Dark photon dark matter constraints at the Taiwan axion search experiment with haloscope

By re-analyzing data from the Taiwan Axion Search Experiment with Haloscope (TASEH) while accounting for scanning timing information, the authors derive world-leading constraints on dark photon dark matter that exceed naive rescaling limits by a factor of two and demonstrate the critical importance of avoiding magnetic-field-only vetoes to prevent discarding valid dark photon signals.

The Gist

Imagine the universe is filled with a mysterious, invisible "fog" called Dark Matter. We know it's there because it holds galaxies together, but we can't see it, touch it, or smell it. Scientists have been trying to figure out what this fog is made of for decades.

One popular theory suggests the fog is made of tiny, invisible particles called Dark Photons. Think of these as "ghostly cousins" to the photons (light particles) that make up the sunlight and the light from your phone screen. These dark cousins are invisible, but they might occasionally whisper to normal light, causing a tiny, detectable ripple.

This paper is about a team of scientists in Taiwan (the TASEH team) who built a super-sensitive "radio" to listen for these whispers. Here is the story of what they found, explained simply:

1. The Radio and the Magnetic Field

To catch these whispers, the scientists used a device called a Haloscope. Imagine a giant, super-cold metal drum (a cavity) that acts like a tuning fork. If a Dark Photon flies by at just the right frequency, it will make the drum vibrate, creating a tiny spark of electricity that the scientists can measure.

• The Old Way (Axion Hunting): For years, scientists have used these drums to hunt for a different particle called an Axion. To catch an Axion, you need a giant magnet. The magnet acts like a "filter" or a "veto." If a signal appears only when the magnet is on, it's likely an Axion. If the signal appears even when the magnet is off, scientists usually throw it away, thinking it's just background noise.
• The New Problem: Dark Photons are different. They don't need a magnet to make the drum vibrate. So, if a scientist sees a signal when the magnet is off, they might mistakenly throw it away, thinking it's noise. But that signal could actually be the Dark Photon they are looking for!

2. The "Rescaling" Mistake

In the past, when scientists wanted to set limits on Dark Photons using data from Axion hunts, they just did a quick math trick called "rescaling." They took the Axion rules and applied them to Dark Photons.

The authors of this paper said, "Wait a minute! That's like using a recipe for a cake to bake a soufflé."

• The Polarization Issue: Axions are simple; they don't have a direction. Dark Photons, however, are like arrows; they have a direction (polarization). As the Earth spins, the direction of the "Dark Photon wind" changes relative to the detector.
• The Timing Issue: The old math assumed the experiment happened instantly. But the TASEH experiment ran for weeks. By carefully tracking when they took measurements and how the Earth rotated, the scientists realized the old math was too cautious. It was like assuming you only have one second to catch a fish, when in reality, you've been fishing for a month.

The Result: By doing the math correctly and accounting for the Earth's rotation, they found that their detector was actually twice as sensitive to Dark Photons as previously thought. They set a new, world-leading record, ruling out a huge chunk of possible Dark Photon "personalities" (specifically, those with a mass between 19.46 and 19.84 micro-electronvolts).

3. The "Ghost" Signal

During their data analysis, the team found something weird. In a specific frequency range (around 4.71 GHz), they saw a signal that was 4.7 times stronger than the usual background noise.

• The Clue: This signal appeared even when the giant magnet was turned off.
• The Interpretation: If this were an Axion, it would be a fake signal (noise). But because Dark Photons don't need a magnet, this signal looked exactly like a potential Dark Photon discovery!
• The Twist: The team did a "sanity check." They looked at data from other super-advanced experiments (HAYSTAC and ORGAN-Q) that had scanned the exact same area with even better equipment. Those experiments saw nothing.
• The Conclusion: The signal in the Taiwan experiment was likely a "ghost"—a glitch in their own equipment or a stray radio wave from a computer in the lab. It wasn't Dark Matter.

Why This Paper Matters

This paper teaches us two big lessons:

1. Don't throw away the weird stuff: If you are hunting for Dark Photons, you can't just use the "Magnet Off = Noise" rule. You have to look at the data with fresh eyes.
2. Details matter: By carefully tracking the timing and the rotation of the Earth, the scientists improved their sensitivity by a factor of two. This shows that in the search for the universe's biggest secrets, the devil is in the details.

In a nutshell: The scientists in Taiwan tuned their super-sensitive radio, realized they had been using the wrong instructions for the job, and found they could hear the "Dark Photon" whispers much better than anyone else. They also found a "ghost" signal that looked promising but turned out to be a false alarm, proving that we need to be extra careful and cross-check our findings with other teams before declaring victory.

Technical Summary

1. Problem Statement

The search for Dark Photon Dark Matter (DPDM) often relies on reinterpreting data from axion search experiments using "naïve rescaling" methods. This approach assumes that the conversion limits for axions can be directly translated to dark photons by adjusting for coupling constants and magnetic field dependencies. However, this paper identifies two critical flaws in this standard practice:

• Polarization Effects: Unlike axions (pseudoscalars), dark photons are spin-1 vector bosons with polarization. The signal strength in a cavity haloscope depends on the angle between the dark photon's polarization vector and the cavity's sensitive axis. Naïve rescaling often assumes an instantaneous measurement or ignores the time-averaged effect of Earth's rotation, leading to overly conservative or inaccurate bounds.
• Magnetic Field Veto Reliance: Axion searches typically use a "magnetic field veto" (checking if a signal disappears when the magnetic field is turned off) to reject instrumental noise. Since dark photon conversion does not require a magnetic field, valid DPDM signals would persist even without the field. Relying on this veto causes experiments to discard genuine dark photon candidates as noise.

2. Methodology

The authors re-analyzed the CD102 dataset from the Taiwan Axion Search Experiment with Haloscope (TASEH), collected between October and November 2021. The analysis involved:

• Theoretical Framework:

  • They modeled the dark photon signal power ($P_{dp}$) in a cavity, incorporating the kinetic mixing parameter ($\epsilon$) and the time-averaged geometric factor $\langle \cos^2 \theta(t) \rangle_T$.
  • They distinguished between two scenarios: Random Polarization (isotropic distribution) and Fixed Polarization (a specific orientation relative to the galactic halo).
  • They derived a rigorous conversion factor to translate axion exclusion limits ($g_{a\gamma}$) into dark photon limits ($\epsilon$), explicitly accounting for the scanning timing and the Earth's rotation ($\omega$).

• Data Processing:

  • Signal Extraction: Raw power spectra were normalized using a Savitzky-Golay (SG) filter to create a Relative Deviation of Power (RDP) spectrum.
  • Combination: Data from 837 tuning steps (scans) were combined using a weighted algorithm to create a single merged spectrum.
  • Re-scanning: Frequency ranges showing initial excesses ($>3\sigma$) were re-scanned 75 additional times to distinguish real signals from statistical fluctuations or instrumental artifacts.

• Case Study Analysis:

  • The team specifically investigated a frequency range ($4.71017 - 4.71019$ GHz) where a signal persisted after the magnetic field was turned off. They analyzed this excess under the dark photon hypothesis to demonstrate how such signals are often missed in standard axion analyses.

3. Key Contributions

• Refined Conversion Formalism: The paper provides a practical, rigorous method for converting axion limits to dark photon limits by incorporating scanning timing information. This corrects the assumption of instantaneous measurements used in previous "AxionLimits" rescalings.
• Polarization Sensitivity: It demonstrates that for fixed polarization scenarios, the exclusion limit is significantly stronger (by a factor of $\sim 2$) when the time-averaged geometric factor is correctly calculated compared to naïve rescaling.
• Veto Critique: It highlights the danger of using magnetic field vetoes for dark photon searches, showing that valid candidates can be erroneously discarded if they mimic the behavior of non-magnetic signals.

4. Results

• New World-Leading Constraints:

  • In the mass range 19.46 – 19.84 $\mu$eV, the study excludes kinetic mixing values of $|\epsilon| \gtrsim 2 \times 10^{-14}$ at 95% confidence level.
  • This limit is approximately twice as strong as the previous "rescaling limit" derived from the same TASEH axion data.
  • The improvement is attributed to the correct modeling of the polarization factor $\langle \cos^2 \theta(t) \rangle_T \approx 0.1$ (for TASEH's specific latitude and scan duration), which is significantly higher than the $\sim 0.02$ assumed in generic short-measurement models.

• Analysis of the 4.71 GHz Excess:

  • A tentative signal was observed at 4.710178 GHz with a local significance of 4.76$\sigma$ (kinetic mixing $|\epsilon| \approx 6.48 \times 10^{-15}$).
  • Crucially, this signal persisted without the magnetic field, mimicking a dark photon signature.
  • However, cross-referencing with recent results from HAYSTAC and ORGAN-Q (which probed this frequency with superior sensitivity and found no signal) led the authors to conclude this excess is an instrumental artifact.
  • This serves as a cautionary case study: while the signal was likely noise, it demonstrated that a dedicated dark photon analysis (without magnetic vetoes) is necessary to properly evaluate such candidates.

5. Significance

• Methodological Advancement: The paper establishes that re-analyzing axion data with dedicated dark photon physics (specifically polarization and timing) yields significantly tighter constraints than simple rescaling.
• Experimental Guidance: It underscores the necessity for future haloscope experiments to:
  1. Record precise timing data to model polarization effects.
  2. Avoid discarding candidates solely based on magnetic field independence.
  3. Perform dedicated re-analyses of existing datasets rather than relying on automated rescaling tools.
• Physics Impact: The new constraints push the boundary of the excluded parameter space for dark photon dark matter, ruling out a region of parameter space that was previously thought to be accessible only with more sensitive future experiments.

In summary, this work bridges the gap between axion and dark photon searches, proving that existing data can provide world-leading constraints on dark photons if analyzed with the correct physical assumptions regarding polarization and experimental timing.