From Neutrino- to Photon-cooled in Three Years: Can Fallback Accretion Explain the X-Ray Excess in GW170817?

Excess X-ray emission from the neutron star merger GW170817 above the predicted afterglow was recently detected t ≈ 3.4 yr post-merger. One possible origin is accretion onto the newly unshrouded black hole (BH) remnant. While fallback of bound dynamical ejecta is insufficient to generate the excess luminosity, LX ∼ 5 × 1038 erg s−1, fallback from the disk wind ejecta—due to their larger mass and lower velocity—remains a possibility. We present hydrodynamic α-viscosity simulations of the post-merger disk evolution that extend to timescales t ≈ 35 s post-merger, necessary to capture the asymptotic evolution into the radiatively inefficient regime. Due to inefficient neutrino cooling, the BH accretion rate decays rapidly at late times (${\dot{M}}_{\mathrm{bh}}\propto {t}^{-{\beta }_{\mathrm{bh}}}$, where βbh ≈ 2.4–2.8), which is incompatible with the late-time excess. However, matter falls back to the inner disk from the equatorial region more gradually, ${\dot{M}}_{\mathrm{fb}}\propto {t}^{-{\beta }_{\mathrm{fb}}}$ with βfb ≈ 1.43 in our α ≈ 0.03 simulations. By the present epoch t ≈ 3.4 yr, the fallback rate has become sub-Eddington and the disk can again accrete efficiently, i.e., ${\dot{M}}_{\mathrm{bh}}\approx {\dot{M}}_{\mathrm{fb}}$, this time due to photon instead of neutrino cooling. The predicted present-day X-ray accretion luminosity, ${L}_{{\rm{X}}}\approx 0.1{\dot{M}}_{\mathrm{bh}}{c}^{2}\approx (2\mbox{--}70)\times {10}^{38}$ erg s−1 for βfb ≈ 1.43–1.66, thus supports (with caveats) an accretion-powered origin for the X-ray excess in GW170817. The suppressed BH accretion rate prior to the sub-Eddington transition, weeks to months after the merger, is key to avoid overproducing the kilonova luminosity via reprocessing.