Collaborative Research: QRM: Microstructure Manifold Analysis Using Hierarchical Set of Morphological, Topological, and Process Descriptors

点击翻译按钮获取中文摘要

The properties and performance of structural materials are dictated by their structure on the microscopic level, referred to as the microstructure. Advanced manufacturing processes seek to tune the microstructure to obtain materials with desired properties. One key requirement to enable such manufacturing processes is the ability to quantitatively describe the microstructures. This award supports the development of mathematical tools to quantify the material microstructure in terms of shape (i.e., morphology), geometry and connectedness (i.e., topology), and property (i.e., performance) within a unified and consistent framework. Such a quantitative approach to characterizing microstructures is used to understand how different manufacturing pathways affect the microstructure and its properties. Researched techniques for efficient and adaptive exploration ensure that this processing-property landscape is explored with a reduced number of experiments. Although this project is geared towards laying the foundations of microstructure quantification, the research has the potential to advance knowledge in several application areas, such as organic electronics, porous electrodes for batteries and fuel cells, and membranes. This research has the potential to accelerate the design of new devices with superior properties by reducing the cost and time-to-market for engineered mesostructure-sensitive materials. Therefore, this work will contribute to enhancing the global competitiveness of the national manufacturing sector. The education and workforce development aspects of the project involve training of the next generation of globally competitive engineers and scientists.The overarching goal of this project is to lay the foundations for using a comprehensive suite of descriptors with incremental dimensionality reduction techniques to adaptively and efficiently build reliable processing-structure-property relationships. A comprehensive suite of topological, morphological and geometric descriptors will be used as markers to adaptively learn the microstructure manifold using modern incremental manifold learning strategies. Process-structure-property relationships can then be naturally parametrized and explored by using this microstructure manifold. This award will provide educational modules about computational and data-driven materials science. Diverse existing mechanisms at the partner institutions will be leveraged to advance goals of minority and women recruitment, undergraduate research, and K-12 outreach.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.