next previous
Up: CHIANTI - an atomic


Subsections

2 The beryllium isoelectronic sequence

2.1 NeVII

 The new R-matrix calculations of Ramsbottom et al. (1994, 1995) are adopted in the CHIANTI 2.0 version. These calculations provide thermally-averaged collision strengths for transitions between the 9 lowest configurations ([2s 2], [2s2p ], [2p 2], [2s3s ], [2s3p ], [2s3d ], [2p3s ], [2p3p ] and [2p3d ]) corresponding to 46 fine-structure energy levels. The experimental energy levels are taken from the NIST database (Martin et al. 1995). Radiative transition probabilities and theoretical energy levels come from the same source as in the CHIANTI v. 1.0 version (Zhang & Sampson 1992; Muhlethaler & Nussbaumer 1976), although the original Ramsbottom et al. (1994, 1995) theoretical energies have been used for fitting the thermally-averaged collision strengths in CHIANTI format.

For some transitions the behaviour of the thermally-averaged collision strengths as a function of electron temperature is quite complex and the 5-point spline interpolation is not able to reproduce with the required accuracy their dependence on $T_{\rm e}$.For this reason values of the thermally-averaged collision strengths for temperatures lower than 104.7 K have been omitted from the database, so that the interpolation technique could reproduce adequately the behaviour of the curve. The omission of these points should not have any consequence as they are outside the temperature range in which Ne VII has significant population.

The use of the R-matrix thermally-averaged collision strengths instead of the Distorted Wave collision strengths has significant effects in the lower level populations and theoretical line intensities.

2.2 MgIX

A problem with the original CHIANTI model of Mg IX was that level 26 ([3p3d ] [3 F 4]) became incorrectly metastable on account of the absence of published allowed transition probabilities to de-populate the level, although the level would clearly decay via allowed transitions to the [2s3d ] [3 D 3] and [2p3p ] [3 D 3] levels. To correct this in v. 1.0 of CHIANTI, the gf values for these transitions were taken from the CHIANTI O V model (Hibbert 1980) and scaled with Mg IX energies to yield A-values for the transitions.

For a more correct approach, it was decided to use SSTRUCT (Eissner et al. 1974; see also Sect. 3 of Paper I) to calculate A-values for all transitions between the 46 levels of the CHIANTI model. This data is now in the CHIANTI 2.0 database. Care was taken to ensure that the SSTRUCT level ordering matched that in the CHIANTI database. The SSTRUCT model included the [3s 2], [3p 2], [3d 2], [3s3p ], [3s3d ] and [3p3d ] configurations in addition to the 9 configurations of the CHIANTI model already mentioned for Ne VII.

2.3 SXIII, ArXV, CaXVII, FeXXIII, NiXXV

The S XIII, Ar XV, Ca XVII, Fe XXIII and Ni XXV atomic models have been extended up to 46 fine-structure energy levels, including the n=3 configurations. These additional configurations were already part of the 1.0 version of the database for the lighter ions. Experimental energy levels come from the NIST database (Martin et al. 1995) for all the ions, but have been complemented with other sources where necessary: Khardi et al. (1994) (S XIII, Ar XV), Martin et al. (1990) (S XIII), Sugar & Corliss (1985) (Ca XVII) and Kelly (1987) (Ca XVII). Theoretical energy levels are taken from Zhang & Sampson (1992) and from Sampson et al. (1984).

Radiative and collisional transition probabilities for transitions from the n=3 levels have been taken from Sampson et al. (1984). The authors provide Coulomb-Born-exchange collision strengths calculated for nine values of the incident electron energy $\epsilon$ in threshold units, in the range $1 \le \epsilon \le 15$. Also electric dipole radiative transition probabilities for transitions from the n=3 levels are supplied.


next previous
Up: CHIANTI - an atomic

Copyright The European Southern Observatory (ESO)