GALACTIC COSMIC-RAYS AND THE EVOLUTION OF LIGHT-ELEMENTS

We reevaluate the contribution of Galactic cosmic ray (GCR) spallation to the abundances of the Li, Be, and B elements at various metallicit ies. We calculate absolute yields, as was formerly done by Reeves and collaborators in their pioneering studies. We discuss all the relevant parameters (spectrum shape and flux, escape length, cross sections, a nd spectral index) in the light of updated cosmic-ray data. We assume a dependence of the flux phi(t) on the supernova rate, phi(t) proporti onal to (dN(SN)/dt)(x), and we introduce the calculated production rat es of LiBeB in a standard model of chemical evolution of the Galaxy. W e find that the light elements are overproduced by GCR spallation with respect to their solar abundances unless (1) the cosmic-ray injection spectrum strongly flattens out at low energies or (2) the dependence of the cosmic-ray flux on the supernova rate is weak, i.e., x less tha n or equal to 1. We argue that constraint (1) should be expected in no nlinear shock acceleration, and is also supported by the relatively we ak ionization rate of the interstellar medium; constraint (2) could re sult from the fact that supernovae occur in associations in superbubbl es, where cosmic-ray fluxes of different supernovae do not add linearl y. We argue that even with substantial modifications, GCRs are not abl e to account for the BeB abundances at low metallicities. We then disc uss the new mechanisms proposed to account for these abundances. When incorporating a new mechanism of spallation, the above constraints are more demanding. Finally, we propose a solution to explain all the abs olute abundances of Li-6, Be-9, B-10, and B-11, in which Galactic cosm ic ray spallation and spallation by fast C and O nuclei originating fr om stellar winds and/or massive star explosions each produce about hal f of the solar/meteoritic abundances; neutrino-driven spallation in Ty pe II supernovae contributes to similar or equal to 15% of the B-11 me teoritic abundance.