Characterization of sodium targets for the study of 23Na(p,a)20Ne reaction at LUNA

Stars produce most of elements in the universe out of primordial nuclei, H and He. Nuclear fusion reactions taking place deep inside stellar cores both provide the energy to keep stars in hydrostatic equilibrium and are responsible for the stellar nucleosynthesis. When the protostar temperature is high enough to start H ignition, it enters the Zero Age Main Sequence and accretes a He core via either predominantly the p-p chain (M_i < 1.2 M_sun) or the CNO cycle (M_i > 1.2 M_sun). When star runs out of H, the leftover moves to outer layers while the core shrinks. TheH continues burning progressively on a thin shell around the He core while the core contracts. When the temperature rises enough to burn He via the 3alpha process, the star will develop a C/O core. When the He-burning phase ends, stars with initial mass between 0.6 and 8 M_sun will enter the Asymptotic Giant Branch (AGB) phase. AGB-stars with a progenitor of at least 4 M_sun are sustained by an alternating He and H shell burning. In particular, the He shell burning proceeds via thermal pulses which drive deep convective phenomena in the (convective) envelope of the star. These mixing events bring on the surface products typical of CNO, NeNa and MgAl cycles, like O and Na. As they reach the star surface, they are effectively dispersed in the interstellar medium via the strong stellar winds typical of AGB-stars. Globular clusters (GC) are objects formed by hundreds of thousands of stars bound by gravity. Several observations of GC, some of which dates back to the seventies, revealed a number of anomalies in the observed chemical abundances, i.e. the Na-O anti-correlation. The activation in AGB stars of the NeNa cycle together with mixing events and mass loss makes them the best candidate to pollute the surrounding interstellar medium from which second generation of stars are formed within the globular cluster. The 23Na(p,alpha)20Ne reaction is placed at the branching point between the NeNa and MgAl cycles and it is responsible for the conversion of 23Na into 20Ne. Since this reaction directly affect the 23Na abundance that will be eventually brought to the surface during the mixing processes and subsequently dispersed in the interstellar medium, it is crucial to precisely determine its reaction rate to constrain the role of AGB in the Na-O anti-correlation. Variations of the cross-section of the 23Na(p,alpha)20Ne reaction can dramatically influence the predicted Na abundance. Resonances play a fundamental role by enhancing the cross-section around the resonance energy E_r. The last uncertainties that affect the 23Na(p,alpha)20Ne reaction rate are related to low-energy resonances below 170 keV. In particular, the resonance at E_r^cm = 138 keV has not been observed directly and is still affected by large uncertainties. An upcoming experiment at LUNA 400kV accelerator aims for the first time at directly measuring its strength. A high intensity proton beam will be delivered from LUNA 400kV to a solid target, while the generated alpha particles will be detected at backward angles by a dedicated array of Si detectors. Backscattered protons will be filtered out by thin foils of aluminised Mylar. The present work focuses on the characterization of the solid targets which will be used in the aforementioned measurement. Two types of targets have been investigated: the evaporated Na_2WO_4 targets produced initially by the Laboratori Nazionali di Legnaro LNL (Italy) and later by the Institute for Nuclear Research (MTA Atomki) in Debrecen (Hungary), and the sputtered NaNbO_3 targets produced at LNL. The relevance of this analysis lies in the fact that targets represent one of the main source of systematic uncertainty. In order to characterize targets, the well known Nuclear Resonant Reaction Analysis technique, consisting of scanning a narrow resonance has been used.