A sealant is required for the solid oxide fuel cell (SOFC) to maintain hermeticity at high operating temperatures, keep fuel and oxidant from mixing, and avoid shorting of the cell stack. Glass and glass–ceramic materials are widely used as a sealant because their properties can be tailored to meet the stringent requirements of SOFC stack, but they are susceptible to cracking. In contrast, a promising concept of self‐repairable glass for seals is pursued for making reliable seals that can self‐repair cracks at the SOFC operating temperatures. This concept is studied through measuring crack‐healing kinetics and independent measurement of glass viscosity for relating to the observed self‐repair. The cracks on the glass surface are created using a Vickers indenter to achieve a well‐defined crack geometry, and then the glass is exposed to elevated temperatures for different length of times to study the crack‐healing kinetics. The crack‐healing kinetics is compared with the predictions of our theoretical model and found to be in good agreement. In addition, glass viscosity is extracted from the healing kinetics and compared with the independent measurement of viscosity measured from the dilatometry and sintering data to further validate the crack‐healing theoretical model. These results are presented and discussed.
- Award ID(s):
- 2001262
- PAR ID:
- 10322620
- Date Published:
- Journal Name:
- Mathematics and Mechanics of Solids
- Volume:
- 27
- Issue:
- 3
- ISSN:
- 1081-2865
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
Abstract -
Abstract Iron oxide copper-gold (IOCG) and iron oxide-apatite (IOA) deposits are major sources of Fe, Cu, and Au. Magnetite is the modally dominant and commodity mineral in IOA deposits, whereas magnetite and hematite are predominant in IOCG deposits, with copper sulfides being the primary commodity minerals. It is generally accepted that IOCG deposits formed by hydrothermal processes, but there is a lack of consensus for the source of the ore fluid(s). There are multiple competing hypotheses for the formation of IOA deposits, with models that range from purely magmatic to purely hydrothermal. In the Chilean iron belt, the spatial and temporal association of IOCG and IOA deposits has led to the hypothesis that IOA and IOCG deposits are genetically connected, where S-Cu-Au–poor magnetite-dominated IOA deposits represent the stratigraphically deeper levels of S-Cu-Au–rich magnetite- and hematite-dominated IOCG deposits. Here we report minor element and Fe and O stable isotope abundances for magnetite and H stable isotope abundances for actinolite from the Candelaria IOCG deposit and Quince IOA prospect in the Chilean iron belt. Backscattered electron imaging reveals textures of igneous and magmatic-hydrothermal affinities and the exsolution of Mn-rich ilmenite from magnetite in Quince and deep levels of Candelaria (>500 m below the bottom of the open pit). Trace element concentrations in magnetite systematically increase with depth in both deposits and decrease from core to rim within magnetite grains in shallow samples from Candelaria. These results are consistent with a cooling trend for magnetite growth from deep to shallow levels in both systems. Iron isotope compositions of magnetite range from δ56Fe values of 0.11 ± 0.07 to 0.16 ± 0.05‰ for Quince and between 0.16 ± 0.03 and 0.42 ± 0.04‰ for Candelaria. Oxygen isotope compositions of magnetite range from δ18O values of 2.65 ± 0.07 to 3.33 ± 0.07‰ for Quince and between 1.16 ± 0.07 and 7.80 ± 0.07‰ for Candelaria. For cogenetic actinolite, δD values range from –41.7 ± 2.10 to –39.0 ± 2.10‰ for Quince and from –93.9 ± 2.10 to –54.0 ± 2.10‰ for Candelaria, and δ18O values range between 5.89 ± 0.23 and 6.02 ± 0.23‰ for Quince and between 7.50 ± 0.23 and 7.69 ± 0.23‰ for Candelaria. The paired Fe and O isotope compositions of magnetite and the H isotope signature of actinolite fingerprint a magmatic source reservoir for ore fluids at Candelaria and Quince. Temperature estimates from O isotope thermometry and Fe# of actinolite (Fe# = [molar Fe]/([molar Fe] + [molar Mg])) are consistent with high-temperature mineralization (600°–860°C). The reintegrated composition of primary Ti-rich magnetite is consistent with igneous magnetite and supports magmatic conditions for the formation of magnetite in the Quince prospect and the deep portion of the Candelaria deposit. The trace element variations and zonation in magnetite from shallower levels of Candelaria are consistent with magnetite growth from a cooling magmatic-hydrothermal fluid. The combined chemical and textural data are consistent with a combined igneous and magmatic-hydrothermal origin for Quince and Candelaria, where the deeper portion of Candelaria corresponds to a transitional phase between the shallower IOCG deposit and a deeper IOA system analogous to the Quince IOA prospect, providing evidence for a continuum between both deposit types.
-
Abstract This paper considers estimation and prediction of a high-dimensional linear regression in the setting of transfer learning where, in addition to observations from the target model, auxiliary samples from different but possibly related regression models are available. When the set of informative auxiliary studies is known, an estimator and a predictor are proposed and their optimality is established. The optimal rates of convergence for prediction and estimation are faster than the corresponding rates without using the auxiliary samples. This implies that knowledge from the informative auxiliary samples can be transferred to improve the learning performance of the target problem. When the set of informative auxiliary samples is unknown, we propose a data-driven procedure for transfer learning, called Trans-Lasso, and show its robustness to non-informative auxiliary samples and its efficiency in knowledge transfer. The proposed procedures are demonstrated in numerical studies and are applied to a dataset concerning the associations among gene expressions. It is shown that Trans-Lasso leads to improved performance in gene expression prediction in a target tissue by incorporating data from multiple different tissues as auxiliary samples.
-
This work presents a novel technique for constructing spatially resolved ion densities from Transient Insertion Langmuir Probe (TIL Probe) measurements in a flame. Similar to a tomographic transformation, this technique is used to deduce the spatial distribution of ions in a flame from many individual measurements that are integrated along a probe's length. We demonstrate the approach in the oxyfuel cutting torch preheat flame, which presents two severe challenges for electrical measurements: (1) temperatures over 3,000K destroy most probes made from alloys with appropriate chemical stability, and (2) the relevant length scales are on the order 0.15 mm. Presented here are (1) a Fourier series formulation for the current density, (2) a least-square problem for calculating the coefficients, (3) criteria for the highest wavenumber allowed in the expansion, (4) description of an experiment used to measure probe currents in an oxyfuel cutting torch preheat flame, (5) solution for spatially resolved current density in the oxyfuel cutting torch flame. Images of ion current density are produced with a resolution of 0.15 mm (0.0059 in), exhibiting peak current densities around 14 $\mu$A/mm. It is found that low-signal regions in the ``shadow'' of high-signal regions can suffer from signal-to-noise ratio problems due to natural fluctuations in the flame, and improvements are proposed to mitigate the effect. It is found that the numerical cost of setting up the resulting Hermitian-matrix linear problem far exceeds the numerical cost of inversion. High-level packages like Python and MATLAB are far too slow, so a multi-threaded algorithm is implemented in C, and the LAPACKE C library is used for efficient linear algebra support.more » « less
-
This work presents a novel technique for constructing spatially resolved ion densities from Transient Insertion Langmuir Probe (TIL Probe) measurements in a flame. Similar to a tomographic transformation, this technique is used to deduce the spatial distribution of ions in a flame from many individual measurements that are integrated along a probe's length. We demonstrate the approach in the oxyfuel cutting torch preheat flame, which presents two severe challenges for electrical measurements: (1) temperatures over 3,000K destroy most probes made from alloys with appropriate chemical stability, and (2) the relevant length scales are on the order 0.15 mm. Presented here are (1) a Fourier series formulation for the current density, (2) a least-square problem for calculating the coefficients, (3) criteria for the highest wavenumber allowed in the expansion, (4) description of an experiment used to measure probe currents in an oxyfuel cutting torch preheat flame, (5) solution for spatially resolved current density in the oxyfuel cutting torch flame. Images of ion current density are produced with a resolution of 0.15 mm (0.0059 in), exhibiting peak current densities around 14 $\mu$A/mm. It is found that low-signal regions in the ``shadow'' of high-signal regions can suffer from signal-to-noise ratio problems due to natural fluctuations in the flame, and improvements are proposed to mitigate the effect. It is found that the numerical cost of setting up the resulting Hermitian-matrix linear problem far exceeds the numerical cost of inversion. High-level packages like Python and MATLAB are far too slow, so a multi-threaded algorithm is implemented in C, and the LAPACKE C library is used for efficient linear algebra support.more » « less