Vie et mort des étoiles qui capturent des trous noirs primordiaux ✛ ✦

Résumé original

tel que rédigé par les auteurs, en anglais arXiv:2606.02700v1

Primordial black holes (PBHs) in the asteroid mass window ($$10^{17}-10^{23}\,{\rm g}$$) remain viable dark matter candidates and can be captured by stars. We develop the first global framework for the evolution of stars that capture PBHs, combining analytic calculations, stellar evolution models, 3D general-relativistic magnetohydrodynamic simulations, and Monte Carlo population synthesis. We find that the fate of these systems bifurcates: PBHs that form an accretion disk before consuming the host drive explosive disruption, whereas PBHs captured too late or growing too slowly consume the star quietly. Capture is dominated by three-body interactions with planetary or stellar companions. For a solar-type host with a Jupiter analog, inspiral within a main-sequence lifetime requires $$M_{\rm BH}^{\rm crit}\gtrsim 10^{22}\,{\rm g}$$, while lighter PBHs generally require tighter companions. Once deposited at the center, the PBH grows through inefficient quasi-spherical Bondi accretion; if it reaches the angular-momentum threshold before consuming the host, the inflow circularizes into a disk. Our Monte Carlo calculations yield sizable quiet-consumption and explosive-disruption populations, with final PBH masses $$M_{\rm BH}\sim0.01-1\,M_\odot$$ and disk-forming PBH spins $$a_\ast\approx0.8$$. Disk formation is the point of no return: disk winds and relativistic jets of $$\sim10^{45}-10^{50}\,{\rm erg\,s^{-1}}$$ disrupt the star within minutes. The resulting transients may include a $$\sim$$day-long UV/blue signal, radio afterglow, and, if the jet escapes, an X-ray-flash/low-luminosity gamma-ray-burst (XRF/llGRB) signal. For an $$O(1)$$ PBH dark matter fraction and optimistic capture assumptions, the event rate can reach that of llGRBs. The low-mass, high-spin remnants offer a complementary PBH probe and possible source for subsolar BH mergers.