No hot and luminous progenitor for Tycho's supernova

Abstract

Type Ia supernovae have proven vital to our understanding of cosmology, both as standard candles and for their role in galactic chemical evolution; however, their origin remains uncertain. The canonical accretion model implies a hot and luminous progenitor which would ionize the surrounding gas out to a radius of $\sim$10--100 parsecs for $\sim$100,000 years after the explosion. Here we report stringent upper limits on the temperature and luminosity of the progenitor of Tycho's supernova (SN 1572), determined using the remnant itself as a probe of its environment. Hot, luminous progenitors that would have produced a greater hydrogen ionization fraction than that measured at the radius of the present remnant ($\sim$3 parsecs) can thus be excluded. This conclusively rules out steadily nuclear-burning white dwarfs (supersoft X-ray sources), as well as disk emission from a Chandrasekhar-mass white dwarf accreting $\gtrsim 10^{-8}M_{\odot}$yr$^{-1}$ (recurrent novae). The lack of a surrounding Str\"omgren sphere is consistent with the merger of a double white dwarf binary, although other more exotic scenarios may be possible.