We report a curated dataset that brings together composition, processing conditions, microstructural details, and mechanical properties for 396 combinations of alloy composition and processing condition drawn from 100 peer-reviewed research articles on precipitate-containing multi-principal element alloys (MPEAs). The dataset was created by first utilizing a generative large language model for information extraction, followed by expert review to ensure accurate recovery of materials data. Compositional information was taken directly from tables and text, while processing routes - including homogenization, rolling, recrystallization, and aging - were converted into uniform temperature and time metrics. Microstructural descriptors, including precipitate phases and sizes, were consolidated into a consistent labeling scheme to accommodate the wide range of terminology used in published literature. Finally, mechanical property data, such as strength and ductility, were compiled together with the temperatures at which they were measured. These data provide a coherent view of the composition-processing-microstructure-property features explored in existing MPEA research and establish a resource that supports data-driven alloy design as well as future development of automated materials information-extraction methodologies. The complete dataset is available on Zenodo.
Raj et al. (Fri,) studied this question.