⚛️ phenomenology

Cosmic Axion Background Detection Using Resonant Cavity Arrays

This paper proposes a strategy for detecting the broadband Cosmic Axion Background using multi-cavity arrays that exploit spatial correlations of the axion-induced electric field, finding that while current ADMX upgrades lack coherent enhancement, optimized stacked, wide-base geometries could improve sensitivity by an order of one.

Original authors: Soobeom Chung, Jeff A. Dror

Published 2026-03-16

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Soobeom Chung, Jeff A. Dror

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

The Big Picture: Hunting for a Ghostly Wind

Imagine the universe is filled with a "wind" made of invisible particles called axions.

Dark Matter Axions: Most scientists are looking for axions that are heavy and slow, like a thick, slow-moving fog. These are the candidates for "Dark Matter."
The Cosmic Axion Background (CaB): This paper is about a different kind of axion. These are light, fast, and zipping around at nearly the speed of light. They are like a high-speed, chaotic breeze that has been blowing since the Big Bang.

The problem? This "wind" is broadband. Imagine trying to hear a specific whisper in a room where a jet engine is roaring. The jet engine (the background noise) is so loud and covers so many frequencies that it drowns out the whisper. Because the axion wind is so chaotic and spread out, it's incredibly hard to distinguish from the static noise in our detectors.

The Old Tool: The Tuning Fork

Scientists use devices called resonant cavities to catch these axions.

The Analogy: Think of a resonant cavity like a tuning fork. If you have a specific note (frequency) in the air, the tuning fork will vibrate loudly. If the air is silent or has a different note, the fork stays still.
The Filter: These cavities act as a very strict filter. They only "listen" to a tiny, narrow slice of the universe's noise. This helps block out some of the jet engine noise, but the axion wind is still so messy that it's hard to tell if the fork is vibrating because of the wind or just random static.

The New Idea: The Choir of Forks

The authors of this paper propose a clever new strategy: Don't use one tuning fork; use an array of them.

Imagine you are in a large field trying to hear that whisper.

The Single Listener: If you have one person (one cavity) listening, they might hear a noise and think, "Was that the whisper or just a leaf rustling?" It's a guess.
The Choir: Now, imagine you have 18 people standing in a specific pattern, all listening at the same time.
- The Trick: The "axion wind" has a special property. Even though it's chaotic, if two people stand close enough together, the wind will hit them in a correlated way. The wind pushes on Person A and Person B in a synchronized pattern.
- The Noise: Random background noise (like the rustling leaves or electronic static) is different for every person. It's uncorrelated.

By comparing the signals from all the forks (cavities) at once, the scientists can look for the synchronized pattern. If all the forks vibrate in a specific, coordinated dance, it's the axion wind. If they are all vibrating randomly, it's just noise.

The Geometry Problem: Stacking vs. Spreading

The paper does a lot of math to figure out the best shape for this "choir."

The "Fat" Cavity: The authors found that the cavities need to be "fat" (wide and short) rather than "tall and skinny."
- Analogy: Imagine trying to catch rain. A wide, shallow bucket catches the rain better than a tall, narrow pipe because the rain is falling from all angles. Similarly, these "fat" cavities catch the axion wind better.
The Stacked Array: The best arrangement they found is stacking these fat cavities vertically, like a tower of pancakes.
- Why? When you stack them, they are close enough that the axion wind hits them in perfect sync. This creates a "coherent" signal where the whole tower vibrates together, making the signal much louder.
The Current Plans (ADMX): The paper looks at the current plans for the ADMX experiment (a real-world axion hunter). They plan to use a flat array of 18 cavities (like a honeycomb).
- The Verdict: The authors say this is good, but not perfect. Because they are spread out flat, they don't get the full "coherent boost" that the stacked tower would get. However, it's still better than nothing and could improve sensitivity by a factor of 2 or so.

The Bottom Line

This paper is a blueprint for how to build a better "axion microphone."

The Challenge: The axion background is too noisy and broad to be heard by a single detector.
The Solution: Use a team of detectors (an array) and look for the synchronized dance between them.
The Secret Sauce: The arrangement matters! Stacking "fat" cavities on top of each other is the most efficient way to amplify the signal, turning a faint whisper into a shout that we can finally hear.

While the current experiments (like ADMX) aren't using the perfect stacked geometry, this research shows that by tweaking the shape and arrangement of the detectors, we can significantly boost our chances of finding this elusive cosmic wind. It's about turning a solo performance into a powerful, synchronized choir.

1. Problem Statement

The detection of the Cosmic Axion Background (CaB) presents a unique challenge compared to the search for axion dark matter.

Nature of the Signal: Unlike axion dark matter, which is non-relativistic and produces a narrowband signal (relative bandwidth $\sim 10^{-6}$ ), the CaB consists of relativistic axions produced in the early universe (via thermal collisions, inflationary resonance, or topological defects). Consequently, the CaB has a broadband energy distribution (relative bandwidth $\sim \mathcal{O}(1)$ ).
Detection Challenge: Standard resonant microwave cavities (like those used in ADMX) act as narrowband filters. Because the CaB is broadband, the signal power is spread out, and the intrinsic coherence time and length of the raw CaB field are extremely short.
Limitations of Current Methods: Previous attempts to detect CaB relied on anisotropy (e.g., dark matter decay producing a directional flux) to modulate the signal as the Earth rotates. This method fails for an isotropic CaB. Furthermore, the broadband nature makes it difficult to distinguish the signal from thermal noise using a single detector.

2. Methodology

The authors propose a new strategy utilizing multi-cavity arrays to exploit spatial correlations in the axion-induced electric field, rather than relying on anisotropy.

Theoretical Framework

Relativistic Form Factor: The authors derive the two-point correlation function for the electric field coefficients in spatially separated cavities. They introduce a relativistic form factor ( $F_{ij}$ ) that encodes the geometric dependence of the signal correlation between cavity $i$ and cavity $j$ .
Role of the Quality Factor ( $Q$ ): A key insight is that while the raw CaB field is incoherent over experimental scales, the cavity resonance acts as a narrowband filter. The filtering process extends the coherence length of the measured signal to be set by the cavity's quality factor ( $Q$ ) rather than the axion's intrinsic bandwidth. This allows the filtered signals in nearby cavities to remain correlated even if the source is broadband.
Likelihood Analysis: The authors construct a likelihood function for an $N$ -cavity array. They define a covariance matrix that includes both the uncorrelated background noise and the correlated signal terms (proportional to $F_{ij}$ ). They use the Asimov dataset approach to project 95% confidence level upper limits on the axion energy density ( $\Omega_0$ ).

Geometric Optimization

The study evaluates how cavity geometry affects the correlation signal:

Fat Cavities ( $L \ll R$ ): Theoretical limits suggest "fat" cavities (short height, large radius) maximize sensitivity.
Vertical Stacking: Cavities stacked vertically along the magnetic field axis ( $\hat{z}$ ) show strong correlations.
Coplanar Arrays: Cavities arranged side-by-side (like current ADMX configurations) show weaker correlations that decay rapidly with separation.

3. Key Contributions

Derivation of Correlation Functions: The paper provides the first rigorous derivation of the two-point correlation function for relativistic axion fields in resonant cavities, explicitly accounting for the cavity's transfer function and quality factor.
Identification of Coherent Enhancement: The authors demonstrate that multi-cavity arrays can achieve coherent signal enhancement if the geometry is optimized (specifically, vertically stacked "fat" cavities), whereas standard coplanar arrays behave incoherently.
Scaling Laws: They establish the scaling behavior of sensitivity ( $\Omega_0$ $Ω_{0}$ ) with the number of cavities ( $N$ $N$ ):
- Incoherent (Coplanar): Sensitivity scales as $N^{-1/2}$ .
- Coherent (Vertically Stacked): Sensitivity can scale as $N^{-1}$ initially, though it saturates as the array grows due to geometric constraints.
Application to ADMX: The formalism is applied to current and future ADMX configurations (Run-2A and the proposed EFR upgrade), quantifying their projected limits on the CaB.

4. Results

Form Factor Behavior: The relativistic form factor $F_{ij}$ oscillates and decays with cavity separation. For vertically stacked cavities, the correlation remains significant even at larger separations compared to coplanar arrangements.
Sensitivity Projections:
- ADMX Run-2A (4 cavities): Due to its coplanar geometry and large aspect ratio ( $L/R \approx 12$ ), it operates in an incoherent regime. The sensitivity improvement over a single cavity is minimal ( $\sim \sqrt{N}$ ).
- ADMX-EFR (18 cavities): Despite having more cavities and a stronger magnetic field, the hexagonal coplanar arrangement still lacks coherent enhancement. The sensitivity gain is limited by the geometry.
- Optimized Vertical Arrays: A hypothetical setup of a 1-meter tall cavity partitioned into many "fat" segments (vertically stacked) shows the best potential. In the coherent limit, sensitivity improves significantly with $N$ until saturation.
Parameter Dependencies: The projected upper limit on the energy density $\Omega_0$ scales as:
$\Omega_0 \propto B^{-2} \cdot T_s \cdot Q^{-1} \cdot N^{-\alpha}$
where $\alpha$ is $0.5$ for incoherent arrays and approaches $1.0$ for coherent arrays.
Cosmological Reach: To reach the cosmologically interesting regime ( $\Omega_0 \sim 10^{-5}$ , constrained by CMB $\Delta N_{eff}$ and BBN), current or near-future ADMX upgrades are insufficient. Achieving this would require extreme parameters (e.g., $B \approx 20$ T, $Q \approx 10^6$ , $T_s \approx 47$ mK) or a fundamentally different geometry.

5. Significance

New Detection Paradigm: This work shifts the focus from single-detector anisotropy searches to spatial correlation in multi-detector arrays, offering a viable path to detect isotropic broadband CaB signals.
Geometric Constraints: It highlights a critical limitation of conventional microwave cavity arrays: the correlation length is tied to the cavity size and aspect ratio. Standard "tall" cavities suppress the relativistic form factor.
Future Directions: The authors suggest that Superconducting Radio-Frequency (SRF) cavities or LC resonators (which can achieve $Q \sim 10^{10}-10^{12}$ and are not strictly bound by geometric mode structures) may be better suited for CaB detection. These systems could maintain phase coherence over larger distances without the mechanical constraints of stacked microwave cavities, potentially allowing for more favorable scaling with detector numbers.

In summary, the paper establishes that while multi-cavity arrays offer a theoretical pathway to detect the Cosmic Axion Background via spatial correlations, current experimental geometries (like ADMX) are suboptimal. Significant gains require either a transition to vertically stacked "fat" cavity configurations or the adoption of high- $Q$ non-microwave resonators.