Development of a methodology for determining reaction pathways and rate constants of complex combustion networks

在许多化工和燃烧工程应用中，要求化学反应机理能够准确地对过程和装置进行建模和优化。由于大量的反应和物种，通常不可能从实验上获得的数据确定所有基本反应的速率系数。因此，在过去的几年里，从头计算扩大了实验。然而，这样的计算需要大量的化学直觉，大量的人工和计算机时间，而且它们的准确性有时仍然受到所用模型的简化的影响。在这项工作中，我们建议通过使用反应力场进行分子动力学模拟来自动识别反应路径和过渡态来克服这些问题。后者将通过量子力学的自动程序进行精炼，并用于计算反应速率。分子模拟也可以用来放松通常使用的过渡态理论模型中的一些假设。因此，我们希望开发一种方法，使反应模型的确定比目前可能的更有效和更准确。这种方法将被用来研究新型生物燃料的点火特性和污染物的形成。

In many chemical and combustion engineering applications, chemical reaction mechanisms are required to model and optimize processes and devices accurately. Due to the large number of reactions and species, often the determination of rate coefficients of all elementary reactions is not possible from experimentally accessible data. Therefore, experiments have been augmented by ab initio calculations in the last years. Such calculations, however, require much chemical intuition, large amounts of human and computer time, and still their accuracy sometimes suffers from the simplifications made in the models used. In this work, we suggest to overcome these issues by performing molecular dynamics simulations with a reactive force field to automatically identify reaction paths and transition states. The latter ones will be refined in an automated procedure by quantum mechanics and used to compute reaction rates. Molecular simulations can also be used to relax some of the assumptions made in the transition state theory models usually used. Therefore, we expect to develop a methodology that will allow the determination of reaction models more efficiently and with higher accuracy than it is presently possible. This methodology will be used to investigate the ignition properties and the formation of pollutants of novel biofuels.