The influence of various laser powder bed fusion (LPBF) process parameters on the crystallographic textures and mechanical properties of a typical Ni-based solid-solution strengthened alloy, Hastelloy-X, was examined. Samples were classified into four groups based on the type of crystallographic texture: single crystalline-like microstructure with //build direction (BD) (-SCM), single crystalline-like microstructure with //BD (-SCM), crystallographic lamellar microstructure (CLM), or polycrystalline microstructure (PCM). These four crystallographic textures were realized in Hastelloy-X for the first time here to the best of our knowledge. The mechanical properties of the samples varied depending on their texture. The tensile properties were affected not only by the Schmid factor but also by the grain size and the presence of lamellar boundaries (grain boundaries). The lamellar boundaries at the interface between the //BD oriented main layers and the //BD-oriented sub-layers of CLM contributed to the resistance to slip transmission and the increased proof stress. It was possible to control a wide range of crystallographic microstructures via the LPBF process parameters, which determines the melt pool morphology and solidification behavior.

This study demonstrates the effects of recrystallization on tensile properties and the anisotropy of IN738LC, a typical γ’ precipitation-strengthened alloy, at both room and high temperatures via the laser powder bed fusion process. The nonrecrystallized columnar microstructure, subjected to standard IN738LC heat treatment up to 1120 °C, and the almost fully recrystallized microstructure, heat-treated at 1204 °C, were compared. The tensile properties strongly depend on whether recrystallization was completed as well as the tensile direction. This can be explained by microstructure characterization, featuring the Taylor factor in the tensile direction, average grain size estimated by ellipse approximation, and the relationship between the grain shape and tensile direction. The shape of the recrystallized grains and the distribution of coarse MC carbides inside the recrystallized grains were determined by the microstructure in an as-built state. In high-temperature tensile tests conducted in the horizontal direction, the separation of the columnar grains caused a brittle fracture. In contrast, dimples were observed at the fracture surface after recrystallization, indicating scope for further improvement in ductility.

Durability of thermal barrier coating (TBC) systems is important because of recent rising of turbine inlet temperature (TIT) for improved efficiency of industrial gas turbine engines. However, high-temperature environment accelerates the degradation of the TBC as well as causes spalling of the top coat. Spalling of the top coat may be attributed to several factors, but evidently the growth of thermally grown oxide (TGO) should be considered as an important factor. One method for reducing the growth rate of TGO is to provide a dense α-Al2O3 layer at the boundary of the bond coat and top coat. This α-Al2O3 layer will suppress the diffusion of oxygen to the bond coat and consumption of aluminum of the bond coat is suppressed. In this study, we focused on thermal pre-oxidation of the bond coat as a means for forming an α-Al2O3 barrier layer that would be effective at reducing the growth rate of TGO, and we studied the suitable pre-oxidation conditions. In the primary stage, we analyzed the oxidation behavior of the bond coat surface during pre-oxidation heat treatment by means of in situ synchrotron X-ray diffraction (XRD) analysis. As a result, we learned that during oxidation in ambient air environment, in the initial stage of oxidation metastable alumina is produced in addition to α-Al2O3, but if the thermal treatment is conducted under some specific low oxygen partial pressure condition, unlike in the ambient air environment, only α-Al2O3 is formed with suppressing formation of metastable alumina. We also conducted transmission electron microscope (TEM) and XRD analysis of oxide scale formed after pre-oxidation heat treatment of the bond coat. As a result, we learned that if pre-oxidation is performed under specific oxygen partial pressure conditions, a monolithic α-Al2O3 layer is formed on the bond coat. We performed a durability evaluation test of TBC with the monolithic α-Al2O3 layer formed by pre-oxidation of the bond coat. An isothermal oxidation test confirmed that the growth of TGO in the TBC that had undergone pre-oxidation was suppressed more thoroughly than that in the TBC that had not undergone pre-oxidation. Cyclic thermal shock test by hydrogen burner rig was also carried out. TBC with the monolithic α-Al2O3 layer has resistance to >2000 cycle thermal shock at a load equivalent to that of actual gas turbine.

Creep testing is time-consuming and costly, leading institutions to limit the number of tests conducted to the minimum necessary for their specific objectives. By pooling data from each institution, it is anticipated that predictive models can be developed for a wide range of materials, including welded joints and degraded materials exposed to service conditions. However, the data obtained by each institution is often highly confidential, making it challenging to share with others. Federated learning, a type of privacy-preserving computation technology, allows for learning while keeping data confidential. Utilizing this approach, it is possible to develop creep life prediction models by leveraging data from various institutions. In this paper, we constructed global deep neural network models for predicting the creep rupture life of heat-resistant ferritic steels in collaboration with eight institutions using the federated learning system we developed for this purpose. Each institution built a local model using only its own data for comparison. While these local models demonstrated good predictive accuracy for their respective datasets, their predictive performance declined when applied to data from other institutions. In contrast, the global model constructed using federated learning showed reasonably good predictive performance across all institutions. The distance between each institution's data was defined in the space of explanatory variables, with the NIMS data, which had the largest dataset, serving as the reference point. The global model maintained high predictive accuracy regardless of the distance from the NIMS data, whereas the predictive accuracy of the NIMS local model significantly decreased as the distance increased.

High temperature strength is one of the critical technical issues to apply metal injection molding (MIM) to Nickel-based superalloy IN713. In this study, through investigating microstructure and high temperature strength of IN713-MIM materials in detail, we developed the technique to promote grain growth. In the experiments using IN713C-MIM materials, the high temperature tensile strength and the creep strength were remarkably inferior to the castings. The creep deformation mechanism of IN713C-MIM materials is revealed that grain boundary diffusion creep was dominant as the stress exponent was determined as n = 2.72. Therefore, we found out that it was important to reduce the amount of grain boundary by promoting grain growth. In the examinations aiming to reduce carbides on grain boundary, which inhibited grain boundary migration, the carbon content of IN713LC-MIM-sintered material could be suppressed to C = 0.05%. Subsequently, when IN713LC-MIM-HIP-ed material was subjected to additional heat treatment at 1280°C for 12 hr, significant grain growth was observed. This Paper was Originally Published in Japanese in J. Jpn. Soc. Powder Powder Metallurgy 66 (2019) 17–22.

Creep testing is time-consuming and costly, leading institutions to limit the number of tests conducted to the minimum necessary for their specific objectives. By pooling data from each institution, it is anticipated that predictive models can be developed for a wide range of materials, including welded joints and degraded materials exposed to service conditions. However, the data obtained by each institution is often highly confidential, making it challenging to share with others. Federated learning, a type of privacy-preserving computation technology, allows for learning while keeping data confidential. Utilizing this approach, it is possible to develop creep life prediction models by leveraging data from various institutions. In this paper, we constructed global deep neural network models for predicting the creep rupture life of heat-resistant ferritic steels in collaboration with eight institutions using the federated learning system we developed for this purpose. Each institution built a local model using only its own data for comparison. While these local models demonstrated good predictive accuracy for their respective datasets, their predictive performance declined when applied to data from other institutions. In contrast, the global model constructed using federated learning showed reasonably good predictive performance across all institutions. The distance between each institution's data was defined in the space of explanatory variables, with the NIMS data, which had the largest dataset, serving as the reference point. The global model maintained high predictive accuracy regardless of the distance from the NIMS data, whereas the predictive accuracy of the NIMS local model significantly decreased as the distance increased.