G4CASCADE: a data-driven implementation of neutron capture $\gamma$-rays in Geant4

The sensitivity of dark matter and neutrino experiments is largely driven by their target volume and their background event rate. The latter requires accurate models of background sources to help account for them in analyses and to develop discrimination techniques. Neutrons pose a particularly important background to experiments looking for signals in the keV to several MeV energy ranges, due to their ability to scatter on target nuclei or capture on nuclei in the detector and produce 2-11 MeV $\gamma$-rays in their de-excitation cascades. These $\gamma$-rays may cause backgrounds to MeV-scale signals, or they may be used to tag keV-scale neutron-induced nuclear recoils. Geant4 is one of the most-used Monte Carlo simulation toolkits in particle and nuclear physics; its currently-available modules for simulating (n, γ) signals, G4NDL and G4PhotoEvaporation, do not correctly reproduce known $\gamma$-ray lines and their correlations, leading to large discrepancies between data and simulations. We here present G4CASCADE, a custom-developed, data-driven Geant4 module that simulates (n, γ) de-excitation pathways with options for handling shortcomings in nuclear data. Benchmark comparisons to measured $\gamma$-ray lines show a significant improvement in the relative intensities of the $\gamma$-ray lines and consistent energy conservation, making G4CASCADE a helpful tool in ongoing and future dark matter and neutrino searches.