On the atomic data for Cl III

Valentina Luridiana (IAC, España)

Valentina Luridiana

We present a comparison among some published transition probabilities and collision strengths for Cl III, which display large differences in the density diagnostics. Complete references are given below.

Available A-values

M83 is used by the photoionization code Cloudy (v. c10.0), KS86 by the IRAF package for the analysis of emission lines nebular (except for the 4-3 transition (2D5/22P1/2), which is not given in KS86 and is thus taken from M83; this combination is referred to as M83KS86 in a table further below). They are compared in this table and here for a quick glance:

Trans. M83 KS86
A(2,1) 4.83E-3 3.44E-3
A(3,1) 7.04E-4 8.07E-4
A(3,2) 3.22E-6 3.08E-6
A(4,1) 0.305 0.122
A(4,2) 0.303 0.360
A(4,3) 0.100
A(5,1) 0.754 0.691
A(5,2) 0.323 0.339
A(5,3) 0.316 0.387
A(5,4) 7.65E-6 7.08E-6

There are differences of 10-20% in most transitions and especially large differences in A(2,1) (40%) and A(4,1) (250%). A(2,1) is particularly relevant since it directly influences density determinations.

Available effective collision strengths (upsilons)

  • M83 (oldest, only one value for all T; range: unknown)
  • BZ89 (T range 2,000-20,000 K, plus a point at 50,000 K which is defined as uncertain by the authors)
  • RBK99 (table data provided by the authors)

Cloudy (v. c10.0) uses M83; nebular uses the BZ89 data in the T = 2,000 to 20,000 K range.

Effective collision strengths are nearly flat around T = 10,000 K; here is a comparison at T = 10,000 K:

Trans. BZ89 M83 RBK99
ϒ(2,1) 1.359 1.26 1.432
ϒ(3,1) 2.046 1.88 2.149
ϒ(4,1) 0.418 0.627 0.531
ϒ(5,1) 0.837 1.26 1.063
ϒ(3,2) 4.519 3.19 4.601
ϒ(4,2) 1.653 1.24 1.690
ϒ(5,2) 2.192 1.91 2.434
ϒ(4,3) 1.563 1.38 1.747
ϒ(5,3) 4.203 3.33 4.439
ϒ(5,4) 1.755 1.34 1.744

The typical difference between M83 and KS86 is ~50%. There are no dramatic differences between BZ89 and RBK99. Some differences below 10,000 K, but generally less than 20%. RBK99 also give an account of such differences. BZ99 appear to be adequate, unless one wants to go to higher T.

Effects on density diagnostics

The following is a comparison at T = 10,000 K (20,000 K) of the [Cl III] density predicted by nebular with different combinations of the input atomic data (“F” = calculation failed):

λ5517/λ5537 M83KS86 + M83 M83 + M83 M83KS86 + BZ89 M83 + BZ89 M83K86 + RBK99 M83 + RBK99
0.25 F 129650 F 102200 F 97500
0.50 28400 14460 22250 11100 21280 10700
0.50 (30040) (16660) (24400) (13210) (23020) (12520)
0.75 8200 5640 6480 4390 6190 4210
1.00 3150 2390 2570 1920 2420 1820
1.25 840 700 790 630 710 580

Summing up, it seems that the examined collision strengths are equivalent in the relevant range for nebular analysis (particularly BZ89 and RBK99), whereas the A-values make a large difference, especially at large densities. If you’re aware of any other dataset worth including in the comparison, please post a comment.


This entry was posted in Atomic data applications and tagged , . Bookmark the permalink.

One Response to On the atomic data for Cl III

  1. Pascal Quinet says:

    Transition probabilities for forbidden lines within the ground configuration of low-Z P-like ions (including Cl III) were also computed by Fritzsche et al., ApJ 518, 994 (1999). These results should be the most accurate ones presently available.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s