All Figures

Figure 1: MM02 stellar yields for 14N, for the whole stellar mass range, for different metallicities. The yields of MM02 for stellar models where rotation is not taken into account are shown as open squares. Filled squares stand for models with rotation. Stellar yields are shown for 3 different values of metallicities (solid lines: solar, dashed line: Z= 0.004 and long dashed line: Z= 0.00001). The asterisks connected by the long-dashed line show the stellar yields adopted only for the lowest metallicity case in our heuristic model (see text).

Open with DEXTER


In the text

Figure 2: Solar vicinity diagram log(N/O) vs. log(O/H)+12. The data points are from Israelian et al. 2004 (large squares), Spite et al. (2005) (asterisks). Also shown is the very metal-poor star found by Christlieb et al. (2004). Solid curves show the prediction of MW models computed with vdHG+WW yields (thin solid line) and MM02 yields (thick solid line). The latter flattens for due to the contribution by massive stars to the nitrogen production at low metallicities. The dotted curve shows a model computed according to the suggestion of Matteucci (1986), where all massive stars, in all metallicities, contribute a fixed amount of primary nitrogen of 0.065 . The dashed line shows the prescriptions of our heuristic model computed with MM02 yields but assuming that massive stars with metallicities less or equal to 0.00001 produce much more nitrogen than the quantities computed by MM02, as shown in Fig. 1.

Open with DEXTER


In the text

Figure 3: Log(C/O) vs. log(O/H) diagram. The data are from Spite et al. (2005 - asterisks), Israelian et al. (2004 - squares), Nissen (2004 - filled pentagons). The different curves show our model predictions computed with different stellar yields as follows: a) dot-dashed curve - a model computed according the prescriptions of Akerman et al. (2004); b) thin solid curve - vdHG97 and WW95, where the latter refer to their solar tables and c) dashed line - vdHG97 and WW95 (where in this case the oxygen as a function of metallicity was adopted as suggested by François et al. 2004). This figure shows an alternative to the model suggested by Akerman et al. (2004) to obtain an upturn of C/O at low metalliticies.

Open with DEXTER


In the text

Figure 4: The data points are from Cayrel et al. (2004), Spite et al. (2005) (asterisk) and Israelian et al. (2004) (squares). The dashed line shows the predictions of our "heuristic model''. The dot-dashed lines refer to models computed with Chieffi & Limongi (2002, 2004) for metallicities below 10-6 and a top-heavy IMF (see text). In the lower panel a model that assumes a constant N production in massive stars of all metallicities (Matteucci 1986) is also shown. In this figure the models are normalized to the solar N and Fe of Holweger (2001) and to the C of Allende-Prieto et al. (2001).

Open with DEXTER


In the text

Figure 5: Abundance gradients of C/O, N/O and C/N predicted by models adopting vdHG+WW yields (as model 7 of Chiappini et al. 2003 - thin solid line) and the same models computed with MM02 yields (as described in CMM03 - thick solid line). The dashed line barely seen in the bottom diagram corresponds to the predictions of our heuristic model (see text). In the middle panel both models overlap. Therefore, it is clear that the dominant factor of the N/O gradient in the MW is nitrogen production in LIMS and not the primary nitrogen in massive stars. For the abundance data see Chiappini et al. (2003), and references therein. Here we added the recent abundance data of Daflon & Cunha (2004, large asterisks).

Open with DEXTER


In the text