Massive-star binaries can undergo a phase where one of the two stars expands during its advanced evolutionary stage as a giant and envelops its companion, ejecting the hydrogen envelope and tightening its orbit. Such a common envelope phase is required to tighten the binary orbit in the formation of many of the observed X-ray binaries and merging compact binary systems. In the formation scenario for neutron star binaries, the system might pass through a phase where a neutron star spirals into the envelope of its giant star companion. These phases lead to mass accretion onto the neutron star. Accretion onto these common-envelope-phase neutron stars can eject matter that has undergone burning near to the neutron star surface. This paper presents nucleosynthetic yields of this ejected matter, using population synthesis models to study the importance of these nucleosynthetic yields in a galactic chemical evolution context. Depending on the extreme conditions in temperature and density found in the accreted material, both proton-rich and neutron-rich nucleosynthesis can be obtained, with efficient production of neutron rich isotopes of low Z material at the most extreme conditions, and proton rich isotopes, again at low Z, in lower density models. Final yields are found to be extremely sensitive to the physical modeling of the accretion phase. We show that neutron stars accreting in binary common envelopes might be a new relevant site for galactic chemical evolution, and therefore more comprehensive studies are needed to better constrain nucleosynthesis in these objects.

We investigated the structure of the low density regions of the inner crust of neutron stars using the Hartree-Fock-Bogoliubov (HFB) model to predict the proton content $Z$ of the nuclear clusters and, together with the lattice spacing, the proton content of the crust as a function of the total baryonic density $\rho_b$. The exploration of the energy surface in the $(Z,\rho_b)$ configuration space and the search for the local minima require thousands of calculations. Each of them implies an HFB calculation in a box with a large number of particles, thus making the whole process very demanding. In this work, we apply a statistical model based on a Gaussian Process Emulator that makes the exploration of the energy surface ten times faster. We also present a novel treatment of the HFB equations that leads to an uncertainty on the total energy of $\approx 4$ keV per particle. Such a high precision is necessary to distinguish neighbour configurations around the energy minima.