Bryan Conry

Tristructural-isotropic (TRISO) fuel and other coated particle nuclear fuels have been selected as candidate fuels for a wide range of microreactor concepts for terrestrial use and space exploration. The fabrication of these fuels largely relies on spout fluidized bed chemical vapor deposition, a process in which a batch of fuel kernels is subjected to flows of fluidizing and pyrolyzing gases, which are precursors to the desired coating composition. Kernel sphericity, particle size distribution (PSD), and surface roughness are critical parameters that can affect fluidization, coating quality, and ultimately fuel performance. Understanding the impact of PSD of the initial fuel kernel could allow the use of kernels with a wider fuel specification. If possible, reducing requirements on sphericity and particle size uniformity without compromising coating performance could reduce process cost, time, and waste generation. Therefore, a better understanding of their impact on the fluidization and coating characteristics is needed. This work used computational fluid dynamics coupled with discrete element modeling to assess the impact of polydispersity of the initial kernel population on fluidization for two mean particle sizes: 250 and 400 um. The initial particle size is normally distributed with standard deviations varying from 12 um (matching experimental conditions) to 63 um. The results show that preferential fluidization effects are increasingly present with wider PSDs and are more prevalent for the smaller mean particle size of 250 um. Increasing the PSD leads to an increase in coating thickness variations for the smallest particles and smaller coating thickness variations for the largest particles. The results have important implications to fabrication efforts, where consistency of coating thickness is of importance.

Tristructural-isotropic (TRISO) fuel particles have demonstrated robust performance under irradiation; however, rare failures have been linked to crack initiation within the porous carbon buffer layer. A mechanistic understanding of these failure pathways requires quantitative characterization of buffer porosity across length scales relevant to stress localization. Previous studies using focused ion beam scanning electron microscopy (FIB-SEM) tomography have provided high-resolution insight into nanoscale porosity but are inherently limited in volume and may not capture the largest and most structurally significant pore features. In this work, X-ray computed tomography (XCT) was used to nondestructively characterize the 3D pore structure of the TRISO buffer layer over volumes several orders of magnitude larger than those accessible by FIB-SEM. A total of more than 1,000,000 pores were segmented from five particles of a common batch and analyzed to quantify pore size, morphology, orientation, and spatial distribution as a function of radial position within the buffer. The results reveal a highly heterogeneous pore population, with gradients in pore size and morphology along the radial thickness of the buffer. In particular, large, anisotropic, and low-solidity pores are preferentially concentrated near the buffer–inner pyrolytic carbon (IPyC) interface and exhibit circumferential alignment. These features are expected to promote localized stress concentrations under irradiation-induced kernel swelling and buffer densification, providing a microstructural basis for previously observed circumferential cracking and buffer–IPyC debonding. Comparison with prior FIB-SEM characterization indicates that volume-limited techniques underrepresent the largest pore structures that are likely most relevant to fracture initiation, propagation, and transmission into subsequent layers. This work demonstrates the critical role of large-volume characterization in capturing the dominant features of buffer porosity and provides statistically robust microstructural data for incorporation into advanced fuel performance models aimed at predicting TRISO failure behavior.