An atomic physics viewpoint of stellar abundance analysis

by Jennifer S. Sobeck

Institution: University of Texas – Austin
Department: Physics
Degree: PhD
Year: 2015
Keywords: Atomic physics; Stellar abundance; Transition probabilities
Record ID: 2060868
Full text PDF: http://hdl.handle.net/2152/29675


Element abundance trends with overall metallicity contain vital clues to the formation and evolution of the Galaxy. Abundances may be used to elucidate nucleosynthesis mechanisms and to ascertain rates of Galactic enrichment. To obtain accurate abundances, several crucial inputs such as high-quality spectroscopic observations, rigorous calculations of line transfer, and precise atomic data (e.g. transition probabilities) are necessary. The current work endeavors to improve abundance values for key elements with a four-fold approach: accumulation of hundreds of high-resolution stellar spectra in order to commence a systematic and thorough Manganese abundance derivation in cluster and halo field stars; re-determination of the neutral chromium oscillator strengths and application of this data to stellar abundance analyses; modification of a radiative line transfer code in order to yield accurate abundances from evolved stars; and semi-empirical derivation of transition probabilities to allow for the utilization of spectral features in the red visible and infrared wavelength ranges for abundance determinations. The first comprehensive investigation of manganese in globular clusters is done in this work. A subsolar Mn abundance trend for both halo globular cluster and field stars is found. The analysis shows that for the metallicity range -0.7>(Fe/H)>-2.7 stars of 19 globular clusters have a a mean relative abundance of <(Mn/Fe)>= -0.37±0.01 (σ=0.10), a value in agreement with that of the field stars: <(Mn/Fe)>= -0.36± 0.01 (σ=0.08). Remarkably, the <(Mn/Fe)> ratio remains constant in both stellar populations over a 2 orders of magnitude span in metallicity. Next, the present study employed branching fraction measurements from Fourier transform spectra in conjunction with published radiative lifetimes to determine transition probabilities for 263 lines of neutral chromium. These laboratory values are used to derive a new photospheric abundance for the Sun: log [element of](Cr I)⊙= 5.64±0.01 (σ=0:07). In addition, oscillator strengths for singly-ionized chromium recently reported by the FERRUM Project are employed to determine: log [element of](Cr II)⊙ = 5.77±0.03 (σ= 0.13). No indications of departures from LTE are found in the neutral chromium abundances. The current work then takes advantage of the fact that transition metals exhibit relatively pure LS coupling and employs standard formulae to yield semi-empirical oscillator strengths. These data were then compared to experimental gf values in order to assess accuracy. Finally, this study undertakes a new abundance investigation of the RGB and RHB stars of the M15 globular cluster. A detailed examination of the both the metallicity and n capture elements is performed. This work appears to confirm that star-to-star abundance variations do occur among the M15 giants (which was initially observed by Sneden et al. 1997, 2000).