THE ORIGIN OF COSMIC-RAYS ABOVE 10(18.5) EV

We discuss the origin of ultra-high-energy cosmic rays (UHECRs) in the energy range above 10(18.5) eV where it is indicated that the spectrum becomes protonic and extends smoothly up to at least 10(19.5) eV and is consistent with a smooth extension to 10(20.5) eV. The acceleration of the 10(19.5) eV component must occur within similar to 1 Gpc. We rule out the production and escape of protons and neutrons from active galactic nuclei. Composition arguments make unlikely any origin in metal-rich environments such as rich clusters and the inner regions of galaxies. We dismiss the canonical extended halo models since such halos are almost never observed although diffuse halos have been seen in QSO absorption-line studies of metallic absorption lines. Large-scale shocks from explosions and winds are analyzed including those originating both recently and at earlier cosmological epochs. Large-scale shocks can work well only if they occur in microgauss fields. Hot spots and cocoons of radio sources are a plausible source for UHECRs with the principal uncertainty in both cases being the adopted or inferred magnetic field strength. Cosmic shacks are formed as structure develops during gravitational collapse of primordial perturbations such as are found in the standard cosmological models in, for example, both pancakes and collisions of hierarchical merging subunits. Cosmic shocks can be good sites for UHECR acceleration if there is a primordial field of order greater than or equal to 10(-9) G or, again, if microgauss fields can be self-generated in shocks. Table 1 summarizes the results of our analysis of all the conventional possibilities and indicates that jets, radio source cocoons, structure formation in clusters superclusters, and large-scale structures are all reasonable sources for the production of UHECRs. We examine the possibility of a second-stage diffuse acceleration process from an ensemble of shocks that might explain the highest energy particles above 100 EeV by boosting their energy from 30 to 300 EeV into a diffuse isotropic component. Table 2 summarizes our results and indicates that at the present time there is as yet no such candidate ensemble of shocks that has both the power and the volume-filling factor to achieve this effect. The highest energy cosmic rays must therefore have been accelerated in one stage from their parent object.