Transition rates, branching ratio, and lifetime of the 2p33sS2∘5 metastable state in O I

Abstract

Based on the multiconfiguration Dirac-Hartree-Fock method, the accurate line strengths, transition rates, branching ratio, branching fraction, and lifetime of the $2{p}^{3}3s\phantom{\rule{0.16em}{0ex}}^{5}S_{2}^{\ensuremath{\circ}}$ metastable state of neutral oxygen atom (O i), together with the uncertainties evaluated from the residual electron correlation effects, are determined by taking into account the effects of core-core, core-valence, and valence correlations, the Breit interaction, and QED effect. It is shown that the core correlation effect has a certain influence if the separation between $^{3}S^{\ensuremath{\circ}}$ and $^{5}S^{\ensuremath{\circ}}$ in the ${p}^{3}s$ configuration is accurately calculated. It is found that the calculations of branching ratio of $^{5}S_{2}^{\ensuremath{\circ}}$ remain stable in different electron correlation models. We infer that it is a peculiar intrinsic property leading to results useful for plasma diagnostics and other applications. The transition rates from $^{5}S_{2}^{\ensuremath{\circ}}$ change considerably as a result of the Breit interaction, because the line strengths of $^{3}P_{2}$--$^{5}S_{2}^{\ensuremath{\circ}}$ ($\ensuremath{\lambda}135.560$ nm) and $^{3}P_{1}$--$^{5}S_{2}^{\ensuremath{\circ}}$ ($\ensuremath{\lambda}135.851$ nm) are different in sensitivity to this effect. This affects the calculations of the branching ratio and lifetime. There exist large discrepancies among several calculations of lifetimes, such as the MCHF calculation by [C. Froese Fischer and G. Tachiev, At. Data Nucl. Data Tables 87, 1 (2004)]. The main reasons are ascribed to the neglected electron correlation and relativistic effects. In addition, by checking different calculations with various sets of multireference configurations, we find that the calculations of the electric dipole matrix elements for $^{5}S_{2}^{\ensuremath{\circ}}$ in the length gauge are much more sensitive to electron correlation effects than in the velocity gauge. This may also lead to some errors in the calculations of the lifetime of $^{5}S_{2}^{\ensuremath{\circ}}$ in the length gauge.