基于化学反应神经网络的纳米Al-PTFE复合体系反应动力学建模研究

A Reaction Kinetic Model of Nano Al-PTFE Composite from Chemical Reaction Neural Networks

为深入了解纳米铝粉-聚四氟乙烯(Al-PTFE)的化学反应机理,以热重(TG)实验数据为基础,采用化学反应神经网络(CRNN)对纳米Al-PTFE反应机理进行动力学建模;提出的建模方法可以建立至少包含4种反应中间体和4个关键反应的动力学模型,通过此模型可以准确预测Al-PTFE反应过程中的热重曲线;再结合已有的Al-PTFE体系动力学过程,推测反应体系的主要反应路径和中间产物。结果表明,纳米Al-PTFE的热解过程可能存在5步基本反应,其中C_(2)F_(4)的分解和气化反应、薄膜Al_(2)O_(3)的破损、内部Al的释放为其主要反应,活化能为200.9kJ/mol。中间物质可能包括CF、CF2、CF3等物质,固态产物为Al_(4)C_(3)和C,气态产物可能为CO_(2)和CO。表明化学反应神经网络能够对纳米Al-PTFE的热解反应进行建模,预测可能存在的反应机理和相应的动力学参数。

To understand the chemical reaction mechanism of nano aluminum-polytetrafluoroethylene(Al-PTFE),based on the experimental data of thermogravimetry(TG),the chemical reaction neural network(CRNN)is used to model the reaction mechanism of nano Al-PTFE.The CRNN method can establish a kinetics model with 4 intermediate products and 4 reaction pathways at least,which is able to predict the TG curves of nano Al-PTFE.Combined with the existing kinetic process of Al-PTFE system,to speculate the main reaction pathways and intermediate products of the reaction system.The results show that there may be 5-step basic reactions in the pyrolysis process of Al-PTFE,the decomposition and gasification of C_(2)F_(4),the breakage of Al_(2)O_(3) film,the release of internal Al as the main reaction,and the activation energy is 200.9kJ/mol.Intermediate substances may include CF,CF2,CF3 and other substances,solid products are Al_(4)C_(3) and C,and gaseous products may be CO_(2) and CO.Chemical reaction neural network can model the pyrolysis reaction of nano Al-PTFE,and predict the possible reaction mechanism and corresponding kinetic parameters.