Quantifying intermediate‐frequency heterogeneities of <scp>SOFC</scp> electrodes using X‐ray computed tomography

Hydrogen embrittlement (HE) causes sudden, costly failures of metal components across a wide range of industries. Yet, despite over a century of research, the physical mechanisms of HE are too poorly understood to predict HE-induced failures with confidence. We use non-destructive, synchrotron-based techniques to investigate the relationship between the crystallographic character of grain boundaries and their susceptibility to hydrogen-assisted fracture in a nickel superalloy. Our data lead us to identify a class of grain boundaries with striking resistance to hydrogen-assisted crack propagation: boundaries with low-index planes (BLIPs). BLIPs are boundaries where at least one of the neighboring grains has a low Miller index facet—{001}, {011}, or {111}—along the grain boundary plane. These boundaries deflect propagating cracks, toughening the material and improving its HE resistance. Our finding paves the way to improved predictions of HE based on the density and distribution of BLIPs in metal microstructures.

show abstract

“…Samples were characterized using near-field HEDM (nf-HEDM) 80 , 81 and XRAT 76 , 82 at APS beamline 1-ID. Both techniques are non-destructive.…”

Section: Methodsmentioning

confidence: 99%

Crystallographic character of grain boundaries resistant to hydrogen-assisted fracture in Ni-base alloy 725

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…As described in our earlier works, [40][41][42] custom inhouse Python codes were used to calculate relevant microstructural properties for each subvolume, including: volume fractions; the average and standard deviation of diameter from the volumeweighted size distribution for each phase (based on an inscribed sphere method); interfacial surface area between each pair of phases; geometric tortuosity factors for each phase; and total and active (connected) TPB densities. For each phase in each subvolume, we also calculated the formation factor K i , which represents the ratio between the effective and bulk diffusivity or conductivity in phase i.…”

Section: Three High-density Regions In the Density Map Were Then Manually Bounded To Indicate Whatmentioning

confidence: 99%

Microstructure Generation via Generative Adversarial Network for Heterogeneous, Topologically Complex 3D Materials

Hsu

Epting

Kim

et al. 2020

JOM

Self Cite

View full text Add to dashboard Cite

Using a large-scale, experimentally captured 3D microstructure data set, we implement the generative adversarial network (GAN) framework to learn and generate 3D microstructures of solid oxide fuel cell electrodes. The generated microstructures are visually, statistically, and topologically realistic, with distributions of microstructural parameters, including volume fraction, particle size, surface area, tortuosity, and triple-phase boundary density, being highly similar to those of the original microstructure. These results are compared and contrasted with those from an established, grain-based generation algorithm (DREAM.3D). Importantly, simulations of electrochemical performance, using a locally resolved finite element model, demonstrate that the GAN-generated microstructures closely match the performance distribution of the original, while DREAM.3D leads to significant differences. The ability of the generative machine learning model to recreate microstructures with high fidelity suggests that the essence of complex microstructures may be captured and represented in a compact and manipulatable form.

show abstract

“…As described in our earlier works [40][41][42], custom in-house Python codes were used to calculate relevant microstructural properties for each subvolume, including: volume fractions; the average and standard deviation of diameter from the volume-weighted size distribution for each phase (based on an inscribed sphere method); interfacial surface area between each pair of phases; geometric tortuosity factors for each phase; and total and active (connected) TPB densities. For each phase in each subvolume, we also calculated the formation factor K i , which represents the ratio between the effective and bulk diffusivity or conductivity in phase i.…”

Section: A 2d Density Map Was Generated Plotting the Grayscale Intens...mentioning

confidence: 99%

Microstructure Generation via Generative Adversarial Network for Heterogeneous, Topologically Complex 3D Materials

Hsu¹,

Epting²,

Kim³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Using a large-scale, experimentally captured 3D microstructure dataset, we implement the generative adversarial network (GAN) framework to learn and generate 3D microstructures of solid oxide fuel cell electrodes. The generated microstructures are visually, statistically, and topologically realistic, with distributions of microstructural parameters, including volume fraction, particle size, surface area, tortuosity, and triple phase boundary density, being highly similar to those of the original microstructure. These results are compared and contrasted with those from an established, grain-based generation algorithm (DREAM.3D). Importantly, simulations of electrochemical performance, using a locally resolved finite element model, demonstrate that the GAN generated microstructures closely match the performance distribution of the original, while DREAM.3D leads to significant differences. The ability of the generative machine learning model to recreate microstructures with high fidelity suggests that the essence of complex microstructures may be captured and represented in a compact and manipulatable form.

show abstract

Quantifying intermediate‐frequency heterogeneities of SOFC electrodes using X‐ray computed tomography

Cited by 27 publications

References 52 publications

Crystallographic character of grain boundaries resistant to hydrogen-assisted fracture in Ni-base alloy 725

Crystallographic character of grain boundaries resistant to hydrogen-assisted fracture in Ni-base alloy 725

Microstructure Generation via Generative Adversarial Network for Heterogeneous, Topologically Complex 3D Materials

Microstructure Generation via Generative Adversarial Network for Heterogeneous, Topologically Complex 3D Materials

Contact Info

Product

Resources

About