Ramsbottom et al. (1998)
have recently presented thermally-averaged
collision strengths for Ne IV calculated in the R-matrix
approximation. Values were presented over the temperature range
. This data replaces the Distorted Wave
calculations of
Bhatia (1996,
unpublished)
that were previously in the
database.
For the same reasons explained in Sect. 2.1, thermally-averaged collision strengths for temperatures lower than 104.4 K have been omitted from the database.
The new model for the ion includes the 22 fine-structure levels of the [2s 2][2p 3], [2s2p 4], [2p 5] and [2s 2][2p 2][3s ] configurations, whose experimental energies come from the NIST database and Bhatia & Kastner (1988). Theoretical energy levels and thermally-averaged collision strengths come from the calculations of Ramsbottom et al. (1998). Radiative transition probabilities have been calculated using the SSTRUCT package including 17 configurations: [2s 2][2p 3], [2s2p 4], [2p 5], [2s 2][2p 2][3l ], [2s2p 3][3l ] (l=s, p, d), [2s 2][2p 2][4l ] and [2s2p 3][4l ] (l=s, p, d, f).
Comparison between Ramsbottom et al. (1998) and Bhatia (1996) has shown that differences of up to 40% arise between theoretical emissivities calculated from the two models.
The new Distorted Wave calculations of Bhatia & Young (1998) have been fitted and replace the data of Bhatia & Mason (1980b) in the database. The new calculations include the configurations [2s 2][2p 3], [2s2p 4], [2p 5] and [2s 2][2p 2][3s ] and were computed at several values of the incoming electron energy. Radiative data are also taken from Bhatia & Young (1998). The level energies of Edlen (1984) are used for the [2s 2][2p 3], [2s2p 4] and [2p 5] configurations, while the [2s 2][2p 2][3s ] energies are from the NIST database (Martin et al. 1995).
A comparison between the Bhatia & Young (1998) collision strengths and the R-Matrix thermally-averaged collision strengths calculated by Ramsbottom & Bell (1997) did not show any significant differences in level populations and line intensities.
Due to the lack of atomic data in the literature, collision strengths, radiative transition probabilities and energy levels have been interpolated along the isoelectronic sequence for both ions. The data used for interpolation were taken from the most abundant ions of the sequence, already included in version 1.0 of the CHIANTI database. For Ni XXII, the Zn XXIV data from Bhatia et al. (1989) have been included in the interpolation.
The interpolation program has been developed by one of the authors (EL) and allows the calculation of theoretical energy levels, radiative transition probabilities and thermally-averaged collision strengths through the interpolation of CHIANTI data for isoelectronic ions. Extensive checks have been made in order to assess the reliability of the results: the use of the program on ions already present in the CHIANTI database yielded results within 5% of the original data.
The 13 level Fe XX model described in Paper I has been extended to 72 levels through the inclusion of the the [2s 2][2p 2][3l ] (l = s, p, d) configurations, corresponding to 72 fine-structure energy levels. The experimental energy levels have been taken from the NIST database (Martin et al. 1995), replacing also the values of the lower n=2 levels adopted in CHIANTI v. 1.0 from Bhatia & Mason (1980a). Theoretical energy levels are from Bhatia & Mason (1980a) (n=2 configurations) and Bhatia et al. (1989) (n=3 configurations).
Radiative and collisional transition probabilities for the n=3 transitions
are taken from the single-energy
Distorted Wave calculations of
Bhatia et al. (1989).
The availability of
only one energy point for did not allow to take into account the dependence of
on the
electron energy and therefore the calculation of the thermally-averaged collision
strengths is somewhat arbitrary. For transitions
with nonzero values of the oscillator strenght, the scaled collision
strengths were obtain by a linear fit to the given collision strength
and the high energy limit determined from the oscillator strength. For
forbidden lines, the scaled collision strength was assumed to be
constant with energy.
The Bhatia & Mason (1980a) radiative and collisional transition probabilities have been maintained.
Copyright The European Southern Observatory (ESO)