Pulsations of primordial supermassive stars induced by a general relativistic instability; visible to JWST at z$>$12

Abstract

The origin of high-redshift quasars and their supermassive black hole engines is unclear. One promising solution is the collapse of a primordial supermassive star. Observational confirmation of this scenario may be difficult, but a general relativistic instability supernova provides one avenue for such. Previous studies have found that a general relativistic instability supernova has a potentially decades-long plateau phase visible to JWST at high redshift. In this work, we examine stars with mass just below the general relativistic instability supernova mass range. These stars pulsate, ejecting a portion of their envelopes. They then contract quasi-statically back to an equilibrium temperature, at which point they again become unstable and pulsate once more. Because each pulse consumes a small amount of the available nuclear fuel, there exists the possibility of multiple pulsations. We present simulations of the contracting phase, the pulsation, and the light-curve phase. We find that the lower mass pulsating models are even brighter than the higher mass supernovae because the pulsations occur in the late helium burning phase when the stars have extremely large radii. The fact that the pulsations are more luminous and occur in a wider mass range than the supernovae bodes well for observation.

Light-curves and nucleosynthesis of CNO-rp driven general relativistic instability supernovae in metal enriched supermassive protostars

“…Besides the high luminosity and temperature at shock breakout, there is another difference between the lightcurves of the pr-GRSNe and the lightcurves of standard GRSNe (𝛼 process, Moriya et al 2021;Nagele et al 2022a). This is the longer duration of the plateau phase which follows hydrogen recombination, nearly an order of magnitude longer than our previous GRSN models.…”

“…The photosphere remains at the surface of the hylotrope for ten days as the shock propagates through the envelope, after which the photosphere expands and the effective temperature drops. From a peak value of 𝐿 bol = 6.86 × 10 46 [ergs/s], the luminosity then falls nearly monotonically (except for a slight increase at hydrogen recombination) in behavior reminiscent of the pulsations of large radius supermassive stars in Nagele et al (2022a). The luminosity of the hylotropic model falls at a slower rate than for the fiducial model, passing 𝐿 bol = 10 43 [ergs/s] only after 9.95 years in the rest frame.…”

“…It includes artificial viscosity, an equation of state with a Saha solver for ionization of hydrogen and helium, and equilibrium flux-limited photon diffusion using OPAL. As in (Nagele et al 2022a), we port our HYDnuc models to SNEC slightly before shock breakout and for the outer layer of the SNEC model, we use the HOSHI progenitor which enables increased surface resolution necessary for properly following the lightcurve. We terminate the SNEC simulations after 10 9 seconds at which point any plateau phase in the luminosity has finished.…”