D3SC: Modern Data Analysis Tools for Prediction and Understanding in Catalyst Discovery

传统上，优化催化反应是一个高度经验性的过程，可能是劳动和材料密集型的，并且在科学上不能令人满意。合成领域的化学家的目标是通过开发将反应输出（例如反应速率或对特定产物的选择性）与分子结构相关联的方法来简化这种催化反应。由此产生的数学关系可以用来预测新催化剂的性能，而无需花费劳动力或材料来制造化合物。为了真正实现这种新兴的数据驱动的优化方法，必须开发有效描述或“参数化”分子的新方法。为此，在这个由化学部化学催化计划资助的项目中，犹他州大学的Matthew Sigman博士和他的团队正在定义能够在高精度水平上检查催化剂和起始材料在一系列催化反应中的结构特征的参数。通过这种优化方法生产的催化剂可以应用于具有社会重要性的反应，从医学相关化合物的先进制造到新型材料。该计划是高度合作的，因为几位研究人员参与了将Sigman博士的技术应用于催化的现代目标。Sigman博士积极参与以他的研究为基础的外联活动。 这些活动包括在夏季在他的实验室指导本科生研究人员，建立数字课程模块供本科生和研究生在学习中使用，并在当地小学进行演示。这些多样化的活动旨在鼓励处于不同教育阶段的学生追求他们对STEM领域的兴趣。Sigman博士正在构建用于配体筛选的多样化结构训练集、参数库和增强型预测工具，以快速识别用于催化的改进配体。这项工作的重点是特权配体支架，包括各种二胺配体（联吡啶，吡啶恶唑啉和双恶唑啉），单齿和双齿膦，膦基恶唑啉，和亚磷酰胺。在Sigman实验室内以及与广泛的研究团队合作获得了构建统计学显著模型的结果。最终的目标是使用与传统的物理有机工具相结合的反应输出的相关性，以发展一个更好的理解的基本现象负责如何一个特定的催化剂/底物组合执行。Sigman博士积极参与STEM外展计划，包括在夏季在他的实验室指导本科研究人员，建立数字课程模块供本科生和研究生在学习中使用，该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查进行评估，被认为值得支持的搜索.

Optimizing catalytic reactions has traditionally been a highly empirical process that can be labor and material intensive and scientifically unsatisfying. A goal of chemists in the synthesis field is to streamline this catalytic reactions by developing methods to correlate reaction outputs (such as reaction rates or selectivity for a certain product) to molecular structure. The resulting mathematical relationships then can be used to predict the performance of new catalysts without spending the labor or materials to make the compounds. To truly enable this emerging data-driven approach to optimization, new methods to describe or "parameterize" molecules effectively have to be developed. To this end, in this project, funded by the Chemical Catalysis Program of the Chemistry Division, Dr. Matthew Sigman of the University of Utah and his team are defining parameters that are capable at a high level of precision of examining structural features of catalysts and starting materials in a range of catalytic reactions. The catalysts produced by this optimization method can be applied to reactions of societal importance ranging from advanced manufacturing of compounds of medical relevance to novel materials. This program is highly collaborative in that several researchers are involved in applying Dr. Sigman's techniques to modern goals in catalysis. Dr. Sigman is actively engaged in outreach activities that build upon his research. These activities include mentoring undergraduate researchers in his laboratory during the summer, building digital curriculum modules for use by undergraduates and graduate students in their studies, and performing demonstrations at local elementary schools. These diverse activities are directed at encouraging students at different stages in their education to pursue their interest in STEM fields.Dr. Sigman is building diverse structural training sets, parameter libraries, and enhanced prediction tools for ligand screening to enable rapid identification of improved ligands for catalysis. The focus of this effort is on privileged ligand scaffolds, including various diamine ligands (bipyridines, pyridine oxazolines, and bisoxazolines), monodentate and bidentate phosphines, phosphino-oxazolines, and phosphoramidites. Results for building statistically significant models are acquired both within the Sigman laboratory as well as in collaboration with a wide range of research teams. The ultimate goal is to use the resultant correlations of the reaction outputs in combination with traditional physical organic tools to evolve a greater understanding of the underlying phenomenon responsible for how a particular catalyst/substrate combination performs. Dr. Sigman is actively engaged in STEM outreach programs including mentoring undergraduate researchers in his laboratory during the summer, building digital curriculum modules for use by undergraduates and graduate students in their studies, and performing demonstrations at local elementary schools.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.