Recently the lifetime of the excited $3d^{9}4s^2{}^2{D}_{3/2}$ in ${\mathrm{Ni}}^{-}$ was measured to be $15.1\pm 0.4$ s [Phys. Rev. A 93, 012512 (2016)]. This deviates from results from nonrelativistic, within the $LS$-approximation, calculations for this forbidden magnetic-dipole transition of 18.88 s [J. Phys. B 50, 025001 (2017)]. We here present elaborate and fully relativistic multiconfiguration Dirac–Hartree–Fock calculations, to explore this difference. Our calculated transition energy $1485.65{\mathrm{cm}}^{-1}$ is in excellent agreement with the experimental $1485\pm 3{\mathrm{cm}}^{-1}$ [Phys. Rev. A 58, 2051 (1998)]. However the lifetime of 18.86 s, while agreeing well with the nonrelativistic analytical value, deviates from experiment by 25%. The uncertainties of our calculated wavelength and lifetime are estimated to be better than 0.1% and 1%, respectively. In spite of including a careful investigation of contributions from correlation as well as higher-order relativistic effects, we cannot find any explanation from the structure theory for this deviation.

• Received 5 January 2017

DOI:https://doi.org/10.1103/PhysRevA.95.042504