Collaborative Research: High-Throughput Exploration of Microstructure-Sensitive Design for Steel Microstructure Optimization to Enhance its Corrosion Resistance in Concrete

碳钢在混凝土中的腐蚀是型钢混凝土结构最常见、最昂贵的劣化机制。美国的腐蚀成本约相当于国内生产总值(GDP)的3%至4%。据估计，每年仅公路桥梁的腐蚀就会给美国经济造成230亿至310亿美元的损失。此外，腐蚀缩短了民用基础设施的使用寿命，导致材料使用量增加。这反过来会增加建筑业的碳足迹，并影响气候变化缓解战略。因此，开发和利用创新、廉价和有效的耐蚀钢材至关重要，以将这种对美国经济和环境的负担降至最低。碳钢是混凝土中使用最多的加固材料，因其可获得性好、成本低而受到广泛关注。支持这一合作研究项目的中心假设是，碳钢的微观结构可以优化，以增强其在混凝土环境中的耐腐蚀性。由于相应的微观结构设计空间很大，因此研究微观组织与腐蚀性能之间的定量关系是具有挑战性的。传统的设计方法远远不能系统地探索如此大的设计空间并确定最佳解决方案。对微结构敏感的设计和材料知识系统采用全面和定量的微结构处理，与新兴的机器学习工具一起可以应对上述重大挑战。该项目的另一个同样重要且新颖的部分在于利用高通量策略来收集和整理高价值的实验数据。为了满足这一需求，将设计和实施新的高通量策略，包括合成跨越各种不同微结构的材料样本库以及评估其微结构和腐蚀性能。这项研究旨在通过使研究人员和材料开发人员拥有基本工具来假设、设计、优化和测试新材料，以经济高效的方式缓解与钢筋混凝土结构中的钢筋腐蚀相关的问题，从而产生深远的社会、政治和经济影响。该方法的科学新颖性在于它能够预测碳钢的微观组织对其腐蚀性能的影响。这些见解可以用来调整组织，在不改变钢的化学成分的情况下优化钢的耐腐蚀性。这项研究的主要推动力来自于需要(1)阐明在碱性混凝土环境中碳钢的腐蚀和微观结构之间鲜为人知的联系，以及(2)弥合与优化钢的微观结构敏感的耐腐蚀性相关的关键知识差距。这项工作集中在四个方面：(1)高通量样品合成，(2)高通量腐蚀性能表征，(3)组织特征工程和构建机器学习模型，以及(4)设计和制造具有最佳微观结构的钢。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。

Corrosion of carbon steel in concrete is the most common and costly deterioration mechanism of steel-reinforced concrete structures. Corrosion costs in US are equivalent to about 3 to 4 percent of the gross domestic product (GDP). The annual cost of corrosion of just highway bridges to the US economy is estimated to be US$23-31 billion. Furthermore, corrosion reduces the lifetime of civil infrastructure and leads to increased use of material. This, in turn, increases the carbon footprint of the construction industry and affects climate change mitigation strategies. Thus, it is critical to develop and utilize innovative, inexpensive, and effective corrosion-resistant steel to minimize this burden on the US economy and on the environment. Carbon steel is the most used reinforcing material in concrete due to its availability and low cost. The central hypothesis underpinning this collaborative research project is that the carbon steel microstructure can be optimized to enhance its corrosion resistance in a concrete environment. Studying the quantitative correlations between microstructure and corrosion properties is challenging since the corresponding microstructure design space is very large. Traditional design approaches are woefully inadequate for systematically exploring such large design spaces and identifying optimal solutions. Microstructure-sensitive design and materials knowledge systems employ a comprehensive and quantitative microstructure treatment, which together with emergent machine learning tools can address the grand challenge described above. An equally important and novel component of this project lies in exploiting high-throughput strategies to collect and curate high-value experimental data. In order to address this need, novel high-throughput strategies, both in synthesizing material sample libraries spanning a wide range of distinct microstructures and evaluating their microstructures and corrosion performances, will be designed and implemented. This research aims to have far-reaching social, political, and economic impacts by enabling researchers and material developers with the fundamental tools to hypothesize, design, optimize, and test new materials to mitigate issues associated with steel corrosion in reinforced concrete structures in a cost-effective way. The scientific novelty of the approach lies in its ability to predict the influence of the microstructure of carbon steel on its corrosion performance. These insights can be used to tune the microstructure to optimize the corrosion resistance of the steel without changing the steel chemistry. The main impetus for this research comes from the need to (1) elucidate the poorly understood linkages between corrosion and the microstructure of carbon steel in an alkaline concrete environment, and (2) bridge a critical knowledge gap related to optimizing the microstructure-sensitive corrosion resistance of steels. This work is focused on four thrusts: (1) high-throughput synthesis of samples, (2) high-throughput characterization of corrosion performance, (3) microstructure feature engineering and building machine learning models, and (4) designing and fabricating steel with an optimal microstructure.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.