International Collaboration in Chemistry: CDS&E: Multiscale Simulations of Bifunctional Catalysis

加州大学圣地亚哥分校的Andreas W Goetz和特拉华大学的Dionisios G Vlachos获得了化学理论、模型和计算方法(CTMC)计划以及化学部计算和数据科学与工程(CDS&Amp；E)计划颁发的奖项的支持。高级网络基础设施部(ACI)是该奖项的联合资助者。Goetz和Vlachos为双功能催化开发和应用计算工具。该项目是与法国里昂高等学院的Philippe Sautet、Paul Fleurat-Lessard和Carine Michel进行的国际合作，他们提供免费的专业知识，并得到法国ANR相应奖项的支持。该项目开发了计算模型和软件，能够处理在溶液相中需要双功能催化剂的催化过程所固有的多尺度性质和复杂性。双功能催化剂对于将生物质转化为液体燃料和化学品非常重要，因此对于不依赖日益减少的石油资源的可持续未来也很重要，并最大限度地减少具有重大社会影响的全球变暖。计算机模拟可以在了解这些催化剂的功能以及指导改进的催化剂和工业上可行的工艺的开发方面发挥关键作用。所开发的计算方法被集成到可免费获得的开源软件库中，并与广泛使用的分子模拟包一起分发。研究生和本科生以及高中生和教师都参与了这个项目，他们通过实习和研究体验计划来培训下一代科学家。这项工作涉及从化学到化学工程到生物科学的多个领域，并有助于生物精炼的发展，对经济增长和减少二氧化碳排放具有明确的推动作用。该项目的发展包括用于快速筛选催化反应网络的新的力场和线性能量关系，用于提取动力学重要步骤的微观动力学模拟，允许粒子跨越QM/MM边界进行自适应交换的分子动力学模拟，以确定重要反应的活化能，以及密度泛函紧束缚理论的参数化，以最大化可访问的时间尺度。这些方法的结合首次能够探索在溶液中使用双功能(金属和酸碱)催化剂转化生物质的途径中的组合爆炸。最初，该项目的重点是甘油加氢脱氧为丙二醇，这方面有重要的实验数据。这些模拟有助于开发改进的双功能催化剂，最终可以改善生物炼油厂的工艺。作为该项目的一部分，在广泛使用和免费提供的开源软件中集成高效的多尺度模拟方法可以影响多个应用领域。

Andreas W Goetz of the University of California, San Diego and Dionisios G Vlachos of the University of Delaware are supported by an award from the Chemical Theory, Models and Computational Methods (CTMC) program and the Computational and Data-Enabled Science and Engineering (CDS&E) program in the Chemistry Division. The Division of Advanced Cyberinfrastructure (ACI) is co-funding this award. Goetz and Vlachos develop and apply computational tools for bifunctional catalysis. The project is an international collaboration with Philippe Sautet, Paul Fleurat-Lessard and Carine Michel of the Ecole Normale Superieure de Lyon in France who provide complimentary expertise and who are supported by a corresponding award of the French ANR. This project develops computational models and software capable of handling the multiscale nature and the complexity inherent to the catalytic processes entailing bifunctional catalysts in solution phase. Bifunctional catalysts are important for the conversion of biomass into liquid fuels and chemicals and therefore for a sustainable future that does not rely on dwindling petroleum sources and minimizes global warming with significant societal impact. Computer simulations can play a key role in understanding how these catalysts function and in guiding development of improved catalysts and industrially viable processes. The computational methods developed are integrated into freely available open source software libraries and distributed with a widely used molecular simulation package. Both graduate and undergraduate students are involved in the project, as well as high school students and teachers via internships and research experience programs to train the next generation of scientists. The work has relevance for multiple domains ranging from chemistry to chemical engineering to biosciences and aids the development of biorefineries with a clear impetus on economic growth and reduced CO2 emissions.The developments in this project encompass a new force field and linear energy relations for fast screening of catalytic reaction networks, microkinetic simulations to extract the kinetically important steps, molecular dynamics simulations with quantum mechanics/molecular mechanics (QM/MM) algorithms that allow an adaptive exchange of particles across the QM/MM boundary to determine the activation energies of the important reactions, and parameterizations of density functional tight binding theory to maximize the accessible time scales. The combination of these methods enables for the first time to explore the combinatorial explosion in pathways in the transformation of biomass using bifunctional (a metal and an acid/base) catalysts in solution. Initially the project focuses on the hydrodeoxygenation of glycerol into propanediol, for which significant experimental data is available. These simulations aid in the development of improved bifunctional catalysts that can ultimately lead to improved processes in biorefineries. The integration of efficient multi-scale simulation approaches in widely utilized and freely available open source software as part of this project can impact multiple application domains.