Reliable predictions of light charged particle production in spallation reactions are important to correctly assess gas production in spallation targets. In particular, the helium production yield is important for assessing damage in the window separ
ating the accelerator vacuum from a spallation target, and tritium is a major contributor to the target radioactivity. Up to now, the models available in the MCNPX transport code, including the widely used default option Bertini-Dresner and the INCL4.2-ABLA combination of models, were not able to correctly predict light charged particle yields. The work done recently on both the intranuclear cascade model INCL4, in which cluster emission through a coalescence process has been introduced, and on the de-excitation model ABLA allows correcting these deficiencies. This paper shows that the coalescence emission plays an important role in the tritium and $^3He$ production and that the combination of the newly develop
The spallation of $^{56}$Fe in collisions with hydrogen at 1 A GeV has been studied in inverse kinematics with the large-aperture setup SPALADIN at GSI. Coincidences of residues with low-center-of-mass kinetic energy light particles and fragments hav
e been measured allowing the decomposition of the total reaction cross-section into the different possible de-excitation channels. Detailed information on the evolution of these de-excitation channels with excitation energy has also been obtained. The comparison of the data with predictions of several de-excitation models coupled to the INCL4 intra-nuclear cascade model shows that only GEMINI can reasonably account for the bulk of collected results, indicating that in a light system with no compression and little angular momentum, multifragmentation might not be necessary to explain the data.
