The impact of recombination energy on simulations of the common-envelope binary interaction

During the common envelope binary interaction, the expanding layers of the gaseous common envelope recombine and the resulting recombination energy has been suggested as a contributing factor to the ejection of the envelope. In this paper we perform a comparative study between simulations with and without the inclusion of recombination energy. We use two distinct setups, comprising 0.88-M$_{\odot}$ and 1.8-M$_{\odot}$ giants, that have been studied before and can serve as benchmarks. In so doing we conclude that (i) the final orbital separation is not affected by the choice of equation of state. In other words, simulations that unbind but a small fraction of the envelope result in similar final separations to those that, thanks to recombination energy, unbind a far larger fractions. (ii) The adoption of a tabulated equation of state results in a much greater fraction of unbound envelope and we demonstrate the cause of this to be the release of recombination energy. (iii) The fraction of hydrogen recombination energy that is allowed to do work should be about half of that which our adiabatic simulations use. (iv) However, for the heavier star simulation we conclude that it is helium and not hydrogen recombination energy that unbinds the gas and we determine that all helium recombination energy is thermalised in the envelope and does work. (v)~The outer regions of the expanding common envelope are likely to see the formation of dust. This dust would promote additional unbinding and shaping of the ejected envelope into axisymmetric morphologies.
Comment: 18 pages, 11 figures. We welcome any comments or questions