Method for measurement of TRISO kernel and layer volumes by X-ray computed tomography

https://doi.org/10.1016/j.jnucmat.2020.152255Get rights and content

Abstract

Layer dimensions are key parameters for as-fabricated tristructural-isotropic (TRISO) particle fuel as well as for post-irradiation examination of particle performance. Layer thicknesses are typically measured by optical microscopy of the particle cross section near mid-plane, while layer volumes are estimated with serial sectioning and microscopy. This method for measuring layer volumes is limited due to the resolution limit imposed by slice thickness and the effect of the polishing process on delicate irradiated particle microstructures. Image processing software has been developed to segregate the three-dimensional (3D) TRISO particle images provided by x-ray computed tomography (XCT) into high-resolution volumetric data for the kernel, all particle layers, and any internal voids introduced during irradiation. These data can be used to analyze key in-reactor behaviors of TRISO particles such as kernel swelling or buffer shrinkage, which are important inputs for fuel performance modeling.

Introduction

Tristructural-isotropic (TRISO) particle fuels are the focus of significant experimental work due to widespread interest in TRISO-based reactor designs. The dimensions of the kernel and the coating layers are key parameters measured in TRISO characterization. Mean layer thicknesses are commonly measured in as-fabricated particles. Optical microscopy is used to measure particles polished to their mid-plane, while mean kernel diameters are commonly measured using optical shadow imaging [1]. More detailed data regarding kernel volumes and coating layers are commonly acquired using serial sectioning coupled with optical microscopy [2]. While this method is sufficient for basic characterization, it has two main shortcomings. First, the resolution of serial sectioning is limited by the step-size between imaging steps. This limitation may be mitigated by reducing the step-size, but there is a practical limit to the number of steps, particularly when working with irradiated particles in hot cell environments. Second, the process of serial sectioning may damage fragile irradiation microstructures, particularly in the buffer region, which commonly features internal fissures and tenuous connections to the adjoining inner pyrolytic carbon (IPyC) layer.

The alternative method for nondestructive characterization of TRISO particles in common use is x-ray computed tomography (XCT). In XCT, two-dimensional (2D) projections of a sample—or radiographs—are generated at varying sample rotations relative to the x-ray source and detector. Brightness and contrast within these radiographs are determined by the relative x-ray transmission through the sample. The radiographs are reconstructed into a single 3D image of the sample based on their specific angles. Historically, XCT of TRISO particles has been used mostly for qualitative analysis of as-fabricated and irradiated particles, with a focus on identification of defects and anomalies for further materialographic analysis [3,4].

An image analysis method has been developed to process tomographic datasets for TRISO particles and generate quantitative information regarding their geometry and structure. Compared to serial sectioning, this method provides higher resolution results while preserving the as-irradiated microstructures. These advantages are particularly useful in quantifying irradiation effects such as kernel swelling or buffer shrinkage. Combined with individual particle irradiation capabilities such as those offered by the Oak Ridge National Laboratory (ORNL) MiniFuel irradiation capsule [5], this image analysis method enables high-resolution measurement of kernel and layer dimensions on the same particle pre- and post-irradiation to quantify changes in a specific time, temperature, and flux environment.

Section snippets

Loading of particles

The key requirements for loading TRISO particles for XCT imaging are (1) the particle must remain stationary during imaging, and (2) sufficient radiation shielding must be provided for handling irradiated particles. When imaging unirradiated particles, two loading methods were used. In the first method, a single particle was placed in a Lucite holder with a notched conical cavity, as shown in Fig. 1. In the second loading method, a Kapton tube of an appropriate diameter was loaded with a series

Layer segmentation

The kernel and TRISO layers were segmented using software written in MATLAB. The segmentation process was designed to take advantage of the nested sphere geometry of TRISO particles to produce the best possible results.

Calculation of geometric parameters

Once segmentation of the kernel and TRISO layers was complete, geometric parameters characterizing each were calculated. These parameters were calculated at sets of approximately equidistant spherical angles from the centroid of the kernel. Kernel radius at each of 5000 spherical angles was taken to equal the position of the kernel/buffer boundary at that angle. Layer thicknesses were calculated at each of 5000 spherical angles by taking the difference between their outer and inner boundaries

Example results

Results illustrating the capabilities of this method in several example particles are given in the following subsections. While these examples show the type of data which can be gathered by this analysis, a full-scale implementation over extensive batches of particles has not yet been performed.

Conclusions

Characterization of individual TRISO particles by XCT has been expanded from primarily qualitative analysis of particle features to detailed quantitative analysis of particle geometry. Digital segmentation of the kernel, the coating layers, and additional particle features in tomographic datasets enables measurement of thicknesses, volumes, and curvatures to fully characterize each feature. This method may be applied to both as-fabricated and irradiated particles, and can support experiments

Data availability

The raw/processed data required to reproduce these findings cannot be shared at this time due to technical or time limitations.

CRediT authorship contribution statement

Grant W. Helmreich: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Software, Supervision, Validation, Visualization, Writing - original draft. Dylan Richardson: Data curation, Investigation, Methodology, Writing - review & editing. Singanallur Venkatakrishnan: Investigation, Methodology, Writing - review & editing. Amir Ziabari: Investigation, Methodology, Writing - review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was sponsored by the U.S. Department of Energy, Office of Nuclear Energy, through the Advanced Reactor Concepts ARC-Xe program and through the Idaho National Laboratory Advanced Reactor Technologies Technology Development Office as part of the Advanced Gas Reactor Fuel Development and Qualification Program.

Cited by (0)

This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

View full text