References for Fe XI

.elvlc file (energy levels)

%energy levels(12,13,41): Jupen et al., MNRAS, 264, 627, 1993
%energy levels:  NIST Database for Atomic Spectroscopy, Version
   1.0, NIST Standard Reference Database 61, 1995.
%theoretical energy levels 1-9: Gupta & Tayal, 1999, ApJ, 5`0, 1078
%theoretical energy levels: Bhatia & Doschek (1996), ADNDT,64, 183
%produced as part of the Arcetri/Cambridge/NRL atomic data base
   collaboration 'CHIANTI'
   by P.R. Young  Feb. 1996 & Enrico Landi Apr. 1999
%comment:
   The third column of energies contains scaled superstructure
   values.

   Have had problems with the identification of some of the Fe XI
   lines. The following are proprosed.

   Level 37 has been proprosed as giving the 188.30 line (Jupen et
   al.). Don't think this is right, as the 3P2 - 3S1 transition is
   quite weak. Can not find any lines which correspond to the 3S1
   transitions, so have put the scaled superstructure value in.

   Level 38 (3P2) is identified from the strong line at 188.23.

   Level 39 (1P1) is identified as the strong line at 188.30.

   There's only one line from level 40 (3P0), whose identification
   I'm uncertain about (near 189.10 AA?). Have used the scaled 
   superstructure energy relative to the 3P1 level.

   Level 41 (3P1) has been identified from the 184.704 line seen by
   Jupen et al., two other transitions from this level match up
   nicely with the 189.123 and 189.733 lines seen by Behring et
   al.(1976).

.splups file (collision data)

%collision strengths:  Bhatia AK & Doschek GA, ADNDT 64, 183, 1996
% effective collision strengths levels 1-9: Gupta & Tayal, 1999, ApJ, 510, 1081
%comment:
   Selected transitions have been fitted in order to reproduce the 
   level balance of the complete collisional data. Essentially all 
   significant collision strengths involving the levels 1-5 and 
   14,20,23,24,25,30.

   This file contains fits to the SCALED Bhatia omega data. It was
   found that the 4 configuration collisional model does not
   accurately predict some important collision strengths, so it was
   decided to scale all the 3s2 3p2 - 3s2 3p 3d collision strengths
   (for which there was a corresponding oscillator strength) by the
   factor:

	(accurate "superstructure" oscillator strength)
	-----------------------------------------------
	     (original Bhatia oscillator strength)

   In particular, this improves the agreement of the 188.22, 188.30 lines
   with theory.
%
% produced as part of the Arcetri/Cambridge/NRL atomic data base collaboration 'CHIANTI'
%
% P.R.Young 30-Aug-97 and E.Landi 8-Apr-99

.wgfa file (radiative data)

%Produced for the CHIANTI atomic database by Peter Young, 7-Jul-2000.
                contact: pyoung@cfa.harvard.edu
%A values: 
   Obtained from a 13 configuration model of Fe XI used in "superstructure", 
   see comment below
%comment:
   The configurations used were:

	3s2 3p4, 3s 3p5, 3s2 3p3 3d, 3p6
	3s2 3p3 {4s, 4p, 4d, 4f}
	3s 3p4 3d, 3s2 3p2 3d2
	3p5 3d, 3s 3p3 3d2, 3s2 3p 3d3

   The most striking difference between this model and the 4 conf
   model occurs for the (2D*) 3P1, 3S1 and 1P1 levels. Essentially,
   the 3S1 level is pushed closer to the other two levels,
   increasing the amount of interaction between them. The data
   below, from "superstructure" shows this.

   The first column gives the levels; the second and third the
   theoretical energies for the 4 conf model and the 13 conf model;
   the next three columns give the percentage contributions of the
   three levels to each other: "a/b" means a is percentage for 4
   conf model, while b is percentage for 13 conf model.

				3S1	3P1	1P1
	3S1	532878	547264	97/71	-/-	-/10
	3P1	555764	553263	1/14	50/42	11/7
	1P1	564755	542936 	-/12	21/16	17/16

   Clearly the change of energies gives rise to different
   interactions. Most notably, 1P1 starts mixing with 3S1, whereas
   it did not before