5 Conclusions

We have presented a theoretical investigation of the 2.22 MeV line emitted from X-ray binary systems neutrons are captured in the secondary's atmosphere. These neutrons are produced in the accretion disk by nuclear spallation of helium nuclei. The rate of production of neutrons depends on various parameters set by the accretion disk model, i.e. the temperature, density, and composition of the accreted material. Calculations of the line intensity normalized to the neutron production rate have been performed for two simple secondary models and for several X-ray binary geometries.

We showed that, due to the rotation of the secondary, the intensity of the 2.22 MeV radiation is periodic, and the line centroid is shifted by the Doppler effect. The mean intensity does not vary significantly with the direction of observation with respect to the binary system frame.

According to our model, the spectral (Figs. 9 and 10) and temporal (Figs. 6-8) analyses of the 2.22 MeV line flux would provide valuable insights into the characteristics of the binary system (separation, inclination with respect to the observer, masses, accretion rate and disk structure, etc.). If one further knows the X-ray luminosity, the geometry of the X-ray binary system, and its distance from Earth, then a measure of a 2.22 MeV line flux from it can set constraints on the neutron production rate and consequently on the accretion disk models (ADAF, ADIOS, SLE, etc.).

At 2.22 MeV the narrow $\gamma$-ray line sensitivity of the SPI spectrometer of the INTEGRAL mission is expected to be 7-10 $\times 10^{-6}$ photons s^-1 cm^-2 for an observation time of 10⁶ s (Jean et al. 1999). The 2.22 MeV fluxes estimated for the simple secondary-star models and for the various accretion disk models used here are found to be measurable by the spectrometer SPI in some cases: mostly if the accretion rate is large enough ( $\dot M \approx$ 10 $^{-8}~M_{\odot}$/yr) and the viscosity is significant ( ). In case of a detection, the SPI spectral resolution ($\approx$3 keV at 2.22 MeV) would allow the measurement of the broadening of the line, which is due to the rotation of the secondary, if the separation and the inclination of the X-ray binary are not too large.

In a future work, the 2.22 MeV line flux and shape will be estimated for known X-ray binary systems (e.g. A0620-00, etc.) using all accurate, available information on their characteristics (distance, separation, composition, etc.). The calculated fluxes will be compared with upper-limit fluxes obtained with COMPTEL (Van Dijk 1996), and correlation with other gamma-ray lines will be investigated. Indeed, the neutrons that irradiate the secondary star can also produce Be and Li isotopes in its atmosphere. This process was proposed by Guessoum & Kazanas (1999) to explain the overabundance of Li in some soft X-ray transients. Both ⁷Li and ⁷Be emit gamma-ray lines (at 0.478 and 0.429 MeV), and ⁷Be decays into ⁷Li^* (with a half-life of 56 days), the latter then produces 0.478 MeV radiation upon de-exciting; such a delayed gamma-ray emission, although weak, would be a clear signature of the process. We can thus expect a certain correlation of the 2.22 MeV emission with these gamma-ray lines, providing supplementary insights into the physics of the binary system.

Acknowledgements

We thank D. Kazanas for his useful comments. N. Guessoum would like to acknowledge a short visiting professor invitation by Université Paul Sabatier and the American University of Sharjah for a summer research grant; he also wishes to thank Prs. P. von Ballmoos and G. Vedrenne for the facilities provided at the Centre d'Étude Spatiale des Rayonnements (Toulouse, France) where much of this work was conducted.