The increasing interest that GEANT4 is gaining nowadays, because of its special capabilities, prompted us to address its reliability in neutronic calculation for the realistic and complex spallation target MEGAPIE of the Paul Scherrer Institute of Switzerland. In this paper we have specifically addressed the neutronic characterization of the active zone of this target. Three physical quantities are evaluated: neutron flux spectra and total neutron fluxes on target's z-axis, and the neutron yield as a function of the target's altitude and radius. Comparison of the obtained results with those of the MCNPX reference code and some experimental measurements have confirmed the impact of the geometrical and proton beam models on the neutron fluxes. It has also allowed to reveal the intrinsic influence of the code type. The resulting differences reach a factor of ~2 for the beam model and 4–18% for the other parameters cumulated. The analysis of the neutron yield has led us to conclude that: 1) Increasing the productivity of the MEGAPIE target cannot be achieved simply by increasing the thickness of the target, if the irradiation parameters are not modified. 2) The size of the spallation area needs to be redefined more precisely.

由于 GEANT4 的特殊功能，它现在越来越受到关注，这促使我们解决其在中子计算中的可靠性问题，用于瑞士 Paul Scherrer 研究所现实和复杂的散裂目标 MEGAPIE。在本文中，我们专门讨论了该目标活性区的中子表征。评估了三个物理量：目标 z 轴上的中子通量谱和总中子通量，以及作为目标高度和半径函数的中子产额。将获得的结果与 MCNPX 参考代码和一些实验测量的结果进行比较，证实了几何和质子束模型对中子通量的影响。它还允许揭示代码类型的内在影响。由此产生的差异对于梁模型达到约 2 倍，对于累积的其他参数达到 4-18%。对中子产额的分析使我们得出结论：1) 如果不修改辐照参数，则不能简单地通过增加靶材的厚度来提高 MEGAPIE 靶材的生产率。2) 需要更精确地重新定义散裂区域的大小。