SPIROS: Streamlined, Precise, Intuitive, and Rapid Optical Simulator for particle physics detectors

Tatsuya Kikawa · 2026

Read SPIROS: Streamlined, Precise, Intuitive, and Rapid Optical Simulator for particle physics detectors by Tatsuya Kikawa (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-08279.txt.

Introduction

Kikawa presents SPIROS, a sleek, precise, easy, and rapid optical tool for particle physics detectors. Unlike broad systems like GEANT4, SPIROS offers a light engine and a friendly interface tuned for optical processes: flashes, Cherenkov emission, and photon flow with bounce, bending, spread, uptake, and catching. Detector geometries import straight from 3D CAD models, and every setting, matter, faces, sources, and sensors alike, is set in one human-readable input file. Checks against GEANT4 agree well on photon making.

Catalog record

CREATORTatsuya Kikawa
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE33695
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.08279

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

SPIROS: Streamlined, Precise, Intuitive, and Rapid Optical Simulator for particle physics detectors Tatsuya Kikawa Abstract This paper presents SPIROS (Streamlined, Precise, Intuitive, and Rapid Optical Simulator), a dedicated optical simulation tool developed for the design and analysis of particle physics detectors. Unlike general-purpose frameworks such as GEANT4, SPIROS offers a lightweight simulation engine and a user-friendly interface optimized for optical processes, including scintillation, Cherenkov emission, and photon transport with reflection, refraction, scattering, absorption, and detection. Detector geometries can be directly imported from 3D CAD models, and all configurations including materials, surfaces, sources, and sensors are specified via a single human-readable input file. Validation against GEANT4 shows excellent agreement in photon generation and propagation behaviors, while benchmark tests demonstrate that SPIROS runs more than two times faster for typical detector configurations. The software has already been applied to multiple neutrino experiments, including T2K, NINJA, and AXEL, for detector design, performance studies, and optimization. SPIROS is open-source and freely available at https://github.com/tkikawa/spiros. 1 Introduction Optical simulations play a crucial role in the design and optimization of detectors used in particle physics experiments, particularly those involving scintillation or Cherenkov light detection. Accurate modeling of photon propagation, reflection, scattering, and detection is essential for understanding detector response and improving reconstruction performance. GEANT4 [ 1 , 2 , 3 ] , a widely used toolkit for simulating the passage of particles through matter, includes capabilities for detailed optical photon tracking. However, configuring GEANT4 for optical simulations is often highly complex and time-consuming. It requires in-depth knowledge of the framework and substantial effort to define the detector geometry, materials, surfaces, and physics processes involved. As a result, performing iterative design studies or rapid prototyping with GEANT4 can be prohibitively inefficient in practice. On the other hand, several commercial optical simulation tools exist, but they are typically tailored to applications in engineering or medical imaging, and lack essential features or flexibility required for particle physics experiments. Moreover, their closed-source nature and limited support for custom detector geometries and physics models pose significant challenges. To address these issues, I have developed SPIROS, a streamlined, precise, intuitive, and rapid optical simulator specifically designed for particle physics experiments. This tool enables rapid simulation of optical processes through a lightweight and highly optimized execution engine. It offers an intuitive user experience by supporting human-readable configuration files, minimal setup requirements, and direct import of detector geometries from 3D CAD models. The framework’s streamlined architecture facilitates fast iteration and prototyping, while its physically accurate modeling of optical interactions ensures precise and reliable results suitable for both design and analysis tasks. Simulation results can be explored using a built-in interactive event display for visual inspection of the optical behavior, or exported to data files for detailed quantitative analysis. SPIROS is released as open-source software and is freely available at https://github.com/tkikawa/spiros, along with a user manual and several example configuration files. This paper presents the design philosophy, software architecture, and computational performance of SPIROS. Its simulation results are validated through comparison with GEANT4, and its applicability is demonstrated through use in the design and optimization of actual particle physics experiments. The remainder of this paper is organized as follows. Section 2 describes the design and implementation of SPIROS, including its architecture, configuration system, and geometry handling. Section 3 presents validation results through a comparison with GEANT4 simulations. Section 4 evaluates the computational performance of SPIROS relative to GEANT4. Section 5 demonstrates applications of SPIROS in actual particle physics experiments. Finally, conclusions are given in Section 6 . 2 Design and Implementation 2.1 Architecture Overview The simulation engine is implemented in C++ and leverages two key external libraries: ASSIMP (Open Asset Import Library) [ 4 ] and ROOT (a data analysis framework developed by CERN) [ 5 , 6 ] . ASSIMP is used to import geometry from various 3D CAD file formats, allowing users to define detector layouts using familiar design tools such as SolidWorks or Autodesk Inventor. ROOT is employed for both simulation output and event display, providing structured data storage and interactive visualization capabilities, and is also used internally to generate Landau-distributed fluctuations for scintillation modeling. The software supports both optical-photon and charged-particle source modes. Photon transport includes refraction, reflection (specular and diffuse), absorption (bulk and surface), scattering, and detection, while charged particles can produce scintillation or Cherenkov light based on material properties. These processes are handled through a modular transport engine that governs photon and particle behavior within the geometry. The Mersenne Twister pseudorandom number generator [ 7 ] is used to sample physical processes efficiently, ensuring statistically reliable and reproducible simulations. 2.2 Implementation of Optical and Particle Physics Processes 2.2.1 Optical Photon Transport Each optical photon is propagated step-by-step through the detector geometry based on geometric boundary tracking and interaction probabilities. The polarization state of each photon is also tracked throughout the simulation and updated appropriately at each optical interaction. At material boundaries, the behavior of the photon is determined by the surface properties of the material: • Normal medium surfaces are treated using Fresnel equations to compute the reflection and refraction probabilities based on the angle of incidence and refractive indices of adjacent materials, with the direction of refraction determined by Snell’s law. These interactions include polarization-dependent reflection and transmission coefficients. • Mirror surfaces reflect photons specularly. • Diffuser surfaces apply Lambertian (cosine-distributed) reflection [ 8 ] . • Absorber surfaces terminate photons upon contact. • Detector surfaces also terminate photons, but may register them as "detected" for output analysis based on user-defined detection efficiency. • Converter surfaces randomize the direction of photons isotropically upon entry into the material, effectively simulating the absorption and re-emission processes as in wavelength shifting. SPIROS also supports material mixtures, where the boundary behavior is defined by a probabilistic combination of two or more material types. Upon boundary contact, the photon undergoes a Monte Carlo selection based on these weights and then follows the corresponding interaction model. For example, a surface may be defined as 70% mirror and 30% absorber, meaning that photons incident on the surface will undergo specular reflection with 70% probability and absorption with 30% probability. This mechanism enables simulation of partially coated surfaces, imperfect reflectors, half-mirrors, or rough surfaces without explicit geometric modeling of microstructure. Within each material, bulk optical properties are modeled probabilistically. SPIROS assumes wavelength-independent optical parameters for simplicity, which is sufficient for many detector design and optimization studies. The step length to the next interaction is sampled based on exponential attenuation and scattering lengths. If the step results in absorption or scattering, the photon’s trajectory is updated accordingly. Rayleigh scattering is enabled by default when a non-zero scattering length is provided, and Mie scattering can optionally be activated with specifying the asymmetry factor. This transport continues until the photon either exits the simulation world, is absorbed, is detected, or exceeds the user-defined limit on the number of reflections. 2.2.2 Charged Particle Handling and Secondary Photon Generation When operating in the charged-particle source mode, SPIROS simulates the passage of primary charged particles through the geometry. Their path is used to estimate the expected number of secondary optical photons produced by scintillation and Cherenkov radiation. For scintillation, the expectation is calculated using the energy loss per unit length (d E

Read the complete text