X-ray signals converted from high-frequency gravitational waves emitted by spinning light primordial black hole dark matter

The paper investigates the detectability of high-frequency gravitational waves (GWs) originating from the superradiance of light primordial black hole (PBH) dark matter, specifically through their conversion into X-ray photons in the Galactic magnetic field. The authors find that the signal is significantly enhanced in the X-ray frequency range around 10 18 Hz. For clustered initial conditions, future X-ray observations may detect the converted photons from primordial black holes in the mass range 10−15 M⊙ ∼ 10−13 M⊙. The results indicate that future X-ray observations could provide a new probe of light primordial black hole dark matter through high-frequency gravitational waves.

The study focuses on PBHs with masses in the range 10−15 M⊙ ∼ 10−11 M⊙, which can account for the entire DM abundance according to current observational constraints. The mechanism considered is superradiance: "A bosonic field such as an axion-like particle can cause superradiance of spinning PBHs when its Compton wavelength is comparable to the size of the PBHs, resulting in the emission of coherent GWs whose angular frequency is given by twice the boson mass m b. The GW frequency is given by f ∼ 4.8 × 10 17 Hz (m b/keV) = 3.2 × 10 17 Hz (α/0.1)(2m PBH/(2 × 10−14 M⊙))−1, requiring a bosonic field with a keV-scale mass." The GW amplitude is h 0 ≃ 10−38 (2m PBH/(2 × 10−14 M⊙))(α/0.1) 7 (kpc/r)(∆χ/0.5), and the emission continues for a coherence time τ c ≃ 8.3 × 10−3 s (2m PBH/(2 × 10−14 M⊙))(α/0.1)−15 (∆χ/0.5)−1.

The authors consider spinning PBHs formed through mergers, with parameters evaluated using approximately twice the mass of the pre-merger PBHs. The event rate of PBH superradiance is assumed equivalent to the PBH merger rate. For a non-clustered initial condition, the event rate is R PBH ≃ 8.8 × 10 cubed kpc−3 yr−1 (ρ̄ G/ρ DM)(m PBH/(9.3 × 10 5 × 10−14 M⊙))−5, while for a clustered initial condition, it is enhanced to R PBH max ≃ 6.1 × 10 9 kpc−3 yr−1 (ρ̄ G/ρ DM)(m PBH/(9.3 × 10 5 × 10−14 M⊙))−5. The accumulated GW background from superradiance gives h 2Ω GW ≃ (1–2) × 10−8 for the non-clustered case and h 2Ω GW ≃ 1.0 × 10−2 for the clustered case.

For the conversion of gravitons into photons, the authors model the large-scale Galactic magnetic field with magnitude B G ∼ 1 µG on average over a Galactic length l G ∼ 20 kpc. The conversion probability is P(r) ≃ (r squared B G 2)/(2M pl 2) for r ≤ l os and P(r) ≃ (l os squared B G 2)/(2M pl 2) for l os ≤ r, where the oscillation length is l os ≃ 20 kpc (f/(2.6 × 10 18 Hz)). The conversion probability in the Milky Way is estimated as (20 kpc) 2(1 µG) 2/(2M pl 2) ≃ 3 × 10−16.

The total number of photons from all events within a narrow field of view (FoV) around the Galactic center during observation time t ob is given by Eq. (9), which shows that in the regime r < l os, more distant events collectively make larger contributions to the total photon number. Consequently, the integration is significantly enhanced, particularly when l G ≲ l os. The condition l G ≲ l os corresponds to frequencies above ∼ 10 18 Hz, or to PBH masses below ∼ 10−14 M⊙.

The expected photon number is compared with the estimated sensitivity of a benchmark X-ray telescope setup, using the projected Lynx sensitivity of 1.6 × 10−19 erg/cm 2/s as a reference, with an effective area A = 30 m squared, a FoV of 30 deg squared, an observation time of 1 year, and an energy resolution ∆E/E = 4 × 10−4 determined by Doppler broadening due to the Galactic DM velocity dispersion. The results show that the expected photon number increases with frequency up to the scale given by Eq.(8), and can exceed the estimated sensitivity under the clustered initial condition for PBH masses around 10−15 M⊙ ∼ 10−13 M⊙.

The authors conclude that "such photons may be detectable for PBHs in the mass range 10−15 M⊙ ∼ 10−13 M⊙ if they constitute all of DM and have clustered initial conditions, as expected in scenarios with non-Gaussian primordial fluctuations, with an optimistic setup for future X-ray observations. They also note that from the viewpoint of the string axiverse, it may be natural to have spin-0 fields with masses around 100 eV ∼ 10 keV, which cause the desired superradiance, and that a similar discussion can be applied to bosonic fields with arbitrary spins. Therefore, future X-ray observations could open a new observational window on light PBH DM through superradiant high-frequency GWs."

Improvements to AI Systems Based on This Paper:

1. Multi-messenger event-rate prediction model

• The AI can now estimate the detectability of high-frequency gravitational waves (HFGWs) from primordial black hole (PBH) superradiance by combining PBH mass functions, merger rates, and boson mass parameters.

• It can output expected photon fluxes (in X-ray bands) for arbitrary Galactic magnetic field models and observation geometries, enabling rapid parameter-space scans for future X-ray missions.

2. Frequency-dependent signal-to-noise optimization

• The AI can compute the optimal observational frequency window (around 10 18 Hz) where the conversion probability and event rate jointly peak, given a PBH mass and clustering scenario.

• It can automatically adjust telescope parameters (effective area, field of view, energy resolution) to maximize detection significance for a given PBH mass range.

3. Clustering-condition classifier

• The AI can distinguish between clustered and non-clustered PBH initial conditions based on the predicted photon count and spectral shape.

• It can infer whether non-Gaussian primordial fluctuations are required to explain a detected X-ray excess, by comparing observed photon rates to the theoretical R PBH max vs. R PBH thresholds.

4. Uncertainty propagation for Galactic magnetic field models

• The AI can propagate uncertainties in B G and l G (e.g., from 0.5 µG to 2 µG, or 10–30 kpc) into the conversion probability and final photon number, providing confidence intervals for detection forecasts.

• It can also handle the transition regime r l os to correctly integrate contributions from distant events.

5. Boson mass inference from GW frequency

• Given a candidate HFGW detection (or X-ray counterpart), the AI can invert the relation f about 4.8 times 10 17 (m b/ keV) Hz to estimate the axion-like particle mass, including error bars from PBH mass uncertainty and spin parameters.

• It can cross-correlate this inferred mass with string axiverse predictions (100 eV–10 keV) to rank theoretical priors.

6. Event-rate normalization and merger-rate calibration

• The AI can recalibrate the PBH superradiance event rate using updated merger-rate estimates from LIGO/Virgo or future gravitational-wave detectors, and propagate this into X-ray detectability forecasts.

• It can handle the factor-of-two mass scaling (post-merger vs. pre-merger) to avoid systematic biases in amplitude and frequency calculations.

7. Sensitivity-matching for next-generation X-ray telescopes

• The AI can simulate the expected photon counts for specific instruments (e.g., Lynx, Athena, or proposed X-ray observatories) and output the minimum PBH abundance (or clustering fraction) required for a 5σ detection.

• It can also optimize the field-of-view placement (e.g., toward the Galactic center vs. off-center) to exploit the r squared enhancement in the conversion probability.

8. Background-noise discrimination

• The AI can model astrophysical X-ray backgrounds (e.g., diffuse Galactic emission, point sources) and use the predicted narrow spectral line (from monochromatic GWs) to separate the PBH signal from continuum backgrounds.

• It can apply Doppler broadening corrections (due to DM velocity dispersion) to predict line width and improve matched-filtering detection algorithms.

9. Spin-arbitrariness extension

• The AI can generalize the superradiance calculation to spin-1 or spin-2 fields (e.g., dark photons or gravitons) by adjusting the boson mass–frequency relation and coupling constants, enabling searches for other ultralight particles via the same X-ray conversion channel.

10. Joint PBH–boson parameter estimation

• The AI can perform a Bayesian joint fit of PBH mass, spin, boson mass, and clustering amplitude using both the HFGW background (h squared GW) and the X-ray photon count, breaking degeneracies that single-channel observations cannot resolve.

• It can output posterior distributions for the DM fraction in PBHs and the axion-photon coupling, directly informing particle physics models.

What the improved AI system can do:

• Given a set of PBH masses and clustering assumptions, it can predict the exact X-ray photon flux and spectral shape observable by future telescopes.

• It can automatically design an optimal observational campaign (telescope pointing, energy band, exposure time) to test light PBH DM.

• It can distinguish between PBH superradiance and other high-frequency GW sources (e.g., cosmic strings) by their unique frequency–amplitude–event-rate signatures.

• It can provide real-time detection probability updates as new merger-rate or magnetic-field data become available.