g_{\alpha pp}$ (Proton): The results improve upon previous laboratory constraints (specifically fifth-force searches) across the entire investigated mass range, reaching a limit of $\approx 9 \times 10^{-5}$ GeV$^{-1}$ at $m_\alpha \approx 1.1 \times 10^{-9}$ eV/$c^2$.
  • $g_{\alpha nn}$ (Neutron) and $g_{\alpha ee}$ (Electron): While less stringent than some fifth-force or astrophysical bounds, these results are complementary. Unlike fifth-force experiments (which probe static potentials) or astrophysical bounds (which do not assume DM), this experiment specifically targets the coherently oscillating DM halo.

5. Significance

• Targeting the DM Halo: This work provides a direct laboratory constraint on ALPs as constituents of the Galactic Dark Matter halo, distinguishing it from constraints derived from non-DM scenarios (like fifth-force experiments) or astrophysical observations.
• Methodological Validation: It demonstrates the viability of radio-frequency atomic magnetometers for broadband axion searches, proving that high sensitivity can be maintained over a wide frequency range without the need for cryogenic or heavy magnetic shielding that might attenuate the signal.
• Future Outlook: The authors suggest that sensitivity could be further improved by increasing the effective magnetometric volume (larger beams), optimizing optical pumping to narrow the resonance linewidth, or employing parametric-resonance techniques to suppress spin-exchange relaxation.

In summary, this paper establishes the most stringent laboratory limits to date on ALP-proton couplings in the $10^{-10}$–$10^{-9}$ eV/$c^2$ mass range and provides a robust framework for future broadband searches for ultralight dark matter.