Hybrid Active-Passive Galactic Cosmic Ray Simulator: in-silico design and optimization

TL;DR

This work designs and optimizes a hybrid active-passive GCR simulator at GSI to reproduce a GCR-like field behind lightweight shielding, enabling realistic ground-based studies of mixed-field space radiation. It combines energy-switchable $^{56}$Fe beams with complex and slab modulators, guided by an extensive Geant4-based optimization pipeline and a Basedata framework to target spectra behind 10 Al. A two-stage modulators-optimization strategy, together with a six-configuration weighting scheme, yields spectra closely resembling the reference GCR field, with LET and dose-equivalent metrics assessed via ICRP and NASA quality factors. A publicly available Geant4 phase-space converter facilitates user-specific simulations, broadening the simulator’s utility for shielding studies, endpoint research, and mission planning in deep-space scenarios.

Abstract

High-energy heavy-ion particle accelerators have long served as proxies for the harsh space radiation environment, enabling both fundamental life-science research and applied testing of flight hardware. Traditionally, monoenergetic high-energy heavy-ion beams have been employed for practicality, providing valuable datasets that underpin radiation risk and predictive computational models. However, such beams cannot fully reproduce the mixed-field nature of space radiation, motivating the development of realistic analogs for improved risk assessment and countermeasure evaluation in preparation for future deep-space missions to Moon or Mars. Spearheaded by developments at the NASA Space Radiation Laboratory, the GSI Helmholtzzentrum fuer Schwerionenforschung, supported by the European Space Agency (ESA), has established advanced space radiation simulation capabilities in Europe. Here, we present the design, optimization, and in-silico benchmarking of GSI's hybrid active-passive Galactic Cosmic Ray (GCR) simulator, together with a computationally optimized phase-space particle source for Geant4, which is available to external users for their own simulation studies and experimental planning.

Figures (9)

• Figure 1: The target ($Z=26$) kinetic energy spectrum cannot be reproduced by a single modulator and was therefore subdivided to match the predefined primary beam energies of the accelerator. The high-energy region above 1000 was excluded, corresponding to the accelerator's operational limit. Transitions between adjacent sections were smoothed using an error-function (ERF) to avoid sharp boundaries and minimize artifacts during the optimization.
• Figure 2: To generate the modulator structure for a single pin, the material weights are inverted and re-calculated to circular areas with different target thicknesses and lofted in FreeCAD to create a volume (Panel A). Afterwards, this volume is subtracted from a rectangular base structure (Panel B) with lateral dimensions of 5 x 5 and an energy-dependent height (\ref{['fig:methods:MC_scheme_full_setup']}), resulting in a complex hole and pin-like shape. The cross section of such a modulator structure is shown in Panel C. Depending on the desired lateral dimensions of the modulator, this basic modulation structure is arranged in a suitable sized matrix to reach the final dimensions.
• Figure 3: Schematic representation of the GCR simulator along the beamline, implemented in Geant4, including all components and their relative distances. The beam is generated from a squared spatial distribution with $x \in \left[-40,\, 40 \right] \, \unit{\mm}$ and $y \in \left[-40,\, 40 \right] \, \unit{\mm}$, orthogonal to the beam direction. The first element along the beam path is the steel mesh modulator, followed by complex modulators designed for primary beam energies of 1, 0.7, and 0.35 (shown in green), slab modulators (two steel-304L slabs in blue and a polyethylene (PE) slab in yellow), and the FRANCO modulator. Except for the steel mesh modulator, which is permanently installed, all modulators are positioned along the beamline according to the specific configuration defined for each exposure. Three scoring planes are placed downstream of the FRANCO modulator to evaluate the spatial homogeneity of the radiation field. All dimensions in the sketch are given in .
• Figure 4: Relative elemental abundances in the reference GCR spectrum (light blue) compared with those generated by the GCR simulator (dark blue). The comparison highlights the capability of simulator to reproduce the elemental composition of the galactic cosmic ray environment across a wide range of nuclear species.
• Figure 5: Differential kinetic energy spectra of the reference GCR (light blue) and of the GCR simulator (dark blue), both obtained by summing over all ion species ($Z\leq26$). For the GCR simulator, the separate contributions of protons ($Z=1$, red), helium nuclei ($Z=2$, orange), and HZE nuclei ($Z \geq 3$, dark green) are also displayed.