Dj. Bord et al., GF-VALUES FOR SINGLY IONIZED LANTHANUM BASED ON A NEW CALIBRATION OF NBS MONOGRAPH-145 INTENSITIES, Monthly Notices of the Royal Astronomical Society, 278(4), 1996, pp. 997-1004
Oscillator strengths and transition probabilities for Lan are computed
using a Hartree-plus-statistical-exchange-interaction approximation t
o the standard Hartree-Fock procedure. Relativistic and electron corre
lation corrections are included, and configuration interactions involv
ing all reliably established La II electron configurations are conside
red. Leading eigenvector percentages for the 5d6d and 5d7s configurati
ons are reported here for the first time. Criteria are developed and a
pplied for selecting the more reliable calculated oscillator strengths
. A set of 70 such values is used to calibrate Monograph 145 intensiti
es to produce moderately accurate gf-values. Comparison of an initial
two-parameter fit to an equation of the form log(gf) - log(I(145)lambd
a(3)) = c(1) + c(2) E(upper) (for E(upper) in cm(-1)) with beam-laser
lifetime measurements shows the resulting log(gf) values to be systema
tically too large by about 0.26. The reason for this discrepancy is no
t entirely understood, but it is likely to be due at least in part to
the incomplete treatment of core polarization effects in our calculati
ons. Modification of the original calibration in the light of the expe
rimental lifetime data yields the following recommended values for the
fitting parameters: c(1) = - 16.503 and c(2) = 7.752 x 10(-5). Log(gf
) values computed for La II lines using this calibration of the Monogr
aph 145 intensities should be accurate to about +/- 0.20 dex over the
energy range 0 to similar to 70 000 cm(-1). The predictions of this fo
rmula are in good agreement with published astrophysical log (gf) valu
es for 34 La II features in the solar spectrum longward of similar to
4300 Angstrom, and yield a mean difference (and standard deviation) of
- 0.07 +/- 0.30.