During automatic kinetic model generation a large number of reactions need to have kinetic data assigned to them. One of the main challenges is on how to account for the large scarcity of accurate data. The ever-growing computational power and availability of high performance computing solutions allows to envisage implementing on-the-fly ab initio calculations for a large number of reaction rate coefficients. Today extensive user knowledge and involvement is required to develop fast and effective ab initio based kinetic models. The present work introduces a series of automation procedures to minimize user involvement. It allows automated quantum chemical calculations for a large number of reactions for gas-phase processes with a wide variety in chemical species. 3D coordinates are calculated for all the reactants, products and transition states and automatically submitted to quantum chemistry packages. The latter yield accurate high-pressure limit rate coefficients. All steps are automated without the need for manual interventions, even for recognizing rotational frequencies and for verifying whether the saddle point searches lead to the appropriate transition state. A good agreement to literature data is found, i.e., most of the rate coefficients are within a factor of 2 of literature data.

在自动动力学模型生成过程中，需要为大量反应分配动力学数据。主要的挑战之一是如何解决大量准确数据的不足。不断增长的计算能力和高性能计算解决方案的可用性允许设想为大量的反应速率系数实现动态的从头计算。如今，需要广泛的用户知识和参与才能开发出快速有效的从头开始基于动力学模型。本工作介绍了一系列自动化程序，以最大程度地减少用户的参与。它允许对具有多种化学物种的气相过程的大量反应进行自动量子化学计算。计算所有反应物，产物和过渡态的3D坐标，并自动提交给量子化学程序包。后者得出准确的高压极限率系数。所有步骤都是自动化的，无需人工干预，甚至无需识别旋转频率并验证鞍点搜索是否会导致适当的过渡状态。发现与文献数据非常吻合，即，大多数速率系数在文献数据的2倍以内。