⚛️ phenomenology

Muonium Spectroscopy as a Quantum Sensor for Ultralight Axion Dark Matter

This paper proposes and demonstrates that high-intensity muon beams can enable a Muonium-based Axion Search (MASH) using resonant quantum transitions to constrain axion-muon couplings in the 18–130 $\mu$ eV mass range, thereby offering a new spectroscopic channel that complements and tightens limits beyond existing muon $g\!-\!2$ measurements.

Original authors: Feng Fang, Kim Siang Khaw, Ce Zhang, Qiaoli Yang, Liangwen Chen, Jie Yang, Lei Yang, Zhiyu Sun

Published 2026-09-03

📖 5 min read🧠 Deep dive

CC0 1.0

Original authors: Feng Fang, Kim Siang Khaw, Ce Zhang, Qiaoli Yang, Liangwen Chen, Jie Yang, Lei Yang, Zhiyu Sun

Original paper dedicated to the public domain under CC0 1.0 (http://creativecommons.org/publicdomain/zero/1.0/). ✨ 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

The universe is filled with a mysterious substance known as dark matter, which makes up about a quarter of everything that exists. While we cannot see it or touch it, its gravitational pull holds galaxies together, and scientists are certain it is not made of the ordinary atoms that form stars, planets, and people. Among the many theories about what this invisible stuff might be, one leading idea suggests it consists of extremely light, ghostly particles called axions. These particles would interact so weakly with normal matter that they pass through everything like a whisper through a wall, making them incredibly difficult to catch. Finding them would be a monumental discovery, potentially solving one of the biggest puzzles in modern physics.

To hunt for these elusive particles, a team of researchers has proposed a new kind of experiment using a rare, short-lived atom called muonium. Unlike the hydrogen atoms found in water or air, muonium is a tiny, hydrogen-like system made of a positive muon and an electron. Because muons are fundamental particles that do not feel the strong nuclear force, muonium is a perfectly clean laboratory for testing the laws of physics without the messy complications found in heavier atoms. The researchers suggest that if axions exist and have a specific mass, they could cause the muonium atom to flip its internal spin in a very specific way. By building a machine capable of watching for this flip, scientists hope to find direct evidence of axion dark matter in a controlled setting, offering a fresh perspective on a problem that has stumped physicists for decades.

The core of this proposal is a device designed to act as a highly sensitive sensor for these invisible particles. The experiment begins with a beam of muons, which are created at high-energy particle accelerators. These muons are slowed down and fired into a thin sheet of carbon. As they pass through the carbon, they grab electrons to form muonium atoms. These new atoms fly forward, carrying the memory of their original direction. The researchers then use a precise sequence of lasers and electric fields to prepare the muonium for observation. First, a laser tuned to a specific color of light removes any muonium atoms that are already in the "wrong" spin state, leaving behind only those in the desired starting position. This cleaning step is crucial because it ensures that any change in the atom's state later on is caused by something new, not by a mistake in the setup.

Once the muonium atoms are purified, they enter a long, quiet tunnel surrounded by a powerful magnetic field. This is the heart of the experiment. The magnetic field is carefully adjusted to create a specific energy gap between the two possible spin states of the muonium atom. If an axion from the dark matter halo of our galaxy happens to have a mass that matches this energy gap exactly, it will resonate with the atom. This resonance acts like a gentle nudge, causing the muonium to flip its spin. The researchers calculate that this interaction would happen at a rate proportional to how strongly the axion couples to the muon. Because the axion field is constantly oscillating, this flip would occur only when the magnetic field is tuned to the precise frequency of the axion's mass.

After passing through this magnetic tunnel, the atoms enter a second stage where they are hit by another set of lasers. These lasers are designed to strip the electron away from any muonium atom that has flipped its spin, turning it back into a free muon. The remaining neutral atoms, which did not flip, pass straight through. A powerful electric field then bends the path of the newly freed muons, steering them toward a detector while the neutral atoms continue on a different path. This separation allows the scientists to count only the muons that were part of a spin-flip event. If the detector sees a sudden spike in these counts when the magnetic field is set to a specific value, it would be a strong signal that an axion was present.

To ensure this idea works, the team ran detailed computer simulations of the entire process. They modeled the behavior of the muon beam, the formation of muonium in the carbon foil, and the journey through the magnetic and electric fields. Their calculations showed that with a high-intensity beam of muons, such as those expected from next-generation facilities in China and Europe, the experiment could detect a handful of these rare events over the course of a year. The simulations also confirmed that background noise from other particles, like stray muons or photons, could be filtered out effectively using time measurements and the electric field deflection. The team found that by tuning the magnetic field strength, they could search for axions with masses between 18 and 130 micro-electronvolts, a range that is currently difficult to explore with other methods.

The results of this study suggest that muonium spectroscopy offers a unique and powerful way to search for dark matter. Unlike previous experiments that relied on indirect measurements or complex astrophysical models, this approach looks for a direct, resonant interaction in a controlled laboratory environment. The team's work demonstrates that with the right equipment, it is possible to tighten the constraints on how axions interact with muons, potentially surpassing the limits set by current measurements of the muon's magnetic properties. While the experiment has not yet been built, the simulations provide a clear roadmap for how it could be constructed. If successful, this new channel of observation could open a window into a sector of physics that has remained largely hidden, bringing us one step closer to understanding the invisible fabric of the universe.

Technical Summary: Muonium Spectroscopy as a Quantum Sensor for Axion Dark Matter

Problem Statement
Approximately 27% of the universe's energy density is attributed to dark matter, yet its fundamental nature remains unknown. Among the leading candidates are axions and axion-like particles (ALPs). While couplings between ALPs and electrons or photons are tightly constrained, the axion-muon coupling ( $g_{a\mu}$ ) remains largely unexplored. Existing laboratory constraints on $g_{a\mu}$ are primarily derived from the muon $g-2$ anomaly, which relies on indirect virtual-loop effects and is subject to model-dependent interpretations. Astrophysical limits (e.g., from SN1987A) exist but depend heavily on complex modeling of supernova cores. Consequently, there is a critical need for a direct, laboratory-based probe that can search for ALP-muon interactions independently of astrophysical modeling or virtual-loop assumptions.

Methodology
The authors propose a novel experimental approach termed Muonium-based Axion Search (MASH), utilizing high-intensity muon beams to create muonium (a bound state of a positive muon and an electron). The methodology relies on detecting resonant quantum transitions between muonium hyperfine states induced by the oscillating ALP dark matter field.

Theoretical Mechanism:
- An external Zeeman magnetic field ( $B$ ) is applied to tune the energy splitting ( $\Delta E$ ) between the muonium hyperfine states $|0, 0\rangle$ and $|1, 1\rangle$ .
- When $\Delta E$ matches the ALP mass ( $m_a$ ), resonant transitions occur at a rate proportional to $g_{a\mu}^2$ .
- The transition rate is given by $R \propto g_{a\mu}^2 L^2 \rho_{DM}$ , where $L$ is the interaction length and $\rho_{DM}$ is the local dark matter density.
Experimental Design:
- Beam Preparation: Surface muons are moderated to keV energies and passed through a thin carbon foil to form muonium atoms.
- State Preparation (Cleaning): A dual-laser system (122 nm and 355 nm) selectively ionizes and removes atoms in the $|1, 1\rangle$ state, leaving a pure sample of $|0, 0\rangle$ ground-state atoms.
- Interaction Region: The purified muonium beam enters a solenoid magnet (the "scan" region) where the magnetic field is tuned to match specific ALP mass hypotheses.
- Detection: A second laser cavity selectively ionizes any muonium atoms that have transitioned to the $|1, 1\rangle$ state via ALP interaction. The resulting free muons are deflected by electric fields and detected by a high-efficiency Channel Electron Multiplier (D3), distinguishing them from background muons.
- Background Suppression: The design employs time-of-flight discrimination, electric field deflection to remove penetrating muons and positrons, and a differential "on/off-resonance" analysis strategy to suppress residual background from imperfect state cleaning.
Simulation:
- The experiment was modeled using Geant4 (via the musrSim package), simulating muon interactions, muonium formation, transport, and detection.
- Parameters were optimized for injection energy and magnetic field length ( $L$ ) to balance conversion efficiency against decay losses and beam divergence.

Key Contributions and Results

Feasibility Demonstration: The study validates the feasibility of using muonium spectroscopy as a direct probe for ALP-muon couplings, a sector previously inaccessible to colliders and precision muon experiments due to the pseudoscalar nature of the interaction.
Sensitivity Projections:
- Using projected high-intensity muon sources (e.g., 5 $\times$ 10 $^{10}$ $\mu^+$ /s from CiADS or HiMB), the experiment can achieve a slow muon rate of $\sim$ 3 $\times$ 10 $^6$ /s.
- For an interaction length of $L=10$ m and an injection energy of $\sim$ 14 keV, the setup is projected to detect $\sim$ 1 signal event per month for ALP masses of $\mathcal{O}(10^{-5})$ eV.
- Exclusion Limits: The projected sensitivity for a one-month run (Mu-I) approaches current limits derived from the muon $g-2$ discrepancy. With a three-year data set (Mu-II), the proposed direct ALP search experiment has the potential to surpass the current muon $g-2$ bound in the relevant parameter region, specifically in the axion mass range of 18–130 $\mu$ eV.
Background Control: The simulation indicates that systematic backgrounds (photons, penetrating muons, positrons) can be reduced to negligible levels through the proposed electric field deflection and spatial selection. The dominant background is residual $|1, 1\rangle$ atoms, which can be managed via resonance modulation.

Significance
The paper claims that this work establishes a new spectroscopic channel for searching for axion and axion-like particle dark matter. By leveraging the simple, purely leptonic structure of muonium, the proposed experiment avoids the complications of strong interaction effects found in hadronic systems. Crucially, it offers a direct, model-independent laboratory probe for the axion-muon coupling, complementing and tightening constraints beyond those achievable through indirect $g-2$ measurements or astrophysical observations. The availability of next-generation high-intensity muon facilities is identified as the key enabler for realizing this sensitivity.

Technical Summary: Muonium Spectroscopy as a Quantum Sensor for Axion Dark Matter

More like this