摘要
碳捕集技术的发展对于实现碳达峰与碳中和意义重大.有机胺作为一种常用的二氧化碳(CO2)吸收剂备受关注,然而常用的有机胺溶剂仍存在化学吸收反应动力学较差等问题.该文提出一种集成反应动力学模型与数学规划方法的计算机辅助碳捕集有机胺溶剂设计框架.首先,基于量子化学方法,构建可调控有机胺化学吸收CO2的活化Gibbs自由能的反应动力学模型.其次,利用过渡态初猜结构生成方法,集成反应动力学模型与数学规划方法,自动、反向及优化设计有机胺分子结构,使有机胺分子结构的活化Gibbs自由能最小且满足其他性质要求.最后,通过相互作用区域指示函数验证有机胺设计结果在反应动力学方面的合理性.相比于工业上常用的乙醇胺,该框架设计得到的最优有机胺活化Gibbs自由能降低了10.7％,具有更快的化学吸收反应动力学速率.研究结果验证了设计框架可行有效,有助于碳捕集有机胺的研究.
Abstract
[Objective]In recent years,increasing emissions of greenhouse gases from industrial sources have led to extreme weather events.Thus,carbon capture has become an increasingly prominent concern for human society.Organic amines have emerged as popular carbon dioxide(CO2)chemical absorbents.However,currently used organic amine solvents encounter challenges,such as poor reaction kinetics of chemical absorptions,which need to be overcome using organic amines with higher performance.Moreover,there is limited research on the crucial property of Gibbs free energy of activation related to the reaction kinetics of chemical absorptions.The difficulty lies in the fact that the calculation of this property often involves the search for transition states in elementary reactions,and the initial structure of the transition state generally needs to be manually arranged,which greatly hampers the success of this search.To overcome this difficulty,a method for generating transition state initial guess structures is employed herein to automatically generate transition state structures for target organic amine molecules.Based on this generation method of transition state initial guess structures,a computer-aided framework for designing organic amine solvents is proposed by integrating a reaction kinetic model and a mathematical programming method.[Methods]First,the mechanism of the reaction between organic amines and CO2 is investigated based on quantum chemical methods and the transition state theory to construct a reaction kinetic model that can regulate the Gibbs free energy of activation of amine-based CO2 chemical absorptions.Additionally,a conformational isomer search approach is incorporated into the reaction kinetic model to improve the computational accuracy of the transition state energies.Second,by employing the transition state initial guess structure generation method,the reaction kinetic model is successfully integrated with the mathematical programming method to achieve automatic,reverse,and optimal design of organic amine molecular structures that have the lowest Gibbs free energy of activation for carbon capture reactions and satisfy all other property constraints being studied.Finally,the interaction region indicator function is used to verify the rationality of the designed organic amines regarding the reaction kinetics.[Results]Using the mathematical programming method,45 SMILES(simplified molecular input line entry system)representations that satisfy the design constraints were obtained.Through the transition state initial guess structure generation method,the corresponding transition state structures of organic amine molecules were obtained,and the Gibbs free energies of activation of these molecules were calculated by the reaction kinetic model.The final design results showed that the organic amine with the best performance had faster reaction kinetics than the primary amine used commonly in the industry,namely MEA(monoethanolamine),and its Gibbs free energy of activation for CO2 chemical absorption was 10.7％less than that of MEA.[Conclusions]In summary,the integration of the reaction kinetic model and the conformational isomer search approach yields molecular conformations with lower energies that are more realistic,resulting in more accurate calculations of Gibbs free energy of activation.The transition state initial guess structure generation method can automatically and rapidly generate initial molecular structures for transition states to calculate Gibbs free energy of activation,significantly improving the success rate of transition state searches.From the final design results,it is clear that the method proposed herein can discover promising monoamine molecules within a large chemical space.
基金项目
国家自然科学基金青年科学基金(22208042)
中央高校基本科研业务费专项(DUT22YG218)
中国博士后科学基金(2022M710578)
维普
万方数据