Geant4 based library SCoRe4 for Surface Contamination and Roughness Effects simulations in rare event search experiments

TL;DR

This work tackles the mismatch between real micro-scale surface roughness and flat Geant4 surfaces in rare-event searches. It introduces SCoRe4, a Geant4-based library with modules Surface Roughness, Portal, and Particle Generator/Shift that generate rough surface patches from experimentally measurable parameters across scales from mm to m while keeping computation efficient. A key result demonstrates how spikes of about $10\,\mu\mathrm{m}$ on a $3\times3\mathrm{cm}$ silicon surface alter energy deposition for a $^210\mathrm{Po}$-related $5.3\mathrm{MeV}$ alpha source, consistent with observed background effects and improving modeling fidelity. The open-source library, GPL-3.0, integrates with existing Geant4 setups and provides a framework to simulate surface contamination more realistically in rare-event physics.

Abstract

Surface simulations are important for accurately modeling particle interactions in experiments where background contributions from surface contaminants can significantly affect detector performance. In rare event searches, such as dark matter or neutrinoless double beta decay experiments, standard Geant4 simulations typically assume perfectly smooth surfaces, neglecting the microscopic roughness that exists in real materials. This simplification can lead to inaccurate predictions of energy deposition. To address this limitation, I developed SCoRe4, a Geant4-based library designed to simulate more realistic surface roughness based on experimentally measurable parameters. The code allows users to generate patches of simplified rough surface geometries across a wide range of scales - from square millimeters to square meters - while maintaining computational efficiency. SCoRe4 is open source and can be easily integrated into existing Geant4 setups. This work presents the structure, implementation, and example application of SCoRe4,as well as its potential use in improving the accuracy of background modeling in rare event physics.

Figures (9)

• Figure 1: Surface profile of a diffused crystal surface. Structures in the µm scale are visible. Figure is taken from Gruener2024
• Figure 2: Visualization of different shape implementations for spikes, which can be of any size that is allowed in .: (a) simplest form of a spike, basis is a tetrahedron. (b) multiple layers of tetrahedrons form the spike, the outer surface approximates a squared function. (c) multiple layers of tetrahedrons form the spike, the outer surface approximates $1/x$.
• Figure 3: Visualization of 3x3 spikes placed at the surface of a target volume to simulate a patch of rough surface. The generated spikes can be of any size allowed in . The size is controlled via setting a width and height for the spiked.
• Figure 4: The flowchart shows the generation of a patch of rough surface, starting with passing the user defined surface description to the class. Next the chart is divided in two parts: 1) Describe represents the actions of the class Describer which translates the user description into a list of needed G4Solids inclusive their parameters and location. This list is passed to the Assembler class which generates all the needed solids and adds them to a G4MultiUnion. In parallel the FacetStore is filled.
• Figure 5: 2D sketch of the function of the implemented Portal representing two subworlds. The green dotted line is the trajectory of the particles entering the portal and the subworld on the right side and leaving them on the left. The orange dotted line is the imagined trajectory of the particle crossing the portal. The red dashed line divides the portal into two subworlds. The simulated particle undergoes the following steps: A) it enters the portal on the right side at A$_1$ and is ported to point A$_2$ of the subworld without a change in momentum. B) after crossing the subworld it leaves the volume at B$_1$ and enters the trigger. This activates the periodic teleportation and the particle is set to point B$_2$ without a change in momentum. C) after crossing the subworld for the second time it leaves the subworld's volume at C$_1$ and enters the trigger again. The particle is teleported to C$_2$ where it leaves the portal, without a change in momentum. The dimension of the subworlds and the portal can be of any size allowed by . The trigger volume ensures, a correct activation of particle teleportation.