Astron. Astrophys. 328, 107-120 (1997)

6. C and O lines and line ratio

Since ¹² C and ¹⁶ O are the most abundant nuclei in the winds of massive stars after protons and particles, the two most intense gamma-ray lines predicted are the ¹² C line at 4.438 MeV and the ¹⁶ O line at 6.129 MeV, in agreement with the COMPTEL observations in Orion. These two lines also constitute the main contribution to the total gamma-ray flux emitted in the range 3-7 MeV, to which we have normalised all our calculated fluxes. We have used the updated cross sections of Tatischeff et al. (1996).

6.1. The ¹² C line at 4.438 MeV

The first excited state of ¹² C can be populated by direct excitation from EP-ISM interactions, or by the breaking of heavier nuclei in spallation reactions, mainly of ¹⁶ O and ¹⁴ N. For all the compositions considered here, however, the main contribution to the 4.438 MeV line is the in-flight excitation of energetic ¹² C nuclei on ISM protons, leading to essentially broad line emission. Figs. 6 and 7 show the detailed contribution of all the production mechanisms representing more than of the total gamma-ray line flux, in two extreme cases. The first corresponds to the late-WC composition, which is so poor in protons and alpha particles that more than of the 4.438 MeV line flux is due to inverse reactions involving carbon and oxygen; and the second corresponds to the mean-wind composition of an OB association with twice solar initial metallicity and enhanced mass loss rates (Geneva group's model E). In this case direct reactions contribute to about of the total ¹² C production, and EP's ¹⁴ N nuclei excitation provides the second most important contribution to the gamma-ray line flux, for any break energy MeV/n.

Fig. 6. Detailed production of the 12 C 4.438 MeV line as a function of the injection break energy for a late-WC composition with solar initial metallicity. The labels indicate the reactions considered, the first quoted species being the projectile.

Fig. 7. Same as Fig. 6 for a mean-OB composition with models E and twice solar metallicity.

In order to distinguish the broad and narrow components of the 4.438 MeV line (leaving aside the possible line splitting; Bykov et al. 1996, Ramaty et al. 1997), we show in Fig. 8 the ratio of inverse to direct contributions for various EP compositions, excluding GR and late-WC compositions for which this ratio is greater than 100, as will as solar composition for which it is of order 0.3. The wide range of variation of this ratio demonstrates that any instrument capable of measuring the ¹² C line profile, such as INTEGRAL's SPI, would be able to set strong constraints on the actual EP composition.

Fig. 8. Inverse-to-direct component ratio for the 12 C line at 4.438 MeV, for various EP compositions. The labels indicate the model and the metallicity used. The latter is solar unless explicitely specified.

6.2. The ¹⁶ O line at 6.129 MeV

Fig. 9 and 10 show the detailed production of the 6.129 MeV line resulting from the de-excitation of ¹⁶ O nuclei produced in their first excited level by direct excitation and ²⁰ Ne spallation for two EP compositions corresponding to the same model (mean-OB composition with models C and IMF index ), but with either solar or twice solar initial metallicity. The differences are quite striking.

Fig. 9. Detailed production of the 16 O 6.129 MeV line as a function of the injection break energy for a mean-OB composition at solar metallicity.

In the case of solar initial metallicity, the EPs are much richer in ¹⁶ O than the ambient ISM (O/H versus ). As a result, the encounters of energetic ¹⁶ O nuclei with ISM protons are much more frequent than that of energetic protons with ISM ¹⁶ O nuclei. This is not anymore the case at twice solar metallicity, because the EPs are richer in protons (O/H versus ). As a consequence, direct and inverse excitations involving protons contribute to the gamma-ray line flux at about the same level (Fig. 10). This effect is even more pronounced for the -¹⁶ O reactions, because of the strong enhancement of the mean-wind helium abundance when passing from solar to twice solar metallicity. The direct-to-inverse ratio then reverses, resulting in a dominating narrow line emission. This prediction could be tested by the variation of the ¹⁶ O line profile as a function of the Galactic longitude, since most of the metal-rich W-R stars should be concentrated in the 4 kpc ring.

Fig. 10. Same as Fig. 9 at twice solar metallicity.

Fig. 11 shows the global inverse-to-direct ratio, or equivalently the broad-to-narrow line component ratio for various EP compositions. As in the case of the ¹² C line at 4.438 MeV, it can be seen that this ratio is very sensitive to the source composition.

Fig. 11. Inverse to direct component ratio for the 16 O line at 6.129 MeV, for various EP compositions. The labels indicate the models and the metallicity used (solar metallicity unless explicitely specified.

At any rate, we predict a 6.129 MeV line significantly narrower than the 4.438 MeV line for any of our mean wind compositions (from either individual stars or OB associations). In particular, the direct reactions are always found to contribute to at least of the total line emission, and actually dominate in the case of a twice solar metallicity (inner Galaxy) or for models with enhanced mass loss rates during the main-sequence phase. This represents a distinctive feature of our models with respect to, e.g., the late-WC model which presents a very high ¹⁶ O abundance.

6.3. The ¹² C /¹⁶ O line ratio

Apart from the line profiles discussed above, the ¹² C /¹⁶ O gamma-ray line ratio is one of the most relevant observable for gamma-ray spectroscopy. It should indeed be accessible quite easily to the INTEGRAL's spectrometer SPI in Orion, and also presumably in the diffuse Galactic emission. The data collected by COMPTEL provide a first constraint on this line ratio in Orion (Bloemen et al. 1997). Our rough guess estimate is C /¹⁶ O .

As can be seen in Fig. 12 and Fig. 13, the ¹² C /¹⁶ O ratio is quite sensitive to the chemical composition of the EPs. If our rough estimate is correct, one can already exclude many compositions, among them the solar system (SS), grain (GR), supernova ejecta (SN35), as well as late-WC compositions, whatever the value of the break energy may be. On the contrary, our mean-OB compositions seem to provide a more adequate line ratio. In the case of a star with enhanced mass loss rate (model E), a break energy MeV/n is required.

Fig. 12. 12 C /16 O gamma-ray line ratio as a function of the injection break energy , for different EP compositions.

Fig. 13. Same as Fig. 12 with mean-wind compositions from individual stars.

Although our estimate is admitedly uncertain, we point out that whatever the value of the ¹² C /¹⁶ O line ratio will prove to be, its measurement will allow us to distinguish between mean-WC and late-WC compositions, and will provide a strong argument to exclude (or favour) compositions such as SS, GR or SN35.

Finally, we note that in a general way, the ¹² C /¹⁶ O line ratio decreases for increasing values of the injection break energy . This is due to the high excitation threshold of ¹⁶ O. However, this does not apply to the SN35 composition. In this case, indeed, the ¹⁶ O abundance is so high with respect to ¹² C ( times greater) that the main contribution to the 4.438 MeV line is of spallative origin (except for the lowest values of ). Both ¹² C and ¹⁶ O lines are thus due to the same collisions, namely ¹⁶ O p, so that the previous argument doesn't hold.

© European Southern Observatory (ESO) 1997

Online publication: March 24, 1998

helpdesk.link@springer.de