Astrophysical uncertainties challenge 21-cm forecasts: A primordial black hole case study

TL;DR

This work addresses how astrophysical uncertainties in the first stars affect 21-cm forecasts for exotic energy injections, using accreting PBHs with masses $M_{ m PBH} \sim 1\,M_\odot-10^3\,M_\odot$ as a case study. It employs the analytic Zeus21 framework to model star-formation and radiation backgrounds, including Lyman-α and X-ray backgrounds, while incorporating PBH energy injection and delayed energy deposition through $f_c(z,x_e)$. By exploring three fiducial astrophysical scenarios and two accretion models (Bondi–Hoyle–Lyttleton vs Park–Ricotti), the study shows that the inferred PBH bounds on $f_{\rm PBH}$ can vary by several orders of magnitude, with radiative feedback in the PR model weakening constraints by at least two orders of magnitude relative to BHL. The results underscore the need for independent constraints on early galaxy properties and a multi-probe approach (e.g., JWST, ATHENA) to robustly translate a 21-cm measurement into limits on new physics.

Abstract

The 21-cm signal is a powerful probe of the early Universe's thermal history and could provide a unique avenue for constraining exotic physics. Previous studies have forecasted stringent constraints on energy injections from exotic sources that heat, excite, and ionize the background gas and thereby modify the 21-cm signal. In this work, we quantify the substantial impact that astrophysical uncertainties have on the projected sensitivity to exotic energy injection. In particular, there are significant uncertainties in the minimum star-forming dark matter halo mass, the Lyman-$α$ emission, and the X-ray emission, whose values characterize the fiducial astrophysical model when projecting bounds. As a case study, we investigate the energy injection of accreting primordial black holes of mass $\sim 1~M_\odot-10^3~M_\odot$, also taking into account uncertainties in the accretion model. We show that, depending on the chosen fiducial model and accretion uncertainties, the sensitivity of future 21-cm data could constrain the abundance of primordial black holes to be either slightly stronger, or significantly weaker, than current limits from the Cosmic Microwave Background.

Figures (6)

• Figure 1: Impact of accreting PBHs on the free-electron fraction (left panel), $x_e$, and on the gas temperature (right panel), $T_k$. Results are shown for two different accretion models, the Park-Ricotti and Bondi-Hoyle-Lyttleton models. For both panels we assume a PBH mass of $M_{\rm PBH} = 10^3 \ M_{\odot}$ and a fractional PBH abundance of $f_{\rm PBH} = 0.1$.
• Figure 2: Energy deposition into the IGM from PBHs into Lyman-$\alpha$ (solid lines), heating (dashed lines), and ionization (dotted lines) for the BHL (red lines) and PR (blue lines) accretion models, as a function of redshift. We assume a PBH mass of $M_{\rm PBH} = 10^3~M_\odot$ and a fractional PBH abundance of $f_{\rm PBH} = 0.1$.
• Figure 3: Predictions of the 21-cm global signal (left panels) and the power spectrum (right panels), as a function of $z$, for the less-constraining (first row), benchmark (middle row) and more-constraining (bottom row) astrophysical scenarios. We assume the BHL accretion model and PBHs of mass $M_{\rm PBH} = 10^2 \, M_\mathbf{\odot}$, with the colors corresponding to different values of $f_{\rm PBH}$ as indicated in the legend of the top left panel; for the power spectrum, we fix $k = 0.15 \, {\rm Mpc}^{-1}$. We also show the $1\sigma$ experimental sensitivities, $\sigma_{\rm exp}\left( z , k \right)$, from the HERA experiment, and the SKA AA* and SKA AA4 configurations, computed using 21cmSenseMurray:2024thePober:2012zzPober:2013. The astrophysical parameters for each row are given in the title, with the definitions provided in \ref{['tab:Astroparameters']}. The remaining astrophysical parameters are fixed to those shown in \ref{['tab:All_Astroparams']}.
• Figure 4: Predictions of the 21-cm signal for PBHs for the BHL accretion model (left column) and the PR accretion model (right column), and for a PBH mass of ${M_{\rm PBH} = 10^2~M_\odot}$ and for a range of fractional PBH abundance $f_{\rm PBH}$. Top row: impact on the global differential brightness temperature, $\delta T_b$, as a function of redshift. Middle row: impact of accreting PBHs on the 21-cm power spectrum, $\Delta^2_{21}$, at a fixed scale $k = 0.15~\textrm{Mpc}^{-1}$, as a function of redshift. The $1\sigma$ sensitivities, $\sigma_{\rm exp}\left( z , k \right)$, are shown for the HERA telescope, and for the SKA AA* and SKA AA4 configurations, computed using 21cmSenseMurray:2024thePober:2012zzPober:2013. Bottom row: same as the middle row, but now showing the power spectrum, $\Delta^2_{21}$, at a fixed redshift $z = 15$, as a function of $k$ instead. In each plot we assume the benchmark astrophysics scenario, corresponding to the middle row in \ref{['fig:pess_fid_opt']}, and given in \ref{['tab:Astroparameters']}. The legend provided in the bottom left panel applies to all panels.
• Figure 5: Comparison of the posteriors for our astrophysical and PBH parameters at 68% and 95% probability for a PBH mass of $M_{\rm PBH} = 10^2 \, M_{\odot}$ and for two accretion models, BHL (blue lines and contours) and PR (red lines and contours). The fiducial astrophysical model is assumed to be the benchmark scenario in \ref{['tab:Astroparameters']}.