Resonance neutron capture and transmission measurements and the stellar neutron capture cross sections of Ba-134 and Ba-136

We have made high-resolution neutron capture and transmission measurements on isotopically enriched samples of Ba-134 and Ba-136 at the Oak Ridge Electron Linear Accelerator (ORELA) in the energy range from 20 eV to 500 keV. Previous measurements had a lower energy limit of 3-5 keV, which is too high to determine accurately the Maxwellian-averaged capture cross section at the low temperatures (kT approximate to 8-12 keV) favored by the most recent stellar models of the s process. By fitting the data with a multilevel R-matrix code, we determined parameters for 86 resonances in Ba-134 below 11 keV and 92 resonances in Ba-136 below 35 keV. Astrophysical reaction rates were calculated using these parameters together with our cross section data for the unresolved resonance region. Our results for the astrophysical reaction rates are in good agreement with the most recent previous measurement at the classical s-process temperature kT=30 keV, but show significant differences at lower temperatures. We determined that these differences were due to the effect of resonances below the energy range of previous experiments and to the use of incorrect neutron widths in a previous resonance analysis. Our data show that the ratio of reaction rates for these two isotopes depends more strongly on temperature than previous measurements indicated. One result of this temperature dependence is that the mean s-process temperature we derived from a classical analysis of the branching at Cs-134 is too low to be consistent with the temperature derived from other branching points. This inconsistency is evidence for the need for more sophisticated models of the s process beyond the classical model. We used a reaction network code to explore the changes in the calculated isotopic abundances resulting from our new reaction rates for an s-process scenario based on a stellar model. These calculations indicate that the previously observed 20% discrepancy with respect to the solar barium abundance is reduced but not resolved by our new reaction rates.