The origin of nitrogen

Recent new high-precision abundance data for Galactic halo stars suggest important primary nitrogen production in very metal-poor massive stars. Here, we compute a new model for the chemical evolution of the Milky Way aimed at explaining these new abundance data. The new data can be explained by adopting: a) the stellar yields obtained from stellar models that take into account rotation; and b) an extra production of nitrogen in the very metal-poor massive stars. In particular, we suggest an increase of nearly a factor of 200 in 14N for a star of 60 $M_{\odot}$ and $\simeq$40 for a star of 9 $M_{\odot}$, for metallicities below $Z=10^{-5}$, with respect to the yields given in the literature for $Z=10^{-5}$ and rotational velocity of 300 km s-1. We show that once we adopt the above prescriptions, our model is able to predict high N/O abundance ratios at low metallicities and still explains the nitrogen abundances observed in thin disk stars in the solar vicinity. The physical motivation for a larger nitrogen production in massive stars in very metal-poor environments could be the fact that some stellar models as well as observational data suggest that at low metallicities stars rotate faster. If this is the case, such large nitrogen production seen in the pristine phases of the halo formation would not necessarily happen in Damped Lyman-α systems which have metallicities always above ${\rm [Fe/H]}\simeq -2.5$, and could have been pre-enriched. We also compute the abundance gradient of N/O along the Galactic disk and show that a negative gradient is predicted once we adopt stellar yields where rotation is taken into account. The latter result implies that intermediate mass stars contribute less to the primary nitrogen than previously thought.