Search for: All records

Refractory high entropy alloys (RHEAs) have gained significant attention in recent years as potential replacements for Ni-based superalloys in gas turbine applications. Improving their properties, such as their high-temperature yield strength, is crucial to their success. Unfortunately, exploring this vast chemical space using exclusively experimental approaches is impractical due to the considerable cost of the synthesis, processing, and testing of candidate alloys, particularly at operation-relevant temperatures. On the other hand, the lack of reasonably accurate predictive property models, especially for high-temperature properties, makes traditional Integrated Computational Materials Engineering (ICME) methods inadequate. In this paper, we address this challenge by combining machine-learning models, easy-to-implement physics-based models, and inexpensive proxy experimental data to develop robust and fast-acting models using the concept of Bayesian updating. The framework combines data from one of the most comprehensive databases on RHEAs (Borg et al., 2020) with one of the most widely used physics-based strength models for BCC-based RHEAs (Maresca and Curtin, 2020) into a compact predictive model that is significantly more accurate than the state-of-the-art. This model is cross-validated, tested for physics-informed extrapolation, and rigorously benchmarked against standard Gaussian process regressors (GPRs) in a toy Bayesian optimization problem. Such a model can be used as a tool within ICME frameworks to screen for RHEAs with superior high-temperature properties. The code associated with this work is available at: https://codeocean.com/capsule/7849853/tree/v2. 

Abstract Laser powder bed fusion (L-PBF) additive manufacturing (AM) is an effective method of fabricating nickel–titanium (NiTi) shape memory alloys (SMAs) with complex geometries, unique functional properties, and tailored material compositions. However, with the increase of Ni content in NiTi powder feedstock, the ability to produce high-quality parts is notably reduced due to the emergence of macroscopic defects such as warpage, elevated edge/corner, delamination, and excessive surface roughness. This study explores the printability of a nickel-rich NiTi powder, where printability refers to the ability to fabricate macro-defect-free parts. Specifically, single track experiments were first conducted to select key processing parameter settings for cubic specimen fabrication. Machine learning classification techniques were implemented to predict the printable space. The reliability of the predicted printable space was verified by further cubic specimens fabrication, and the relationship between processing parameters and potential macro-defect modes was investigated. Results indicated that laser power was critical to the printability of high Ni content NiTi powder. In the low laser power setting (P < 100 W), the printable space was relatively wider with delamination as the main macro-defect mode. In the sub-high laser power condition (100 W ≤ P ≤ 200 W), the printable space was narrowed to a low hatch spacing region with macro-defects of warpage, elevated edge/corner, and delamination happened at different scanning speeds and hatch spacing combinations. The rough surface defect emerged when further increasing the laser power (P > 200 W), leading to a further narrowed printable space.