Modelling the Central Constant Emission X-ray component of η Carinae

Abstract

The X-ray emission of $\eta$ Carinae shows multiple features at various spatial and temporal scales. The central constant emission (CCE) component is centred on the binary and arises from spatial scales much smaller than the bipolar Homunculus nebula, but likely larger than the central wind--wind collision region between the stars as it does not vary over the $\sim$2-3 month X-ray minimum when it can be observed. Using large-scale 3D smoothed particle hydrodynamics (SPH) simulations, we model both the colliding-wind region between the stars, and the region where the secondary wind collides with primary wind ejected from the previous periastron passage. The simulations extend out to one hundred semimajor axes and make two limiting assumptions (strong coupling and no coupling) about the influence of the primary radiation field on the secondary wind. We perform 3D radiative transfer calculations on the SPH output to synthesize the X-ray emission, with the aim of reproducing the CCE spectrum. For the preferred primary mass-loss rate $\dot{M}_A\approx8.5\times10^{-4}$ M$_\odot$ yr$^{-1}$, the model spectra well reproduce the observation as the strong- and no-coupling spectra bound the CCE observation for longitude of periastron $\omega\approx252^\circ$, and bound/converge on the observation for $\omega\approx90^\circ$. This suggests that $\eta$ Carinae has moderate coupling between the primary radiation and secondary wind, that both the region between the stars and the comoving collision on the backside of the secondary generate the CCE, and that the CCE cannot place constraints on the binary's line of sight. We also discuss comparisons with common X-ray fitting parameters.