The main goal of this Thesis is to study the formation of the elements heavier than Fe in asymptotic giant branch (AGB) stars. These elements are formed when iron peak nuclei present in the intershell experience a series of slow neutron-captures and β− decays to form isotopes of heavier elements (the slow neutron-capture process, hereafter s-process). During the thermally pulsing AGB phase, convective motion in the envelope (third dredge up) drives the products of the s-process, along with C, from the layer between the H- and He-burning shells to the stellar surface. The C and s-process enriched material is then released in the interstellar medium at the end of the AGB phase, which is subsequently ionized during the planetary nebula (PN) formation. The chemical analysis of the ionized gas, in particular the composition of the neutron(n)-capture elements, thus gives direct information on the nucleosynthesis that occurred during the last stages of the progenitor stars.
We present the spectroscopic analysis of a sample of 8 PNe in the optical and in the near-infrared (NIR) wavelength ranges. We obtained very high-resolution optical spectra (R∼40000) of the PNe NGC 3918 and NGC 5315 with the Ultraviolet Visual Echelle spec- trograph (UVES) attached to the UT2 of the 8.2m Very Large Telescope (VLT), at Cerro Paranal, Chile. For these two objects, we also obtained medium-resolution NIR spectra (R∼4800) with the FIRE spectrograph, attached to the 6.5m Magellan Baade Telescope, at Las Campanas, Chile. In addition, we observed 6 PNe in the NIR range, with the medium-resolution (R∼3500-4800) EMIR spectrograph, attached at the Gran Telescopio Canarias, in La Palma, Spain.
The analyzed spectra have allowed to reach our main goals. We have reduced the large uncertainties computed in previous calculations of n-capture element abundances in PNe, by detecting emission lines from multiple ionic species for a given element. Precise determinations of n-capture element abundances have allowed us to test the predictions of modern AGB nucleosynthesis models as well as to test the accuracy of new atomic data computations and ionization correction factors (ICF) prescriptions.
To achieve these objectives, the obtained raw data have been reduced to finally obtain 1D flux-calibrated spectra. Line fluxes in each spectra have been measured and a detailed line identification process has been performed for each object, which translated in more than 1700 lines measured and identified in our sample. We have measured and identified emission lines of several ions of n-capture elements such as Ge, Se, Br, Kr, Rb, Cd, Te and Xe. Some of these lines have been identified for the first time in astrophysical nebulae. The chemical abundances analysis has been performed using an homogeneous set of atomic data for the whole sample of PNe. We have computed physical conditions (electron temperature and density) and ionic abundances of ions of common elements such as He, C, N, O, Cl, S, Ar and Fe from our optical spectra. For the objects observed only in the NIR with EMIR, we have considered UV, optical and mid- to far-infrared line fluxes available in the literature. Ionic abundances of n-capture elements have been calculated using optical and NIR line intensities measured in our spectra.
The elemental abundances of common and n-capture elements have been calculated with recent ICFs available in the literature. In particular, Kr and Se abundances have been calculated using the complete set of available ICFs, which were recently constructed with detailed photoionization models. We have found good agreement between the Kr abundances derived using different ICFs and the elemental abundances derived for this element are among the most accurate determinations ever made for n-capture elements. For Se we have concluded that the complete set of ICFs created for this element produce reliable results only for high excitation PNe. Dedicated ICFs are not available for Br, Rb, Te and Xe and their elemental abundances were computed using approximate ioniza- tion correction schemes based on the ionic fractions of common ions, which have similar ionization potentials than the detected ions of these n-capture elements.
We have calculated n-capture element enrichments for Se, Br, Kr, Rb, Te and Xe. We conclude that a substantial s-process enrichment occurred on the stellar surfaces of the progenitor stars of NGC 3918, NGC 7027, IC 418 and M 1-11 during the AGB phase. For NGC5315, NGC2440 and DdDm1 we calculate no enrichments of s-processed material. The high mass of the progenitor stars may cause a strong dilution during TDU and/or the presence of an interacting central binary systems may truncate the AGB lifetimes ham- pering the s-process enrichment for NGC 5315 and NGC 2440. For the halo PN DdDm 1 we conclude that its progenitor star has an initial mass too low to experience TDU events.
The enrichments of the n-capture elements have been compared with the theoretical predictions for a wide range of initial progenitor star masses. We have found a good agreement between the estimated initial masses using n-capture and common element abundances in our solar metallicity PNe, while for the very low metallicity objects of our sample our observational results are less consistent with the theoretical predictions. Finally, we have explored the ratio between the enrichments of the elements belonging to the second (Ba, La and Ce) and to the first (Sr, Y and Zr) s-process peak ([hs/ls]), which gives direct information on the time integrated neutron flux during the thermally pulsing AGB phase, finding that the [Xe/Kr] and [Te/Se] are also useful indicators of the time integrated neutron flux.