Collaborative Research: Beyond the Single-Atom Paradigm: A Priori Design of Dual-Atom Alloy Active Sites for Efficient and Selective Chemical Conversions

在化学系化学结构、动力学和机理A(CSDM-A)计划的支持下，杜兰大学的Matthew Montemore和塔夫茨大学的Charles Sykes正在利用量子化学、机器学习和原子尺度表征对一类新型双原子催化剂进行计算和实验研究。这些新材料被称为双原子合金，由嵌入在主体金属表面的一对能量稳定的不同金属原子组成。与单原子合金相比，双原子合金的活性中心预计会更活跃，能够进行更具挑战性的反应，包括靶向断裂。单原子合金是一类活性中心定义明确的类似材料。然而，加速发现新合金需要解决许多挑战，包括从潜在配对的巨大材料空间中识别有希望的金属组合，了解它们的结构/功能关系，以及在实验和计算之间建立有效的反馈回路。Montemore博士和他的学生将使用机器学习和密度泛函理论(DFT)相结合的方法，根据特定化学反应的表面稳定性和低能垒来确定候选结构。赛克斯博士和他的学生将对预测的结构进行实验合成、表征和测试。他们的发现可能导致一种全新类型的催化剂的开发，在这种催化剂中，定义良好的金属原子结构可以与预测模型相结合，以加速一系列多步骤反应机理。除了这些广泛的科学影响，这个项目还将支持一个虚拟现实K-12外联研讨会，为西班牙裔科学家和工程师提供指导，并通过促进反向科学博览会激励服务不足的高中的学生。双原子合金利用单原子合金的几个优势：定义明确的活性中心，允许在计算、表面科学和反应堆研究之间清晰地通信；在反应能量学中偏离线性关联的机会；以及独特的电子结构。就像单原子合金一样，表面科学实验和计算先于催化剂合成和测试，Montemore和Sykes将把理论和表面科学实验结合起来，探索双原子合金的结构-反应空间。这项研究有望导致在一系列定义明确的活性部位上发展结构-功能关系，这将成为更大的多相催化社区的指南。此外，这些研究有可能发现具有高催化性能的新材料类别，特别需要对传统多相催化剂具有挑战性的反应，如选择性交叉偶联。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。

With support from the Chemical Structure, Dynamics, and Mechanisms A (CSDM-A) program in the Division of Chemistry, Matthew Montemore of Tulane University and Charles Sykes of Tufts University are performing computational and experimental investigations of a novel class of dual-atom catalysts using quantum chemistry, machine learning, and atomic-scale characterization. These novel materials, termed dual-atom alloys, consist of an energetically stable pair of different metal atoms embedded in the surface of a host metal. Compared to single-atom alloys, a similar class of materials with well-defined active sites, dual-atom alloy active sites are anticipated to be more reactive and enable more challenging reactions including targeted bond scission. However, accelerating the discovery of novel alloys requires addressing numerous challenges including identifying promising combinations of metals from the large materials space of potential pairings, understanding their structure/function relationships, and creating an effective feedback loop between experiment and computation. Dr. Montemore and his students will use machine learning combined with density functional theory (DFT) to identify candidate structures based on surface stability and low energy barriers for specific chemical reactions. Dr. Sykes and his students will experimentally synthesize, characterize, and test the predicted structures. Their discoveries could lead to the development of an entirely new class of catalysts where well-defined metal atomic structures can be combined with predictive models to accelerate a range of multistep reaction mechanisms. In addition to these broad scientific impacts, this project will also support a virtual reality K-12 outreach workshop, provide mentorship for Hispanic scientists and engineers, and inspire students from underserved high schools by facilitating a Reverse Science Fair.Dual-atom alloys leverage several of the advantages of single-atom alloys: a well-defined active site that allows clear correspondence across computation, surface science, and reactor studies; opportunities to deviate from linear correlations in reaction energetics; and unique electronic structure. Just as with single-atom alloys, in which surface science experiments and computation preceded catalyst synthesis and testing, Montemore and Sykes will couple theory and surface science experiments to explore dual-atom alloy structure-reactivity space. This research is expected to lead to the development of structure-function relationships on a broad set of well-defined active sites, which would then serve as a guide to the larger heterogenous catalysis community. Furthermore, these studies have the potential to uncover new classes of materials with high catalytic performance, a particular need for reactions that are challenging with traditional heterogeneous catalysts, such as selective cross-couplings.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.