⚛️ phenomenology

Dark Photons in the Radio Sky: II. Resonant Conversions in the Intergalactic Medium

This paper details the analysis pipeline for forecasting the Square Kilometre Array's sensitivity to dark photons via resonant conversions in the intergalactic medium, demonstrating that combining SKA with galaxy surveys and 21-cm experiments could detect dark photons with masses between $5\times 10^{-15}$ and $5\times 10^{-12}$ eV and kinetic mixing parameters as low as $10^{-8}$ .

Original authors: Ethan Baker, Hongwan Liu

Published 2026-09-18

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ethan Baker, Hongwan Liu

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 universe is filled with a faint, ancient afterglow known as the cosmic microwave background. This radiation is the oldest light in existence, a remnant from the moment the universe cooled enough for atoms to form, roughly 380,000 years after the Big Bang. For decades, scientists have studied this light with extreme precision, using it as a backlight to understand the history and composition of the cosmos. However, this light might also be hiding a secret. Physicists have long suspected that there is a hidden sector of particles that do not interact with normal matter, except perhaps through a very subtle, ghostly connection. One of the simplest and most compelling ideas for such a hidden particle is the "dark photon." Unlike the photons that make up visible light and radio waves, dark photons would be invisible to our eyes and most detectors, but they might occasionally swap places with normal photons as they travel through space. If this swapping happens, it would leave a tiny, specific fingerprint on the cosmic microwave background, a signal that has so far remained out of reach.

A team of researchers at Boston University has now mapped out a detailed plan to find this fingerprint using the most powerful radio telescope ever built, the Square Kilometre Array, or SKA. In a new study, they describe how this massive instrument, currently under construction, could detect the presence of dark photons by looking for specific patterns in the radio sky. The researchers focused on a process where normal photons from the cosmic microwave background convert into dark photons as they pass through the vast, thin gas that fills the space between galaxies. This conversion does not happen everywhere; it occurs only at specific locations where the density of electrons in the gas matches a precise condition related to the mass of the dark photon. When this match happens, the normal photon disappears from our view, effectively turning into a dark photon that slips away undetected. This disappearance creates tiny, cold patches in the temperature of the radio sky, distinct from the usual fluctuations caused by the distribution of matter in the universe.

To see if these patches could be found, the researchers built a sophisticated simulation pipeline. They started by modeling three different cosmic environments where this conversion could occur: inside the dense clouds of gas surrounding galaxies, in the thin intergalactic gas during the early era when the first stars were turning on, and in the thin gas of the nearby, modern universe. For each environment, they calculated exactly what the signal would look like for dark photons of different masses. They then simulated what the SKA would actually see, including the overwhelming noise from our own galaxy and distant radio sources that usually drown out faint cosmic signals. To separate the signal from this noise, they employed a statistical technique that combines data from multiple radio frequencies. Since the dark photon signal has a unique frequency signature that differs from the background noise, this method allows the researchers to isolate the faint dark photon pattern, effectively filtering out the interference.

The results of this analysis are promising. The study shows that the SKA, when combined with surveys of galaxies, could detect dark photons with masses between 5 × 10⁻¹⁵ and 5 × 10⁻¹² electron volts. This range covers a mass scale that previous experiments, such as those using the Planck satellite, could not probe effectively. The researchers found that the SKA would be sensitive enough to detect a conversion rate as low as one in a hundred million, a level of sensitivity that would allow it to discover dark photons if they exist within this mass range. The key to this success is the telescope's ability to resolve very small details in the sky. Unlike earlier instruments that saw the radio sky as a blurry smear, the SKA can distinguish fine structures, which is crucial because the dark photon signal creates small, localized cold spots. Additionally, the telescope's ability to identify and remove bright radio sources from other galaxies further clears the view, making the faint signal easier to spot.

The researchers also looked at how this search could be performed using other types of radio experiments that measure the "global" signal of the universe, which is the average brightness of the sky at different times in cosmic history. They found that these experiments could also be very sensitive to dark photons, potentially detecting them even if the SKA does not. However, they noted that searching for the signal in the detailed fluctuations of the radio sky, known as the power spectrum, is much more difficult for current instruments because the signal gets mixed up with other foreground noise in a way that is hard to untangle. Despite this challenge, the overall picture is one of great potential. The study demonstrates that radio astronomy is uniquely positioned to explore this hidden sector of physics, offering a new and powerful way to test the limits of our understanding of the universe. By turning the entire sky into a laboratory, the SKA could finally reveal whether the dark photons that physicists have theorized for decades are actually real.

Technical Summary: Dark Photons in the Radio Sky: II. Resonant Conversions in the Intergalactic Medium

Problem Statement
This work addresses the search for dark photons ( $A'$ ), a hypothetical gauge boson associated with a new dark $U(1)'$ symmetry that kinetically mixes with Standard Model photons via a parameter $\epsilon$ . Specifically, the paper focuses on the resonant conversion of Cosmic Microwave Background (CMB) photons into dark photons ( $\gamma \to A'$ ) as they propagate through the Intergalactic Medium (IGM). These conversions occur when the dark photon mass $m_{A'}$ matches the effective plasma mass $m_\gamma$ of the medium, leading to a disappearance of CMB photons. This process generates distinct temperature anisotropies (cold patches) in the radio sky and correlates with the distribution of baryons and galaxies. While previous studies (e.g., Refs. [80, 81]) have constrained these conversions in dark matter halos using Planck data, this paper investigates the sensitivity of the upcoming Square Kilometre Array (SKA) and 21-cm experiments to conversions occurring within the IGM, particularly during the Epoch of Reionization (EoR) and in the late Universe.

Methodology
The authors develop a comprehensive analysis pipeline to forecast the sensitivity of radio experiments to $\gamma \to A'$ conversions. The methodology is divided into signal modeling, mock observation generation, and statistical analysis.

Signal Modeling: The paper models three distinct cosmological environments where resonant conversions occur:
- Environment A (Dark Matter Halos): Conversions in halos at $0.005 \lesssim z \lesssim 4$ . This utilizes the halo model formalism (Paper I) to compute angular power spectra ( $C_\ell$ ) and cross-correlations with galaxy surveys.
- Environment B (EoR IGM): Conversions in the IGM during the EoR ( $5 \lesssim z \lesssim 35$ ). The authors use the semi-numerical code 21cmFAST to generate 3D light cones of the IGM, calculating the plasma mass $m_\gamma$ and determining conversion probabilities along lines of sight. This yields a sky map of the temperature deficit $\Delta T$ .
- Environment C (Late-Time IGM): Conversions in the low-redshift IGM ( $0.005 \lesssim z \lesssim 4$ ). Due to the complexity of hydrodynamical simulations, the authors defer a rigorous simulation-based treatment to future work; instead, they estimate this signal using a lognormal Ansatz for the probability distribution function (PDF) of $m_\gamma$ , following Refs. [62, 110].
Mock Observations and Foreground Removal:
- The authors generate nine mock radio maps at SKA-Mid frequencies (0.41–12.53 GHz) containing realistic astrophysical foregrounds: galactic synchrotron, free-free, spinning dust, and unresolved extragalactic point sources. These are modeled using pysm3 and epspy.
- Thermal noise is added based on LOFAR Deep Field sensitivity levels.
- To extract the signal, the authors employ a Needlet Internal Linear Combination (NILC) algorithm. This technique constructs a minimum-variance linear combination of the multi-frequency maps, constrained to preserve the spectral shape of the dark photon signal (scaling as $\omega^{-1}$ ), effectively suppressing foregrounds.
- Mock galaxy catalogs are generated for cross-correlation: the "K22" Halo Occupation Distribution (HOD) model for low-redshift surveys (Environment A) and 21cmFAST outputs for high-redshift Roman Space Telescope surveys (Environment B).
Statistical Analysis:
- Sensitivity is forecasted by comparing theoretical predictions (autocorrelation functions $\xi_{TT}$ and cross-correlations $\xi_{Tg}$ ) against the "null-signal" mock observations (foregrounds only).
- Likelihood functions are constructed assuming Gaussian statistics. The covariance matrices are estimated using bootstrap resampling (500 iterations) to account for mode mixing and sky masking.
- Upper limits on the kinetic mixing parameter $\epsilon$ are derived at the 95% confidence level.

Key Contributions

Pipeline Development: The paper details the first complete analysis pipeline specifically designed to forecast SKA sensitivity to dark photon conversions in the IGM, including the application of NILC to radio data for this specific spectral signature.
IGM Signal Characterization: It provides detailed theoretical predictions for $\gamma \to A'$ conversions in the EoR (Environment B) using 21cmFAST simulations, and preliminary estimates for late-time IGM conversions (Environment C) using a lognormal PDF approximation, noting the latter requires future hydrodynamical simulations for rigor.
Cross-Correlation Strategy: The work emphasizes the utility of cross-correlating radio sky maps with galaxy surveys. It demonstrates that the cross-correlation signal scales as $\epsilon^2$ , offering potentially less suppression at small $\epsilon$ compared to the autocorrelation signal ( $\epsilon^4$ ).
21-cm Global Signal Analysis: The paper extends the analysis to 21-cm global signal experiments, estimating the impact of $\gamma \to A'$ conversions on the sky-averaged brightness temperature.

Results

Sensitivity Reach: The authors forecast that SKA, combined with galaxy surveys, can discover dark photons with masses in the range $5 \times 10^{-15} \text{ eV} \lesssim m_{A'} \lesssim 5 \times 10^{-12} \text{ eV}$ and kinetic mixing parameters as low as $\epsilon \sim 10^{-8}$ .
Environment-Specific Limits:
- Environment A (Halo): Cross-correlation with low-redshift galaxy surveys improves upon existing Planck-unWISE limits by a factor of 4 for masses around $10^{-12}$ eV.
- Environment B (EoR): SKA cross-correlated with the Roman galaxy survey is sensitive to $\epsilon \sim 5 \times 10^{-7}$ for masses around $2 \times 10^{-14}$ eV, slightly exceeding current limits from CMB spectral distortions.
- Environment C (Late IGM): Preliminary estimates based on a lognormal Ansatz suggest SKA could surpass FIRAS limits for masses between $5 \times 10^{-15}$ and $10^{-13}$ eV, motivating a more detailed study using hydrodynamical simulations.
21-cm Power Spectrum: The paper concludes that searching for dark photons in 21-cm power spectrum experiments (e.g., HERA) is significantly complicated. The signal's smooth spectral dependence causes it to leak into the "wedge" region of the power spectrum, which is typically excised to remove foregrounds, thereby removing the signal of interest.
Global Signal: Conversely, 21-cm global signal experiments could be highly sensitive, provided the signal can be disentangled from foregrounds that share a similar frequency dependence.

Significance and Claims
The paper claims that upcoming radio experiments, particularly SKA, offer a powerful and unique avenue for probing ultralight dark photons, surpassing current constraints from CMB spectral distortions and anisotropy measurements. The authors highlight that the superior angular resolution of SKA (compared to CMB experiments like Planck) allows for the resolution of small-scale physics and more effective point source subtraction, which are critical for detecting the faint anisotropies generated by $\gamma \to A'$ conversions.

The work underscores the potential of radio astronomy to study physics beyond the Standard Model, specifically noting that the combination of high-resolution imaging and cross-correlation with galaxy surveys provides a robust method to detect resonant conversions in the IGM. The authors remain modest regarding the 21-cm power spectrum results, acknowledging that experimental limitations (the wedge) currently hinder detection, and they call for future work to explore alternative analysis strategies for that data. Similarly, the results for Environment C are presented as preliminary estimates motivating more rigorous hydrodynamical simulations.

More like this