化学
脱碳
均分解
键裂
氢原子萃取
光化学
热解
单重态
物理化学
计算化学
激进的
有机化学
激发态
催化作用
物理
核物理学
作者
Du Wang,Zhen‐Yu Tian,Zhi-Hao Zheng,Wang Li,Lingnan Wu,Jiu-Jie Kuang,Jiuzhong Yang
标识
DOI:10.1016/j.combustflame.2023.113240
摘要
As a polar solvent, N, N-Dimethylformamide (DMF) is a potential source of volatile organic compounds during industrial processing. The potential energy surface and pressure-dependent rate coefficients of DMF primary and secondary thermal decomposition were theoretically investigated by automatic ab initio calculations. A singlet aminohydroxycarbene route generating CO is found to be the most energetically favored channel among the unimolecular initiation reactions with a tight transition state. A kinetic model was proposed for DMF pyrolysis prediction and validated against the speciation data newly obtained from synchrotron vacuum ultraviolet photoionization mass spectrometry in a flow reactor within 773 K–1048 K and literature data in a jet stirred reactor within 450 K–900 K at atmospheric pressure. Satisfactory agreements were obtained for the major intermediates and products, but the model failed to capture the formation of some oxygenated compounds, possibly due to the missing reaction pathway for bimolecular addition. During pyrolysis, DMF is mainly consumed by the hydrogen abstraction reaction as the relatively high CN bond energy hinders the barrierless homolytic cleavage. Unlike aldehydes where the carbonyl site is favored, H-abstraction from DMF methyl sites is prevalent, although the acylamino group readily undergoes decarbonylation and β-scission forming oxygenated and nitrogenous parts. Consequently, CO and HCN are the major pyrolysis products. Besides, by incorporating essential high temperature kinetics, extended validation was conducted against the available intermediate temperature autoignition delay times (1000–1350 K, 2 bar) and laminar burning velocity data, and the reason for the discrepancy between the model prediction and experiments was inferred according to the sensitivity analyses.
科研通智能强力驱动
Strongly Powered by AbleSci AI