Excited State Specific Correlation Methods in Quantum Chemistry

在化学系化学理论、模型和计算方法项目的支持下，加州大学伯克利分校的Eric Neuscamman正在开发用于计算机模拟光驱动化学的新工具。无论是研究阳光造成的DNA损伤，还是研究模拟光合作用的工业过程，理解光驱动化学变化的机制都因其发生的小尺度和快节奏而变得困难。现代实验技术可以让我们对发生的事情有所了解，但往往留下一些关键问题没有答案，比如一个分子在吸收阳光的能量后会变成什么样的形状。Neuscamman小组将开发和部署新一代的计算机模型，忠实地模拟这些过程在高优先级的化学领域，目前的方法是有限的。特别是，光驱动过程将电子从分子的一边移动到另一边，或者同时移动多个电子，除了在最小的分子中，目前的工具无法准确模拟，而这些过程的关键应用发生在涉及数百个原子的技术和生物环境中。通过弥合这一差距，Neuscamman小组旨在加深我们对光驱动化学及其关键技术的理解。在这项研究的同时，Neuscamman小组将把它的推广工作扩展到中学生，教授数学优化方法的基本原理，这些方法支持从化学到机器学习的科学优先事项。这种以游戏为基础的拓展也将围绕斜率和曲率等概念播下种子，这样当学生最终发现自己在微积分课堂上时，他们就已经熟悉了微积分在现实世界中令人兴奋和有利可图的应用。通过让本科生担任此次拓展活动的讲师，该活动既可以拓宽中学生的视野，也可以加强对积极参与的本科生同事的理解。了解分子吸收光的跃迁能量是化学的核心。虽然几十年来的理论工作一直致力于预测电子（例如HOMO→LUMO）激发的跃迁能，但现有的理论方法仍然不能很好地服务于某些类别的电子激发。其中两个例子是电荷转移态，这是生物和人工光收集以及许多酶反应机制的核心，以及双重激发，这在色素和发色团的扩展π共轭网络以及许多过渡金属配合物中很常见。Neuscamman小组将在电子激发态平均场处理方面取得最新突破的基础上，构建一套高度准确、价格合理的耦合簇和相关方法，用于模拟大分子和分子组装中的电子激发态。这些方法有可能显著改善电荷转移和双激发态建模的最新技术，对化学、生物学、材料和系统化学具有潜在的广泛的长期科学影响。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

With support from the Chemical Theory, Models and Computational Methods program in the Division of Chemistry, Eric Neuscamman of the University of California at Berkeley is developing new tools for the computer simulation of light-driven chemistry. Whether studying DNA damage from sunlight or industrial processes that mimic photosynthesis, understanding the mechanisms by which light drives chemical change is made difficult by the tiny scales and fast pace at which the action occurs. Modern experimental techniques can offer glimpses of what is going on, but often leave key questions unanswered, such as the full series of shapes a molecule transforms itself through after absorbing energy from sunlight. The Neuscamman group will develop and deploy a new generation of computer models that faithfully simulate these processes in high priority areas of chemistry where current methods are limited. In particular, light-driven processes that move electrons from one side of a molecule to the other or that move multiple electrons at once cannot be simulated accurately by current tools except in the smallest molecules, whereas key applications of these processes occur in technological and biological settings involving hundreds of atoms. By bridging this gap, the Neuscamman group aims to deepen our understanding of light-driven chemistry and the crucial technologies that rely on it. In tandem with this research, the Neuscamman group will expand its outreach work to middle school students, teaching the underlying principles of mathematical optimization methods that support scientific priorities from chemistry to machine learning. This game-based outreach will also plant seeds around concepts like slope and curvature so that students are already familiar with exciting and lucrative real-world uses for calculus when they eventually find themselves in a calculus classroom. By engaging undergraduate students as instructors in this outreach, the activity will both broaden middle school student horizons and reinforce understanding for the active undergraduate co-worker participants. Understanding the energies of transitions in which molecules absorb light is central to chemistry. While decades of theoretical work have been dedicated to predicting the transition energies of electronic (e.g. HOMO→LUMO) excitations in particular, some categories of electronic excitation are still poorly served by available theoretical methods. Two examples of these are charge transfer states, which are central to biological and artificial light harvesting as well as many enzymatic reaction mechanisms, and double excitations, which are common in the extended π-conjugation networks of pigments and chromophores as well as in many transition metal complexes. The Neuscamman group will build on recent breakthroughs in the mean-field treatment of electronically excited states to construct what is anticipated to be a highly accurate, and affordable suite of coupled cluster and related methods for modeling electronically excited states in large molecules and molecular assemblies. These methods have the potential to significantly improve the state-of-the art in modeling both charge transfer and doubly excited states with potential broad long term scientific impacts for chemistry, biology, materials and systems chemistry.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.