LIGO gravitational wave detection, primordial black holes and the near-IR cosmic infrared background anisotropies

TL;DR

The paper argues that LIGO-detected BHs could be primordial black holes constituting dark matter, introducing a Poissonian isocurvature component that dominates small-scale power. This boosts the formation of early halos (z>10), increasing the abundance of luminous sources and potentially explaining the observed near-IR cosmic infrared background anisotropies with very modest formation efficiencies. Additionally, gas accretion onto PBHs in these halos could produce IR and soft X-ray emission, naturally accounting for the CIB–CXB coherence and offering testable predictions about early star-formation and reionization. The conclusions hinge on PBHs comprising all or most of DM and imply modifications to the early radiative history and feedback processes, with observational implications for future surveys and cross-correlations.

Abstract

LIGO's discovery of a gravitational wave from two merging black holes (BHs) of similar masses rekindled suggestions that primordial BHs (PBHs) make up the dark matter (DM). If so, PBHs would add a Poissonian isocurvature density fluctuation component to the inflation-produced adiabatic density fluctuations. For LIGO's BH parameters, this extra component would dominate the small-scale power responsible for collapse of early DM halos at z>10, where first luminous sources formed. We quantify the resultant increase in high-z abundances of collapsed halos that are suitable for producing the first generation of stars and luminous sources. The significantly increased abundance of the early halos would naturally explain the observed source-subtracted near-IR cosmic infrared background (CIB) fluctuations, which cannot be accounted for by known galaxy populations. For LIGO's BH parameters this increase is such that the observed CIB fluctuation levels at 2 to 5 micron can be produced if only a tiny fraction of baryons in the collapsed DM halos forms luminous sources. Gas accretion onto these PBHs in collapsed halos, where first stars should also form, would straightforwardly account for the observed high coherence between the CIB and unresolved cosmic X-ray background in soft X-rays. We discuss modifications possibly required in the processes of first star formation if LIGO-type BHs indeed make up the bulk or all of DM. The arguments are valid only if the PBHs make up all, or at least most, of DM, but at the same time the mechanism appears inevitable if DM is made of PBHs.

Figures (3)

• Figure 1: Black solid line marks the CMBFAST-computed $\Lambda$CDM power spectrum at $z=20$ vs the mass contained within the comoving radius $2\pi/k$ for the cosmological parameters adopted here. Black dashes show the $P_{\Lambda{\rm CDM}}\propto k^{-3}$ extrapolation to scales inaccessible to CMBFAST, but relevant for the first halos collapse. Red horizontal solid line shows the Poissonian power from DM PBHs of $M_{\rm PBH}=30M_\odot$, which clearly dominates the scales relevant for halo collapse at this epoch.
• Figure 2: Curves show the rms density contrast over the halo mass for $M_{\rm PBH}=0$ (thin), 15 (thick), 30 (thickest) $M_\odot$ at $z=$ 30 (red), 20 (green), 15 (blue), 10 (black). Black horizontal line shows $\delta_{\rm col}$, so halos with density contrast $>\delta_{\rm col}$ collapse at that $z$. Vertical dashes with same color notation mark halo mass where $T_{\rm vir}>10^4$K and vertical dash--dotted lines show the same for $T_{\rm vir}>10^3$K (at $z>15$ they are to the left of the box).
• Figure 3: Fraction of collapsed halos (eq. \ref{['eq:f_halo']}) at $T_{\rm vir}>10^4$K (left) and $T_{\rm vir}>10^3$K (right) vs $z$ for standard $\Lambda$CDM power spectrum (red filled circles), DM PBHs with $M_{\rm PBH}=15M_\odot$ (open black circles) and $M_{\rm PBH}=30M_\odot$ (filled black circles). Thick solid curves mark the overall fraction of baryons (effectively $f_*f_{\rm Halo}$) needed to produce the observed CIB per eq. \ref{['eq:fraction_bc']} with $f_{\rm Halo}=1$ with the H-burning radiation efficiency $\epsilon=0.007$ (blue) and BH-type efficiency $\epsilon=0.2$ (black). The mean efficiency of the required conversion of baryons into luminous sources inside each halo would be the ratio of the solid curves to the circles. While $f_*$ is high (even higher than, or comparable to, 100% at $z_>\atop^{\sim}{$_>^∼$} 20$), it remains very modest if the PBHs make up the DM.