Collaborative Research: EAGER: SSMCDAT2023: Data-driven Predictive Understanding of Oxidation Resistance in High-Entropy Alloy Nanoparticles

非技术性总结这个奖项是在一个热切的提议上颁发的。它支持在利哈伊大学举行的SSMCDAT 2023数据马拉松上推进的一个项目的进展。该项目旨在了解高熵合金纳米颗粒的氧化行为，这是一种含有多种元素的大致相等比例的新型合金。工业条件下的氧化对这些合金的性能产生负面影响，限制了它们的广泛使用。在这个项目中，该团队将采用数据驱动的方法，将实验和计算数据集与有针对性的实验合成相结合。我们的目标是开发可靠的预测模型，考虑这两种类型数据中的不确定性，从而更好地理解高熵合金纳米颗粒在纳米尺度上的氧化行为。通过这一跨越材料科学、化学和应用数学的跨学科努力获得基础知识，该项目有可能提高高熵合金纳米颗粒的抗氧化性。它还将为验证数据驱动模型的关键实验研究提供必要的支持。这项研究为合成高熵材料的创新策略开辟了新的可能性，为未来研究和技术应用的令人兴奋的进步铺平了道路。该项目在协作和跨学科的研究环境中为研究生提供材料化学和数据科学的综合研究培训。该项目将参加数学和统计创新研究所的一项长期计划，并组织一个以“化学和材料科学的不确定度量化”为中心的研讨会。利用该项目的成果，该团队旨在推动和振兴数据密集型研究，特别是通过将不确定性量化整合到固态和材料化学领域的预测建模中。技术总结这个奖项是在一个热切的提议上颁发的。它支持在利哈伊大学举行的SSMCDAT 2023数据马拉松上推进的一个项目的进展。本项目旨在通过将实验和计算工具与现代数据科学方法相结合，从机理上了解高熵合金纳米粒子的元素偏析、迁移和抗氧化性之间的相互作用。目标是建立数据驱动的材料设计策略，允许通过成分设计精确控制高熵纳米颗粒的氧化动力学。该项目将利用现有的热力学和吸附能量学的实验和计算数据集，并将这些数据与补充的高通量第一性原理计算和混合分子动力学/蒙特卡罗模拟相结合，为氧化条件下高熵合金纳米粒子中的元素偏析和迁移模型开发机器学习模型。将开发一种新的高斯过程回归模型，该模型固有地包括不确定性量化并允许直观地解释，以预测氧化行为。此外，该项目将合成具有特定组成的高熵合金纳米颗粒，并表征其结构和氧化行为，将结果与模型预测进行比较。这项研究将为氧化环境下高熵合金纳米颗粒的结构-性质关系提供实验验证的基础知识。此外，该项目还将为固体和材料化学领域建立一套有价值的分析和建模工具。这些工具将实现加速高熵合金纳米粒子的实验-计算设计的综合方法，促进在更广泛的化学空间中以理论为指导的多组分材料纳米粒子的合成研究。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

NONTECHNICAL SUMMARYThis award is made on an EAGER proposal. It supports progress on a project advanced at the SSMCDAT 2023 Datathon held at Lehigh University. This project aims to understand the oxidation behavior of high-entropy alloy nanoparticles, which are a new type of alloys containing multiple elements in roughly equal proportions. Oxidation under industrial conditions negatively impacts the performance of these alloys, limiting their broader use. In this project, the team will employ a data-driven approach, combining experimental and computational datasets with targeted experimental synthesis. The goal is to develop reliable predictive models considering uncertainties in both types of data, to lead to a better understanding of the oxidation behavior of high-entropy alloy nanoparticles at the nanoscale. By gaining fundamental knowledge through this interdisciplinary effort spanning materials science, chemistry, and applied mathematics, the project has the potential to enhance the oxidation resistance of high-entropy alloy nanoparticles. It will also provide essential support for critical experimental studies to validate the data-driven models. This research opens new possibilities for innovative strategies to synthesize high-entropy materials, paving the way for exciting advances in future research and technological applications.The project provides comprehensive research training in materials chemistry and data science to graduate students within a collaborative and interdisciplinary research environment. The project will participate in a long program at the Institute of Mathematical and Statistical Innovation and organize a workshop centered around "Uncertainty Quantification for Chemistry and Materials Science". Leveraging the outcomes of the project, the team aims to propel and invigorate data-intensive research, particularly by integrating uncertainty quantification into predictive modeling within the domain of solid state and materials chemistry. TECHNICAL SUMMARYThis award is made on an EAGER proposal. It supports progress on a project advanced at the SSMCDAT 2023 Datathon held at Lehigh University. This project aims to gain a mechanistic understanding of the interplay among elemental segregation, migration, and oxidation resistance of high-entropy alloy nanoparticles by integrating experimental and computational tools with modern data science methods. The goal is to establish data-driven materials design strategies that allow precise control over the oxidation kinetics of high-entropy nanoparticles with composition design. The project will leverage existing experimental and computational datasets on thermodynamic and adsorption energetics, and combine this data with supplementary high-throughput first-principles calculations and hybrid molecular dynamics / Monte Carlo simulations, to develop machine learning models for elemental segregation and migration models in high-entropy alloy nanoparticles under oxidation conditions. A novel Gaussian Process regression model, which inherently includes uncertainty quantification and allows for intuitive interpretation, will be developed to predict oxidation behavior. Furthermore, the project will synthesize high-entropy alloy nanoparticles with specific compositions and characterize their structural and oxidation behavior, comparing the results with model predictions. This this research will provide experimentally validated fundamental knowledge regarding the structure-property relationships of high-entropy alloy nanoparticles under oxidation environments. Additionally, the project will establish a valuable suite of analytical and modeling tools for the field of solid-state and materials chemistry. These tools will enable an integrated approach to accelerate experimental-computational design of high-entropy alloy nanoparticles, facilitating theory-guided synthesis research of multicomponent material nanoparticles across a broader chemical space.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.