Experimental study of configuration mixing in intermediate excited levels of barium.

Abstract

Resonant three-photon ionization of barium, via five closely spaced excited bound states---labeled 6s8s $^{1}S_{0}$, 6${p}^{2}$ $^{3}P_{0}$, 6${p}^{2}$ $^{3}P_{1}$,6s7d $^{1}D_{2}$, and 6${p}^{2}$ $^{3}P_{2}$---has yielded both angular distributions of energy-resolved photoelectrons and branching ratios to the 6s, 5d, and 6p configurations of ${\mathrm{Ba}}^{+}$. In contrast to earlier ab initio and semiempirical theoretical analyses which attributed nearly pure configurations to these five bound states, our experimental findings are consistent with a significant degree of 6${p}^{2}$ character in the 6s8s $^{1}S_{0}$ and 6s7d $^{1}D_{2}$ levels, more in accord with their original configuration labels in the tables of Moore. Branching to the 6p states of the ion predominates in each case, and the angular distributions of the corresponding photoelectrons from the 6${p}^{2}$ $^{3}P_{2}$ and 6s7d $^{1}D_{2}$ levels are described well by a parametric theoretical model approximating these states as pure 6${p}^{2}$ configurations. For the 6s8s $^{1}S_{0}$ level, inferences based on these observations must be made cautiously because of the possibility of an ``anomalously'' low photoionization amplitude for the 6s8s component of the state. Available oscillator strengths suggest the possibility of configuration interaction via spin-orbit coupling. Overall, strong configuration mixing and/or mislabeling of the configurations is implied; we suggest that the choice of configurational labels for these states be reexamined.