[Paper Review] Peculiarities in the Simulation of Optical Physics with Geant4

E. Dietz-Laursonn|arXiv (Cornell University)|Dec 15, 2016

Semiconductor Lasers and Optical Devices1 references17 citations

TL;DR

This paper identifies and explains critical peculiarities in Geant4's optical physics simulation, particularly regarding photon tracking, material properties, and surface definitions. It emphasizes that incorrect handling of energy-dependent spectra, rise/decay times, interpolation methods, and reflectivity models can lead to non-physical results, especially for users unfamiliar with Geant4's internal logic and default behaviors.

ABSTRACT

Geant4 is a complex and widely-used software toolkit for the simulation of the passage of particles through matter and the interactions they undergo. It contains very extensive and flexible optical physics capabilities. These allow for the specification of a lot of properties for materials and surfaces as well as for the optical physics process itself. Because of this large variety of possible adjustments, there is a number of peculiarities of the optical physics in Geant4, which the user has to be aware of in order to avoid incorrect simulation results due to user mistakes. As these peculiarities are not always easy to recognise, this can be a serious problem, especially for less experienced users of Geant4. In order to assist users of optical physics in Geant4 in avoiding mistakes, this paper gives a summary of peculiarities of the optical physics in Geant4.

Motivation & Objective

• To address common user errors in Geant4 optical physics simulations caused by subtle, non-obvious behaviors in the toolkit’s design.
• To highlight how improper specification of optical properties—especially energy-dependent spectra—can lead to incorrect or non-physical simulation outcomes.
• To clarify the distinction between material and surface properties in Geant4, particularly regarding refractive index, complex refractive index, and reflectivity.
• To explain the impact of default settings such as zero rise time and delta-function decay time on signal timing accuracy.
• To guide users in correctly modeling optical processes like Cherenkov radiation, scintillation, and wavelength shifting in complex detector geometries.

Proposed method

• Systematic identification and documentation of Geant4’s optical physics peculiarities through analysis of its internal processing logic and user-reported issues.
• Use of simulation tests (e.g., figure 1) to demonstrate interpolation behavior differences between emission spectra (constant mean interpolation) and other spectra (linear interpolation).
• Explanation of the role of track identification numbers in preventing misinterpretation of multi-segmented trajectories from optical photon generation.
• Clarification of the distinction between REFLECTIVITY and TRANSMITTANCE variables for dielectric-dielectric and dielectric-metal surfaces.
• Demonstration of how to properly activate finite rise and decay times in the PhysicsList via code examples from external references.
• Analysis of the role of WLSABSLENGTH and ABSLENGTH in modeling wavelength-shifting materials with overlapping absorption mechanisms.

Experimental results

Research questions

• RQ1 How do incorrect energy-sorted spectra or mismatched energy ranges in optical properties affect simulation results in Geant4?
• RQ2 Why do default settings for rise time and decay time in scintillation and WLS processes lead to non-physical signal shapes, and how can they be corrected?
• RQ3 What causes unexpected interpolation behavior in emission spectra, and how can users avoid artifacts from low-resolution data?
• RQ4 Why does the REFLECTIVITY variable behave differently for dielectric-dielectric versus dielectric-metal surfaces, and what is the correct alternative for dielectrics?
• RQ5 How does the internal track handling of optical photons affect trajectory interpretation, and what role does the track ID play in avoiding double-counting?

Key findings

• Optical properties in Geant4 must be specified as functions of rising photon energy and on identical energy ranges; mismatched ranges cause simulation errors.
• Rise time is deactivated by default, resulting in zero simulated rise time regardless of user input, requiring explicit activation in the PhysicsList.
• The WLS decay time is a delta function by default; to simulate exponential decay, the user must explicitly enable this feature in the PhysicsList.
• Emission spectra are interpolated using a constant mean value between points, not linear interpolation, requiring high-resolution data to avoid artifacts.
• G4OpticalPhoton creation delays the mother particle’s processing, leading to multiple track segments; this can cause double-counting if track IDs are not used.
• G4SensitiveDetectors respond to both incident and reflected optical photons, a behavior rooted in Geant4’s internal treatment of optical surfaces and detection.