⚛️ high-energy experiments

Ferromagnetic broadband sensing of axionlike dark matter

This paper presents a high-bandwidth levitated magnetometer utilizing a double-resonance mode and hybrid ferromagnetic components to achieve a magnetic-field resolution of 0.7 fT, thereby establishing new direct limits on axionlike dark matter in the 40–3000 Hz frequency range that improve upon previous results by over four orders of magnitude.

Original authors: Chenhao Peng, Dmitry Budker, Yuanning Gao, Xinran Li, Jia Liu, Wei Ji, Jing Shu, Zhenxing Tang, Liheng Wang

Published 2026-06-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Chenhao Peng, Dmitry Budker, Yuanning Gao, Xinran Li, Jia Liu, Wei Ji, Jing Shu, Zhenxing Tang, Liheng Wang

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine trying to hear a single, faint whisper in a very noisy room. That is essentially what scientists are doing when they search for axions, a mysterious type of invisible particle that might make up "dark matter" (the stuff holding galaxies together that we can't see).

This paper describes a new, super-sensitive "ear" built by researchers at Peking University and their international partners to listen for these whispers. Here is how they did it, explained simply:

1. The Problem: The "Goldilocks" Dilemma

In the past, scientists built sensors to find these particles, but they faced a tricky problem.

The "High-Pitch" Sensor: Some sensors were incredibly sensitive but only worked on a very specific, narrow frequency (like a radio tuned to exactly one station). To find an axion, you'd have to tune this sensor to thousands of different frequencies one by one. This would take hundreds of years!
The "Broad" Sensor: Other sensors could listen to a wide range of frequencies at once, but they weren't sensitive enough to hear the faint axion whisper.

The team wanted a sensor that was both super-sensitive and able to listen to a wide range of frequencies at the same time.

2. The Solution: A Levitating Magnet on a "Double-Track"

The researchers created a device called LeMaMa (Levitated Magnet Magnetometer).

The Floating Magnet: Instead of a heavy magnet sitting on a table, they used a tiny magnet (about the width of a human hair) that floats in mid-air. It is held up by magnetic forces and sits inside a vacuum chamber (a box with no air) so nothing bumps into it. Because it floats and doesn't rub against anything, it is incredibly quiet and sensitive.
The "Double-Resonance" Trick: Usually, a floating object has one natural "humming" frequency (like a guitar string). If you push it at that frequency, it swings wildly. The team engineered their magnet to have two slightly different natural frequencies (271 Hz and 276 Hz) that are very close together.
- The Analogy: Imagine a swing set with two swings side-by-side. If you push them at slightly different times, they create a combined effect that covers a much wider range of motion than a single swing could. This allowed the sensor to be sensitive across a "broadband" range (from 40 Hz to 3000 Hz) without losing its super-sensitivity.

3. The Setup: A Fortress Against Noise

To hear the axion, they had to block out all other noise (like the hum of a refrigerator or vibrations from the ground).

The Shield: They wrapped the sensor in a special soft-metal shield that acts like a noise-canceling headphone for magnetic fields, blocking out outside interference.
The Converter: They built a giant ring made of thousands of tiny, hard magnets. This ring creates a strong, invisible magnetic field. If an axion passes through this field, physics says it should turn into a tiny, oscillating magnetic signal. The ring is designed to catch this signal and boost it so the floating sensor can hear it.
The Reader: A laser beam bounces off the floating magnet. If the magnet moves even a tiny bit (because an axion pushed it), the laser bounces differently. A camera-like detector tracks this movement with extreme precision.

4. The Results: A New Record

The team ran this experiment for about 15 hours, scanning frequencies between 40 Hz and 3000 Hz.

What they found: They did not find any axions. (This is actually good news for science; it means they ruled out the possibility that axions exist at those specific strengths).
The Achievement: Even though they didn't find the particle, they set a new record for how small a signal they could detect.
- They improved upon previous limits by more than 10,000 times (four orders of magnitude) in their best frequency range.
- They are now the first to set strict limits on axions in the 500 Hz to 3000 Hz range.
- Their sensor was sensitive enough to detect a magnetic field change as small as 0.7 femtotesla. To visualize this: it's like detecting the magnetic field of a single neuron firing, from a distance, while standing on a vibrating subway train.

5. Why This Matters

This experiment proves that you don't need to be in a freezing-cold laboratory (like many other physics experiments) to get these results. They did it at room temperature.

The paper claims this "hybrid" system (floating magnet + special magnet ring) is a powerful new tool. It opens the door to searching for dark matter much faster than before because it doesn't have to tune one frequency at a time. The authors also mention that this same technology could be used to detect biological signals (like heart or brain activity) or search for other new physics, but the primary claim of this specific paper is the successful demonstration of this new, ultra-sensitive, wide-bandwidth sensor for dark matter hunting.

Technical Summary: Ferromagnetic broadband sensing of axionlike dark matter

Problem Statement
The detection of axion and axionlike particles (ALPs) as dark matter candidates requires sensors capable of both high magnetic sensitivity and large bandwidth. While levitated sensors offer ultra-high sensitivity due to extremely low dissipation, traditional resonant approaches face a fundamental trade-off: high sensitivity is typically achieved at the expense of bandwidth. Previous room-temperature levitated magnetometer (LeMaMa) sensors demonstrated femtotesla-level sensitivity but possessed narrow resonance linewidths (approx. 0.01 Hz), making the scanning of broad axion mass ranges prohibitively time-consuming (potentially requiring years to cover relevant parameter spaces). Furthermore, direct laboratory searches in the intermediate frequency band (40–3000 Hz) have been sparse due to the difficulty of simultaneously achieving strong magnetic fields over large volumes, low environmental noise, and quiet readout.

Methodology
The authors present a hybrid ferromagnetic platform utilizing a levitated magnet magnetometer (LeMaMa) engineered for broadband operation. The experimental setup consists of three primary components:

Levitated Sensor: A cylindrical ferromagnetic magnet ( $r \sim 500$ µm, $h \sim 200$ µm) is levitated above a pyrolytic graphite layer using a lifting magnet, providing diamagnetic stabilization. The sensor operates at room temperature to avoid cryogenic vibration noise. Its mechanical motion is read out via a 780 nm probe laser reflected onto a quadrant photodetector (QPD), reconstructing two-dimensional angular motion.
Axion-Field Converter: A toroidal array of 13,248 hard-ferrite permanent magnets generates a static magnetic field ( $B_0 \sim 0.26$ T) within a central bore. This configuration induces an effective oscillating current density ( $J_a = -g_{a\gamma} \dot{a} B_0$ ) in the presence of an axion field, producing an oscillating magnetic signal ( $B_a$ ) at the sensor location. Ferrite magnets are used to minimize thermal current noise due to their high resistivity.
Shielding and Noise Suppression: A soft-ferromagnetic shield attenuates ambient magnetic noise, while the use of insulating permanent magnets eliminates noise from current fluctuations.

Key Technical Innovation: Double-Resonance Mode
To overcome the bandwidth limitation, the authors engineered the sensor to exhibit two librational modes with slightly split resonance frequencies ( $f_1 \approx 271$ Hz and $f_2 \approx 276$ Hz). By tailoring the dissipation (quality factors $Q_1, Q_2 \approx 1.7\text{--}2.3 \times 10^3$ ) and resonance frequencies, the system achieves an effective sensitivity linewidth ( $\Gamma_{sen}^{FWHM}$ ) of approximately 12 Hz. This represents a broadening of the accessible bandwidth by more than three orders of magnitude compared to single-mode, high-Q LeMaMa sensors, while maintaining high magnetic sensitivity. The detector response is modeled as a two-mode linear system, accounting for the interference between modes and the projection of the signal onto the readout channels.

Results
The experiment accumulated approximately 150 hours of data, with 15 hours dedicated to the final axion-model test and limit setting.

Sensitivity: The system achieved a magnetic-field resolution of 0.7 fT near the upper resonance frequency of 276 Hz.
Search Limits: A direct search for axionlike dark matter via the photon coupling ( $g_{a\gamma}$ ) was conducted over the 40–3000 Hz frequency range. No evidence for an axion signal was found.
Constraints: The study established new 95% confidence-level upper limits. The strongest limit was reached near 273.45 Hz, yielding $g_{a\gamma} < 1.8 \times 10^{-7} \text{ GeV}^{-1}$ . Over the 270–282 Hz interval, the limit remained below $7.0 \times 10^{-7} \text{ GeV}^{-1}$ .
Comparison: These results improve upon previous direct laboratory limits in this frequency band by more than four orders of magnitude. Specifically, the current result surpasses the projected sensitivity of a one-year integration for a levitated-superconductor proposal by over one order of magnitude and exceeds the projection for a superconducting-levitation ferromagnet proposal by over six orders of magnitude, despite the current analysis using only 15 hours of data.

Significance and Claims
The paper claims that this work demonstrates a viable hybrid ferromagnetic platform for axionlike dark matter searches that successfully balances high sensitivity with broad bandwidth. Key claims regarding the significance of the work include:

Bandwidth Breakthrough: The engineered double-resonance mode extends the effective sensitivity linewidth by three orders of magnitude, addressing the primary bottleneck in resonant dark matter searches.
Room-Temperature Operation: The system operates entirely at room temperature, avoiding the vibration noise associated with cryogenic refrigeration systems that often limit low-frequency searches.
Scalability: The authors identify a clear upgrade path, suggesting that optimizing vacuum conditions, optical reflectivity, and increasing the converter array size could improve the reach for $g_{a\gamma}$ by three orders of magnitude, potentially reaching sensitivities of $g_{a\gamma} \sim 10^{-10}$ .
Broader Applicability: Beyond axion searches, the authors note that the broadband, high-sensitivity, room-temperature nature of the LeMaMa sensor makes it suitable for other applications, including biological magnetic signal detection (e.g., heart, brain, and skeletal muscle signals) and searches for exotic spin-dependent interactions.

The paper concludes that the demonstrated high-bandwidth levitated sensor constitutes a significant step forward in the direct detection of ultralight dark matter, offering a competitive alternative to existing indirect bounds and resonant scanning methods.

1. The Problem: The "Goldilocks" Dilemma

2. The Solution: A Levitating Magnet on a "Double-Track"

3. The Setup: A Fortress Against Noise

4. The Results: A New Record

5. Why This Matters

More like this