Improved laboratory transition probabilities for GdII and application to the gadolinium abundances of the sun and three r-process rich, metal-poor stars

Radiative lifetimes, accurate to +/-5%, have been measured for 49 even-parity and 14 odd-parity levels of Gd II using laser-induced fluorescence. The lifetimes are combined with branching fractions measured using Fourier transform spectrometry to determine transition probabilities for 611 lines of Gd II. This work is the largest-scale laboratory study to date of Gd II transition probabilities and the first using a high-performance Fourier transform spectrometer. This improved data set has been used to determine a new solar photospheric Gd abundance, log epsilon = 1.11 +/- 0.03. Revised Gd abundances have also been derived for the r-process-Y-rich metal-poor giant stars CS 22892-052, BD + 17 3248, and HD 115444. The resulting Gd/Eu abundance ratios are in very good agreement with the solar system r-process ratio. We have employed the increasingly accurate stellar abundance determinations, resulting in large part from the more precise laboratory atomic data, to predict directly the solar system r-process elemental abundances for Gd, Sm, Ho, and Nd. Our analysis of the stellar data suggests slightly higher recommended values for the r-process contribution and total solar system values, consistent with the photospheric determinations, for the elements for Gd, Sm, and Ho.