thesis - IRS, The Infrared Spectrograph

42 CHAPTER 3: An ISO and IUE study of Planetary Nebula NGC 2440Ne IVNe VAr IVMg VTe (K)N IIO IIAr IIIO IIIO IVAr VS IIIN III.P. (eV)Figure 3.2–. Electron temperature versus ionization potential. The ratios used for each ion are indicatedin Table 3.6. The N II ratio plotted in this figure, has been derived using the line fluxes given byPerinotto & Corradi (1998), see Table 3.6. The shape of the curve is in accordance with the expectedphysical behavior (see discussion Sect 3.5.2)and Perinotto & Corradi (1998) lead to different T e which can slightly affect the relativeabundances derived for low ionization potentials (IP) ions.To derive the abundances for each ion a T e has been assumed as determined by the curvein Fig. 3.2. The temperature assumption is more important for ions where the abundance hasbeen derived using optical or ultraviolet data. Abundances derived from infrared lines areinsensitive to the adopted temperature. As can be seen in Fig. 3.2 the trend at high IP is notas well determined as at low IP. For ions with low IP the abundance has been derived mainlyusing ultraviolet or optical data, that are sensitive to the T e adopted, but where the curve iswell determined. On the other hand abundances of ions at high IP have been derived frominfrared lines and therefore are insensitive to the adopted T e .3.6 Chemical abundancesTo derive the ionic abundances a N e =4500 cm −3 an a T e according to the IP of each ion (seeSect. 3.5.2) have been assumed.3.6.1 AnalysisThe element abundances for carbon, nitrogen, oxygen, neon, magnesium, argon and sulfurhave been derived. Unfortunately for silicon and chlorine only one stage of ionization wasmeasured which is not enough to properly derive the abundance. The determination of theionization correction factor (ICF) is too uncertain. The ions for which abundances have been