Simon Guichandut

We perform the first multidimensional fluid simulations of thermonuclear helium ignition underneath a hydrogen-rich shell. This situation is relevant to Type I X-ray bursts on neutron stars which accrete from a hydrogen-rich companion. Using the low Mach number fluid code MAESTROeX, we investigate the growth of the convection zone due to nuclear burning, and the evolution of the chemical abundances in the atmosphere of the star. We also examine the convective boundary mixing processes which cause the evolution to differ significantly from previous one-dimensional simulations that rely on mixinglength theory. We find that the convection zone grows outwards as penetrating fluid elements cool the overlying radiative layer, rather than directly from the increasing entropy of the convection zone itself. Simultaneously, these flows efficiently mix composition, carrying carbon out of, and protons into the convection zone even before contact with the hydrogen shell. We discuss the implications of these effects for future modeling of these events and observations.

Motivated by the recent observation by NICER of a type I X-ray burst from SAX J1808.4–3658 with a distinct “pause” feature during its rise, we show that bursts which ignite in a helium layer underneath a hydrogen-rich shell naturally give rise to such pauses, as long as enough energy is produced to eject the outer layers of the envelope by super-Eddington winds. The length of the pause is determined by the extent of the convection generated after ignition, while the rate of change of luminosity following the pause is set by the hydrogen gradient left behind by convection. Using the MESA stellar evolution code, we simulate the accumulation, nuclear burning, and convective mixing prior to and throughout the ignition of the burst, followed by the hydrodynamic wind. We show that the results are sensitive to the treatment of convection adopted within the code. In particular, the efficiency of mixing at the H/He interface plays a key role in determining the shape of the lightcurve. The data from SAX J1808.4–3658 favor strong mixing scenarios. Multidimensional simulations will be needed to properly model the interaction between convection and nuclear burning during these bursts, which will then enable a new way to use X-ray burst lightcurves to study neutron star surfaces.