Collaborative Research: Catalyst Structure, Reaction Mechanism, and Roles of Chlorine for Ethylene Epoxidation

环氧乙烷(EO)是生产各种先进材料、工业溶液和表面活性剂的关键中间体。全球环氧乙烷产量超过270亿公斤，主要来源是原料乙烯与银(银)催化剂上的氧气的选择性反应。银催化剂是由五种或五种以上元素的复杂组合促进的，每种元素在催化中扮演不同的角色。氯是提高乙烯转化为环氧乙烷选择性的主要助剂之一。尽管目前工业工艺技术的环氧乙烷选择性在90%以上，但在提高工艺效率和减少二氧化碳排放的同时，仍存在进一步提高选择性的机会。然而，氯提高选择性的相互作用尚不完全清楚，从而限制了选择性的进一步提高。该项目结合了先进的实验、计算和机器学习方法，以更好地了解环氧乙烷制造的催化机理和氯助剂所起的作用。该项目将确定在与环氧乙烷催化相关的压力和温度下，氯促进的银纳米颗粒上存在的稳态活性相、表面结构和表面中间体。更深入地了解环氧乙烷催化剂的操作和设计原则，以及环氧乙烷选择性的提高，可能会在减少碳排放的同时，推动美国化学工业取得革命性的进步。此外，该项目将加强合作机构的研究生和本科生教育，方法是在理论研究小组和实验研究小组之间对学生进行交叉培训，通过导师计划进行K-12推广，以及与STEM针对高中女性的教育机会相结合。该项目探索了许多先前研究中的活性银催化剂的表面结构不代表在实际条件下运行的催化剂的可能性。因此，氯的主要反应机理和提高选择性的途径仍然不清楚，特别是在工业相关的反应条件和氯覆盖范围内。长期以来，对氯作为促进剂的要求要么与表面氧原子的诱导反应活性有关，要么与氧化物表面存在的导致燃烧的氧空位的选择性滴定有关，但这些解释假定特定的机理途径取决于催化剂的表面结构。该项目将建立环氧化和燃烧的机理和反应途径，以及这些途径涉及氧诱导的表面重建中特定的氧中间体和氧空位的途径。此外，还将深入了解被吸附的氯原子如何影响表面重建、活性氧物种和空位的分布和覆盖，以及反应势垒、动力学和选择性的相关变化。通过实验(时间分辨原位拉曼光谱)和计算方法(基于机器学习的势的宏正则蒙特卡罗和分子动力学模拟)相结合的方法来研究反应条件下催化体系的内在复杂性。更广泛地说，这个项目将提供创造和验证方法(包括实验和计算方法)的机会，以解决催化过程中表面结构的变化，这是该领域的一个主要挑战。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。

Ethylene oxide (EO) is a key intermediate in the production of a wide array of advanced materials, industrial solutions, and surfactants. Worldwide production of EO exceeds 27 billion kilograms, and the primary source involves selective reaction of the raw feedstock, ethylene, with oxygen over silver (Ag) catalysts. The Ag catalysts are promoted with complex combinations of five or more elements, each with different roles in catalysis. Chlorine (Cl) is one of the main promoters used to enhance the selectivity of ethylene conversion to EO. Although the EO selectivity of current industrial process technology is above 90%, opportunity exists to further increase the selectivity while simultaneously improving process efficiency and decreasing CO2 emissions. Yet, the interactions by which Cl increases selectivity are not fully understood, thus limiting further improvements in selectivity. The project combines advanced experimental, computational, and machine learning methods to better understand the catalytic mechanism of EO manufacture and the role played by the Cl promoter. The project will identify the steady-state active phase, surface structure, and surface intermediates present on Cl-promoted Ag nanoparticles at pressures and temperatures relevant for EO catalysis. Deeper understanding of the operation and design principles for EO catalysts, and increases in EO selectivity, could lead to transformative advances in the U.S. chemicals industry while simultaneously decreasing carbon emissions. In addition, the project will enhance graduate and undergraduate student education at the partner institutions, through cross-training of students between theory and experimental research groups, K-12 outreach via mentorship programs, and integration with educational opportunities that target high school women in STEM. The project explores the likelihood that surface structures for active Ag catalysts in many prior studies do not represent catalysts operating at practical conditions. Consequently, the dominant reaction mechanisms and the ways in which Cl improves selectivity remain unclear, particularly at industrially relevant reaction conditions and Cl coverages. The requirement for Cl as a promoter has long been associated with either induced reactivity of surface oxygen atoms or the selective titration of oxygen vacancies responsible for combustion present on oxide surfaces, yet those explanations presume specific mechanistic pathways that are dependent on catalyst surface structure. This project will establish the mechanism and reaction pathways for epoxidation and combustion, and the ways by which those pathways involve specific oxygen intermediates and oxygen vacancies upon oxygen-induced surface reconstructions. Additional insight will be obtained regarding how adsorbed Cl atoms affect surface reconstructions, the distribution and coverages of reactive oxygen species and vacancies, and the associated changes in reaction barriers, kinetics and selectivity. The inherent complexity of the catalytic system under reaction conditions is addressed through a combination of experimental (time-resolved in situ Raman spectroscopy) and computational methodologies (grand canonical Monte Carlo and molecular dynamics simulations employing machine learning-based potentials). More broadly, this project will provide opportunities for creating and validating methods (both experimental and computational) that address changes in surface structure during catalysis, a major challenge in the field.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.