Production of noble gas isotopes by proton-induced reactions on lead and bismuth

We measured integral thin target cross sections for the proton-induced production of He-, Ne-, Ar-, Kr- and Xe-isotopes from lead from the respective reaction threshold up to 2.6 GeV. The production of noble gas isotopes from lead is of special importance for design studies of accelerator driven nuclear reactors and/or energy amplifiers. For all experiments with proton energies above 200 MeV a new mini-stack approach was used instead of the stacked-foil technique in order to minimise the influences of secondary particles on the residual nuclide production. About 420 cross sections for 23 nuclear reactions were determined. The phenomenology of the determined excitation functions enables us to distinguish between the different reaction modes fragmentation, hot and cold symmetric fission, asymmetric fission and deep spallation. Cross sections for the production of Ne-21 and Ar-38 measured below 100 MeV and 200 MeV, respectively, enable us to study nuclide production below the nominal Coulomb-barrier. The experimental data are compared to results from the theoretical nuclear model code INCL4/ABLA. While the model describes the production of He-4 reasonably well, it underestimates the cross sections for Ne- and Ar-isotopes produced via deep spallation and/or multi-fragmentation by up to two orders of magnitude. For the Kr- and Xe-isotopes the agreement between modelled and measured data strongly depends on the reaction mechanisms. While INCL4/ABLA describes the production of n-poor Kr-isotopes via hot-symmetric fission and the production of Xe-isotopes via asymmetric fission reasonably well, i.e. within a factor of 2, the discrepancies between modelled and measured cross sections for the n-rich Kr-isotopes produced via cold symmetric fission are significantly larger. For the Xe-isotopes produced via spallation, i.e. at energies higher than about 600 MeV, the model completely fails to describe the experimental data. Therefore, the comparison of measured and modelled thin target cross sections clearly indicates that experimental data are still needed because the predictive power of nuclear model codes, though permanently improving, does still not allow to reliably predict the cross sections needed for most applications and irradiation experiments remain indispensable. (C) 2004 Elsevier B.V. All rights reserved.