CAREER: Uncertainty Quantification in the Rational Design of Bifunctional Catalysts

职业：双功能催化剂合理设计中的不确定性量化

• 批准号: 1254352
• 负责人: Andreas Heyden
• 金额: $ 40万
• 依托单位: University of South Carolina at Columbia
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-04-15 至 2019-03-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1254352&HistoricalAwards=false
• 关键词: CAREER; Uncertainty; Quantification; Rational; Design

Intellectual Merit For complex reactions with more than one key surface intermediate, bifunctional multiphase catalysts have a significant advantage over conventional monophase catalysts since each phase can potentially be adjusted independently to activate a key reaction step. Unfortunately, our understanding of multiphase catalysts whose activity and selectivity is primarily determined by the interfaces and adjacent surroundings remains relatively poor. Andreas Heyden of the University of South Carolina in this Faculty Early Career Development (CAREER) Program Award proposes utilizing computational studies to establish the underlying science of such bifunctional heterogeneous catalysts whose activity is largely determined by the three-phase boundary (TPB) of a gas-phase, a reducible oxide support, and a noble metal cluster or nanoparticle. The work will focus on identifying the origin or descriptors of the unique activity of Au and Pt catalysts supported on ceria and titania for the water-gas shift (WGS) reaction. The desire to better understand chemical reactions at the TPB is motivated by the fact that most heterogeneous catalysts consist of several solid phases and that although the overall catalyst activity and selectivity is often known to be a result of multiphase effects, the understanding of the chemical function of multiphase systems is relatively poor.Given the complexity of catalyst systems, connecting theoretical calculations to experimental observations becomes extremely challenging for multiphase systems. Heyden proposes to usemodern density functionals and to quantify uncertainty in the computational predictionsbased on a network of reactions. As a result, realistic and meaningful probabilistic predictionscan be made which significantly facilitates connecting theoretical calculations with experimentalobservations. Heyden plans to use modern Bayesian statistical tools to validate reaction sitemodels and to identify a required level of accuracy for a given prediction. Experimental information for comparison is secured through collaborations with scientists at Purdue University and the University of Southern California.Broader Impact Understanding the origin and identifying descriptors for the unique activity of oxide supported noble metal catalysts for the WGS has the potential to lead to the development of improved WGS catalysts for mobile and stationary applications. Furthermore, insights obtained from this study can likely be applied to various chemical reactions since the combination of reducible oxide supports and noble metals catalyze many reactions. In addition, the application of a computational strategy that quantifies uncertainty in computational catalysis predictions is important not only for reactions occurring at TPBs but for most complex reactions occurring at lower temperatures where even small errors in reaction energies lead to large uncertainties in predicted turnover frequencies, apparent activation barriers, and reaction orders.The research results of the proposed project will be integrated into a joint graduate and undergraduate elective ?Multiscale Modeling: From Electrons to Chemical Reactors? as part of the core chemical engineering curriculum to promote active, inquiry based learning. A continuous outreach program will be established with the Engineering Academy of a local urban high school (92% African American students) to increase the participation of underrepresented minorities in the study of engineering. Key components of this program include guest lectures, hands-on learning experiences of Academy students onthe USC campus, and a mentoring program for Engineering Academy students.

对于具有一个以上关键表面中间体的复杂反应，双功能多相催化剂比传统的单相催化剂具有显著的优势，因为每个相都可以独立调节以激活关键反应步骤。不幸的是，我们对活性和选择性主要由界面和邻近环境决定的多相催化剂的理解仍然相对较差。南卡罗来纳州大学的Andreas Heyden在这个教师早期职业发展（CAREER）计划奖中提出利用计算研究来建立这种双功能多相催化剂的基础科学，其活性在很大程度上取决于气相，可还原氧化物载体和贵金属簇或纳米颗粒的三相边界（TPB）。这项工作将集中在确定的起源或描述符的独特活性的Au和Pt催化剂上的二氧化铈和二氧化钛的水煤气变换（WGS）反应。更好地理解TPB处的化学反应的期望是由以下事实激发的：大多数非均相催化剂由几个固相组成，并且尽管通常已知总体催化剂活性和选择性是多相效应的结果，但是对多相系统的化学功能的理解相对较差。将理论计算与实验观察相联系对于多相系统来说是极具挑战性的。Heyden建议使用现代密度泛函，并基于反应网络量化计算预测中的不确定性。因此，可以做出现实的和有意义的概率预测，这显著地促进了理论计算与实验观测的连接。海登计划使用现代贝叶斯统计工具来验证反应位点模型，并确定给定预测所需的准确度。通过与普渡大学和南加州大学的科学家合作，确保了用于比较的实验信息。更广泛的影响了解氧化物负载的贵金属催化剂的WGS的独特活性的起源和识别描述符有可能导致改进的WGS催化剂的移动的和固定的应用程序的发展。此外，从这项研究中获得的见解可能适用于各种化学反应，因为可还原氧化物载体和贵金属的组合催化许多反应。此外，量化计算催化预测中的不确定性的计算策略的应用不仅对于在TPB发生的反应是重要的，而且对于在较低温度下发生的大多数复杂反应也是重要的，在较低温度下，即使反应能量中的小误差也会导致预测的周转频率、表观活化势垒、和反应顺序。拟议项目的研究成果将被整合到一个联合研究生和本科生选修？多尺度建模：从电子到化学反应器？作为核心化学工程课程的一部分，以促进积极的，基于探究的学习。将与当地城市高中的工程学院（92%的非裔美国学生）建立一个持续的外展计划，以增加代表性不足的少数民族在工程研究中的参与。该计划的主要组成部分包括客座讲座，南加州大学校园内学院学生的实践学习经验，以及工程学院学生的辅导计划。