随着酶的定向进化技术（2018年诺贝尔化学奖）的发展和广泛应用，利用特制的生物酶定向制备单一组分、特定分子结构的碳氢燃料为未来喷气燃料的生产提供了一条定制化的路径。目前，我国相关的研究机构已经开始利用特制的生物酶制备不同分子结构的生物基新型碳氢燃料，并从中探索适宜高超声速飞行器需要的具有高密度、高燃烧热值、高热沉、低结焦积碳等性能的优质新型燃料。但在这类生物基新型碳氢燃料的探索研究过程中，新型燃料的物理和热解性质的实验及模拟数据还呈现空白，并且也缺乏一套燃料性质研究方案来认识燃料分子结构与其物理、热解性质之间的基础关系。面对大量的新型燃料，如何快速的获取它们的基本物理、化学特性，评估它们密度、燃烧热值、热沉、结焦积碳等性能，并从中筛选出满足需求的高性能燃料成为一个急需解决的科学问题。
本篇论文初步建立了一套适用于新型燃料热解性质研究和结构优化的研究方案，以研究新型碳氢燃料热解性质与其分子结构之间的关系，从中探索具有低结焦积碳潜力的新型碳氢燃料分子结构，从而为高性能生物基喷气燃料的研发提供帮助。本文建立的研究方案主要包括新型燃料的物性预测、动力学建模及模型优化、热解实验及模拟以及分子结构优化四部分。
在新型燃料的物性预测方面，发展了一套物性预测方法。该方法通过引入高物性估算精度的基团贡献法和热物性计算软件SUPPERTRAPP实现了对任意已知结构的新型燃料的可靠物性预测，从而帮助研究者筛选出满足相关物性要求的新型燃料进行后续研究。
在动力学建模及模型优化方面，通过引入化学反应动力学模型生成程序RMG（Reaction Mechanism Generator）建立了可用的燃料动力学模型；基于灵敏度分析和线性规划提出了动力学模型活性参数选择方法，该活性参数选择方法的特点是在活性参数选择过程中可视化了所选择的活性参数对模型输出可能产生的影响，通过评估活性参数对模型的影响情况来保证所选活性参数的可靠性；并在活性参数选择方法的基础上结合梯度下降法提出了能同时对数百个活性参数同时进行优化的优化方法，以保证选择和优化的活性参数覆盖模型中大部分的重要参数，从而得到较好的优化效果；并且基于Matlab语言和开源的Cantera化学反应计算工具库编写了能够使用上述两种方法的动力学模型优化程序OptChem，OptChem在相应工况上的模拟结果与Chemkin的模拟结果相同，在灵敏度的定性分析上OptChem和Chemkin的效果基本相同。OptChem程序最主要也最具特色的功能便是能对包含数百个物种和数千个化学反应的详细动力学模型中的化学反应参数进行优化，使模拟结果能与实验结果较好的相符。
在热解实验及模拟方面，为新型燃料设计了低压、高压热解实验，研究了新型碳氢燃料热解性质与其分子结构之间的关系。其中，低压热解实验在合肥国家同步辐射实验室质谱分析线站上进行，利用同步辐射真空紫外光电离质谱方法确定了两种生物基新型燃料FA和FB的部分低压热解产物的种类以及这些热解产物摩尔分数随热解温度的变化关系，并且通过热解产物中乙烯和苯的生成量判断了新型燃料FA具有更优的低结焦积碳性能。在高压热解实验上，搭建了能够模拟燃料在发动机冷却槽道中流动热解工况的高压热解实验系统，以获得燃料在实际应用环境下的热解数据。在本文工作中开展了多种生物基新型燃料和典型燃料的高压热解实验，但基于项目要求，仅介绍十氢化萘和JP-10两种典型燃料的高压热解实验，相关实验填补了这两种典型燃料在对应工况下热解性质的研究空白，而且还能为新型燃料在相关工况下性能的分析提供参照。通过RMG和OptChem建立了生物基新型燃料FA和FB的低压热解动力学模型，以及十氢化萘和JP-10的高压热解动力学模型，并基于燃料的动力学模型分析了燃料在对应实验条件下热解时苯、甲苯这类芳香烃的主要生成路径，发现了对于类FA结构的燃料可以基于分子结构调整开展低结焦积碳性质的燃料探索；验证了使用RMG生成类FA结构的燃料低压、高压动力学模型是相对可靠的，这为使用RMG进行类FA结构的燃料的分子结构优化提供了支撑。
在分子结构优化上，以一组类FA结构的新型碳氢燃料分子为代表，研究了这类燃料分子结构与其热解性质之间关系，并从中找到了潜在的低结焦低碳性能的燃料分子结构。在分子结构优化过程中提出了分子结构的优化思路，以寻找低芳香烃产量的燃料分子为例，该优化思路为通过灵敏度分析寻找对燃料热解过程中芳香烃生成产生重要影响的反应，并且从这些重要反应所包含的反应物中筛选出燃料初始热解所产生的基团，然后通过反应速率观察这些基团的产生是否有利于芳香烃生成，利于芳香烃生成的基团结构应当改变，不利于芳香烃生成的基团结构则保留。

With the development and wide application of directed evolution of enzymes (2018 Nobel Prize in Chemistry), using specialized biological enzymes to directionally prepare hydrocarbon fuels with a single component and specific molecular structure provides a customized path for the production of future jet fuels. At present, relevant research institutions in China have begun to use specialized biological enzymes to prepare new bio-based hydrocarbon fuels with different molecular structures, and to explore new high-quality fuels with high density, high combustion heat value, high heat sink, and low coking and carbon deposition suitable for hypersonic aircraft. However, in the exploration and research process of such new bio-based hydrocarbon fuels, the experimental and simulation data of the physical and pyrolysis properties of the new fuels are still blank, and there is also a lack of a fuel property research protocol to understand the basic relationship between the molecular structure of the fuel and its physical and pyrolysis properties. In the face of a large number of new fuels, how to quickly obtain their basic physical and chemical properties, evaluate their density, combustion heat value, heat sink, coking and carbon deposition, and other properties, and select high-performance fuels that meet the needs has become an urgent scientific issue to be solved.
This dissertation initially established a research protocol suitable for the pyrolysis properties and structure optimization of new hydrocarbon fuels to study the relationship between the pyrolysis properties and the molecular structure of new hydrocarbon fuels, and to explore the molecular structure of new hydrocarbon fuel with low coking and carbon deposition potential, thereby providing help for the development of high-performance bio-based jet fuels. The research protocol established in this dissertation mainly includes four parts: physical property prediction, kinetic modeling and model optimization, pyrolysis experiment and simulation, and molecular structure optimization.
In terms of physical properties prediction of new fuels, a physical property prediction method has been developed. By using the group contribution method with high physical property estimation accuracy and the thermophysical property calculation software SUPPERTRAPP, the method can reliably predict the physical properties of new fuel with any known structure, thereby helping researchers to screen new fuels that meet the relevant physical property requirements for subsequent research.
In terms of kinetic modeling and model optimization, the RMG (Reaction Mechanism Generator) program was used to establish the usable fuel kinetic model. Based on sensitivity analysis and linear programming, an active parameter selection method for kinetic model was proposed. The characteristics of the proposed active parameter selection method is preliminary visualization of the possible influence of the selected active parameters on the model outputs in the process of parameter selection, ensuring the reliability of the selected active parameters. On the basis of the active parameter selection method combined with the gradient descent method, an optimization method that can simultaneously optimize hundreds of active parameters was proposed to ensure that the selected and optimized active parameters cover most of the important parameters in the model, so as to obtain better optimization effect. And based on Matlab language and the open source chemical reaction calculation tool library Cantera to write a kinetic model optimization program OptChem, which can use the above two methods. Under the same simulation conditions, the simulation results of OptChem and Chemkin are the same, and the sensitivity analysis results of OptChem and Chemkin are basically the same. The main and most distinctive feature of OptChem is the ability to optimize the chemical reaction parameters in a detailed kinetic model containing hundreds of species and thousands of chemical reactions, so that the simulation results can be in good agreement with the experimental results.
In terms of the pyrolysis experiment and simulation, the low pressure pyrolysis experiment and the high pressure pyrolysis experiment were designed for the new hydrocarbon fuels, and the relationship between the pyrolysis property and the molecular structure of the new hydrocarbon fuels was studied. The low-pressure pyrolysis experiment was carried out on the mass spectrometry station of the National Synchrotron Radiation Laboratory in Hefei. The types of partial low-pressure pyrolysis products of two new bio-based fuels, FA and FB, and the relationship between the mole fractions of these pyrolysis products and pyrolysis temperature were determined by synchrotron radiation vacuum ultraviolet photoionization mass spectrometry. According to the production of ethylene and benzene in the pyrolysis products, it was judged that the new fuel FA had better low coking and carbon deposition performance. In the high-pressure pyrolysis experiment, a high-pressure pyrolysis experiment system was built to simulate the flow pyrolysis condition of fuel in the engine cooling channel, and the pyrolysis data of fuel in the actual application environment can be obtained. The high-pressure pyrolysis experiments of a variety of new bio-based fuels and typical fuels were carried out, but according to the requirements of the project, only the high-pressure pyrolysis experiments of naphthol dechydropyrolysis and JP-10 were introduced in this dissertation. The high-pressure pyrolysis experiments of two typical fuels, decalin and JP-10, filled the research gap of pyrolysis properties of these two fuels under corresponding working conditions and provide reference for the analysis of the performance of new fuels under relevant working conditions. The low-pressure pyrolysis kinetic models of new bio-based fuels FA and FB, and the high-pressure pyrolysis kinetic models of decalin and JP-10 were established by RMG and OptChem. The main production paths of aromatic hydrocarbons such as benzene and toluene during the pyrolysis of fuels under corresponding experimental conditions were analyzed based on the kinetic models of above-mention fuels. It was found that fuels with FA-like structure are able to explor for low coking and carbon deposition properties based on molecular structure adjustment. It is verified that the low pressure and high pressure kinetic models of fuels with FA-like structure generated by RMG are relatively reliable, which provides support for the molecular structure optimization of FA-like fuels using RMG.
In terms of molecular structure optimization, taking a group of new FA-like fuels as representatives, the relationship between the molecular structure of such fuels and their pyolysis properties was studied, and the potential fuel molecular structures with low coking and carbon deposition performance were found. The optimization idea of molecular structure was proposed in the process of molecular structure optimization. Taking the search for fuel molecules with low aromatic hydrocarbon production as an example, the optimization idea is to find the reactions that have important effects on the production of aromatic hydrocarbons in the pyrolysis process of fuel through sensitivity analysis, and screen out the groups produced by the initial pyrolysis of fuel from the reactants included in these important reactions. Then, through the reaction rate, observe whether the generation of these groups is conducive to the production of aromatic hydrocarbons, and the structure of the groups conducive to the production of aromatic hydrocarbons should be changed, and the structure of the groups that are not conducive to the production of aromatic hydrocarbons are retained.