Effects of nuclear cross sections on $^{19}$F nucleosynthesis at low metallicities
The origin of fluorine is a longstanding problem in nuclear astrophysics. It is widely recognized that Asymptotic Giant Branch (AGB) stars are among the most important contributors to the Galactic fluorine production. In general, extant nucleosynthesis models overestimate the fluorine production by AGB stars with respect to observations. In this paper we review the relevant nuclear reaction rates involved in the fluorine production/destruction. We perform this analysis on a model with initial mass M=2 M$_\odot$ and Z=0.001. We found that the major uncertainties are due to the $^{13}$C($α$,n)$^{16}$O, the $^{19}$F($α$,p)$^{22}$Ne and the $^{14}$N(p,$γ$)$^{15}$O reactions. A change of the corresponding reaction rates within the present experimental uncertainties implies surface $^{19}$F variations at the AGB tip lower than 10\%. For some $α$ capture reactions, however, larger variations in the rates of those processes cannot be excluded. Thus, we explore the effects of the variation of some $α$ capture rates well beyond the current published uncertainties. The largest $^{19}$F variations are obtained by varying the $^{15}$N($α$,$γ$)$^{19}$F and the $^{19}$F($α$,p)$^{22}$Ne reactions. The analysis of some $α$ capture processes assuming a wider uncertainty range determines $^{19}$F abundances in better agreement with recent spectroscopic fluorine measurements at low metallicity. In the framework of the latter scenario the $^{15}$N($α$,$γ$)$^{19}$F and the $^{19}$F($α$,p)$^{22}$Ne reactions show the largest effects on fluorine nucleosynthesis. The presence of poorly known low energy resonances make such a scenario, even if unlikely, possible. We plan to directly measure these resonances.