Kinetic analysis guided by artificial intelligence

人工智能正在给包括科学领域在内的社会方方面面带来革命性的变化，化学也不例外。人工智能已经被用于解决化学上的重大挑战，例如设计合成路线[1，2]和根据蛋白质的氨基酸序列预测蛋白质的三维结构[3，4]。此外，人工智能在优化反应和求解电子薛定谔方程方面取得了令人印象深刻的进展[5]。然而，人工智能在催化领域的应用仍然有巨大的潜力。在这个项目中，我们将研究人工智能模型的潜力，以彻底改变动力学分析的设计和执行，以了解反应机理。基于我们最近在《自然》杂志上发表的应用机器学习来阐明单分子反应中的反应机理[7]，我们处于独特的地位来推进这一令人兴奋的领域。该项目将结合传统的湿法实验室技术来筛选和开发催化反应，以及使用现代反应监控工具来收集数据，并对这些数据进行人工智能辅助分析。我们将研究各种现代催化体系在药物合成、先进材料和可持续发展等领域的应用。[1]自然2020 83；[2]自然2018 604；[3]自然2021 583；[4]自然2021 590；[5]自然2021 89；[6]NAT。化学。2020 891；[7]自然2023 689。

Artificial intelligence is revolutionizing all aspects of society, including the field of science, and chemistry is no exception. AI has already been used to solve significant challenges in chemistry, such as designing synthetic routes[1, 2] and predicting the 3D structures of proteins from their amino acid sequences[3, 4]. Additionally, AI has made impressive strides in optimizing reactions[5] and solving the electronic Schrodinger equation[6]. However, there is still significant potential for AI to be applied in the field of catalysis.In this project, we will examine the potential of AI models to revolutionize the design and execution of kinetic analysis for understanding reaction mechanisms. Building upon our recent Nature publication in the application of machine learning for the elucidation of reaction mechanisms in unimolecular reactions,[7] we are uniquely positioned to advance this exciting field.The project will involve a combination of traditional wet-lab techniques to screen and develop catalytic reactions, as well as modern reaction monitoring tools to collect data, and AI-assisted analysis of this data. We will investigate a variety of modern catalytic systems with applications in areas such as drug synthesis, advanced materials, and sustainability.[1] Nature 2020 83; [2] Nature 2018 604; [3] Nature 2021 583; [4] Nature 2021 590; [5] Nature 2021 89; [6] Nat. Chem. 2020 891; [7] Nature 2023 689.