| We measured the abundances of 35 elements in 22 field red giants and a red giant in the globular cluster M92. We found the [Zn/Fe] ratio increases with decreasing [Fe/H], reaching ∼0.3 at [Fe/H] = −3.0. While this is a larger [Zn/Fe] than found by previous investigators, it is not sufficient to account for the [Zn/Fe] observed in the damped Lyα systems. We test different models for the production of the s-process elements by comparing our [Y/Zr] values, which have been produced by the r-process, to predictions of what the s-process does not produce. We find that the models of Arlandini et al. (1999), which calculate s-process production in a model AGB star, agree the best. We then look at the r-process abundances across a wide range in mass. The [Y/Ba] values for most of our stars cluster around −0.30, but there are three outliers with [Y/Ba] values up to 1 dex higher. Thus the heavy element abundances do not show the same pattern from Z = 39 to Z = 56. However, our abundances ratios from Pd (Z = 46) to Yb (Z = 70) are consistent with a scaled solar system r-process pattern, arguing that at least the heavy r-process elements are made in a universal pattern. If we assume that this same pattern hold through thorium, we can determine the ages of our stars from the present abundance of radioactive thorium and an initial thorium abundance based on the abundance of stable heavy elements. Our results for five stars are consistent with those stars being the same age. Our mean age is 10.8 ± 2 Gyr. However that result depends critically on the assumed Th/stable ratio, which we adopt from models of the r-process. For an average age of 15 Gyrs, the initial Th/Eu ratio we would need is 0.590. Finally, the [element/Fe] ratios for elements in the iron group and lower do not show any dispersion, unlike for the r-process elements such as Y and Ba. Therefore the individual contributions of supernovae have been erased for the lighter elements. |
