⚛️ high-energy experiments

Search for Dark Photon Dark Matter at 74--159 $\mu$ eV/ $c^2$ Using Broadband Cryogenic Receivers

Using broadband cryogenic K- and Ka-band receivers, this study searched for dark photon dark matter in the 74–159 $\mu$ eV/ $c^2$ mass range, finding no significant signal and setting new upper limits on the kinetic mixing parameter that improve upon previous constraints by over an order of magnitude in the 74–110 $\mu$ eV/ $c^2$ range while providing the first direct limits for the 110–159 $\mu$ eV/ $c^2$ range.

Original author: Hiroki Takeuchi, Shunsuke Adachi, Shunki Endo, Keigo Hanimura, Michinori Hirose, Shunsuke Honda, Yuki Sakurai, Junya Suzuki, Osamu Tajima

Published 2026-11-10

📖 4 min read🧠 Deep dive

CC BY 4.0

Original author: Hiroki Takeuchi, Shunsuke Adachi, Shunki Endo, Keigo Hanimura, Michinori Hirose, Shunsuke Honda, Yuki Sakurai, Junya Suzuki, Osamu Tajima

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 author. For technical accuracy, refer to the original paper. Read full disclaimer

For decades, astronomers have known that the universe is filled with far more matter than we can see. Stars orbit too fast, and galaxies hold together too tightly, to be explained by the visible stars and gas alone. This invisible substance, called dark matter, makes up most of the mass in the cosmos, yet no one has ever directly detected a single particle of it. While gravity reveals its presence, the microscopic nature of dark matter remains one of the greatest mysteries in physics. One leading idea suggests that dark matter might be made of a hypothetical particle called the dark photon. Unlike the ordinary photons that make up light, these dark photons would have a tiny amount of mass and could occasionally transform into regular light when they hit a metal surface. If this transformation happens, it would create a faint, steady hum of radio waves at a specific frequency determined by the particle's mass. Finding this hum would not only reveal what dark matter is but also open a new window into the fundamental laws of the universe.

A team of researchers in Japan, working under the name DOSUE-RR, set out to listen for this faint signal across a wide range of frequencies. They focused on a specific mass range for the dark photon, between 74 and 159 micro-electron volts per speed of light squared, which corresponds to radio waves between 18 and 38.5 gigahertz. To catch this signal, they built a highly sensitive listening station using two specialized receivers, one tuned to the lower part of this range and the other to the higher part. These receivers were cooled to extremely low temperatures to reduce their own internal noise, allowing them to hear the faintest whispers from the cosmos. The setup involved a metal plate facing a horn antenna; if dark photons were passing through the room, the theory predicted they would convert into ordinary radio waves upon hitting the metal plate and be collected by the antenna. The team used a broadband spectrometer, a device capable of analyzing a vast stretch of the radio spectrum all at once, rather than tuning to one narrow frequency at a time. This approach allowed them to scan a much wider area of the search space than previous experiments could manage in a single run.

The experiment ran for months, accumulating hundreds of thousands of seconds of observation time for each frequency point they tested. The researchers carefully monitored their equipment to ensure that any signal they found was truly from the sky and not from local interference, such as radio clocks or electronic glitches. They even tested their system by replacing the metal plate with a room-temperature absorber to see what the background noise looked like without the potential for dark photon conversion. After sifting through the data and removing any frequencies contaminated by known sources of interference, they analyzed the remaining signal for any sign of the expected dark photon hum. The result was clear: they found no statistically significant signal. The data showed no excess of radio waves that could be attributed to dark photons converting at the metal surface.

Despite the lack of a discovery, the experiment provided a crucial advance in the search for dark matter. By finding nothing, the team was able to set strict new limits on how strongly dark photons could interact with ordinary light. They determined that if dark photons exist in this mass range, their ability to mix with regular photons must be incredibly weak, far weaker than many previous theories had allowed. Their results improved upon earlier limits by more than ten times for the lower part of their search range and provided the first direct constraints for the higher part of the range, pushing the boundaries of what is known about these particles. The findings suggest that if dark photons make up all the dark matter in our local neighborhood, their interaction with the visible world is so faint that it remains beyond the reach of current technology, or they simply do not exist in the quantities or forms the researchers were looking for. This work narrows the field of possibilities, guiding future experiments toward new strategies or different mass ranges in the ongoing quest to understand the invisible universe.

Technical Summary: Search for Dark Photon Dark Matter at 74–159 µeV/c²

Problem and Motivation
Dark matter's microscopic nature remains unknown despite astrophysical evidence of its gravitational effects. The dark photon, a hypothetical massive vector boson that kinetically mixes with ordinary photons, is a viable dark matter candidate. If produced by quantum fluctuations during inflation, dark photon dark matter (DPDM) could account for the full local dark matter abundance, requiring masses on the order of 100 µeV/c² for inflationary scales near current cosmic microwave background limits. While cosmological observations constrain DPDM over a broad mass range, direct searches have historically been limited to narrow mass windows due to the bandwidth constraints of resonant enhancement techniques. Consequently, most masses above 110 µeV/c² lacked direct experimental limits.

Methodology
The DOSUE-RR Collaboration conducted a broadband search for DPDM in the mass range of 74–159 µeV/c² (corresponding to frequencies of 18–38.5 GHz). The experiment utilized the conversion of DPDM into ordinary photons at a metal surface via kinetic mixing ( $\chi$ ). The apparatus consisted of a metal plate facing a horn antenna, where conversion photons are emitted nearly perpendicular to the surface. The signal frequency $\nu$ satisfies $h\nu \simeq m_{DP}c^2$ .

Key experimental components included:

Receivers: A cryogenic system employing both K-band (18–26 GHz) and a new Ka-band (26–38.5 GHz) receiver. System noise temperatures ranged from 40–100 K (K-band) to 60–120 K (Ka-band).
Spectrometer: The experiment utilized dSpec, a broadband spectrometer developed to increase integration time per frequency bin by a factor of 2,000 compared to previous spectrum analyzers used by the group.
Observation Strategy: Measurements were performed at the zenith with horns oriented at an azimuth of ~165°. The search covered the frequency range using local oscillator (LO) pairs separated by 20 MHz to distinguish spurious signals. Total integration time was approximately $4.8 \times 10^5$ seconds per frequency bin.
Signal Model: The expected signal power $P_{DP}$ is proportional to $\chi^2$ , the local dark matter density $\rho$ , the effective detection area $A_{eff}$ , and a polarization factor $\alpha$ . The signal is expected to have a fractional linewidth of $\Delta\nu/\nu \sim 10^{-6}$ due to Galactic velocity dispersion.
Data Analysis: Spectra were analyzed by fitting a linear background to the expected signal shape (convolved with the spectrometer's power response). Statistical uncertainties were derived from "null spectra" generated by pairing and sign-flipping consecutive measurements. Spurious signals originating from the mixer, LO, or clock harmonics were identified and excluded.

Key Contributions and Results
The search yielded no statistically significant signal. The analysis established the following results:

Upper Limits: The experiment set 95% confidence-level upper limits on the kinetic mixing parameter $\chi$ in the range of $(3.1\text{--}17) \times 10^{-12}$ over 74–159 µeV/c².
Coverage: These limits cover 99.8% of the search range, assuming DPDM constitutes all local dark matter.
Systematic Uncertainties: The total systematic uncertainty in $\chi$ was evaluated to be between 5.4% and 25.4%, dominated by uncertainties in the effective detection area ( $A_{eff}$ ) and gain calibration.
Spurious Signal Handling: Frequency intervals contaminated by identified spurious signals (e.g., clock harmonics) were excluded, leaving 99.8% of the search range valid for the final limits.

Significance
The paper claims that these results represent a significant improvement over previous direct searches:

Improved Sensitivity: Over most of the 74–110 µeV/c² range, the new limits improve upon the collaboration's previous results by more than an order of magnitude.
First Direct Limits: For the majority of the 110–159 µeV/c² range, this work provides the first direct experimental limits.
Comparison to Theory: The achieved limits are approximately two orders of magnitude below existing cosmological bounds in the 110–159 µeV/c² region, demonstrating the capability of broadband cryogenic receivers to probe parameter space previously inaccessible to direct detection methods.

Technical Summary: Search for Dark Photon Dark Matter at 74–159 µeV/c²

More like this