Methodologies for imaging a used nuclear fuel dry storage cask with cosmic ray muon computed tomography

TL;DR

This paper explores methodologies for imaging used nuclear fuel dry storage casks using cosmic ray muon computed tomography, addressing challenges like low muon flux and path modeling to improve non-invasive inspection techniques.

Contribution

It investigates three muon path models, two projection methods, and two reconstruction techniques within a Geant4 simulation to enhance muon CT for nuclear fuel cask monitoring.

Findings

Different models and methods were evaluated for accuracy and efficiency.

Simulation results demonstrate potential improvements in imaging quality.

Insights into practical implementation constraints for muon CT in nuclear safety.

Abstract

Muons interact with matter via two major interaction mechanisms: ionization and radioactive process, and multiple Coulomb scattering leading to energy loss and trajectory deflection, respectively. For a monoenergetic muon beam crossing an object, the scattering angle follows a Gaussian distribution with a zero mean value and a variance that depends on the atomic number of the material object it traversed. Thus, the measured scattering angle may be used to reconstruct the geometrical and material information of the contents inside the dry storage cask. In traditional X-ray computed tomography, the projection information used to reconstruct the attenuation map of the imaged objects is the negative natural logarithm of the transmission rate of the X-rays, which is equal to the linear summation of the X-ray attenuation coefficients along the incident path. Similarly, the variance of the muon scattering angle is also the linear integral of the scattering density of the objects crossed by the muons. Thus, a muon CT image can be built by equating scattering density with attenuation coefficient. However, muon CT faces some unique challenges including: 1) long measurement times due to low cosmic muon flux, 2) insufficiently accurate muon path models, and 3) the inability to precisely measuring muon momentum. In this work, three different muon path models, two different projection methods, and two different reconstruction methods were investigated for use in muon CT of dry storage casks. The investigation was conducted in a validated Geant4 workspace, both in an ideal case and with relevant engineering restrictions considered. The results of these investigations and the expected benefits for fuel cask monitoring are reported herein.