【作者】 杨立斌；

【导师】 辛峰；

【作者基本信息】 天津大学 ， 化学工程， 2006， 硕士

【摘要】 环己酮肟是生产己内酰胺的关键中间体,在传统工艺中大多采用环己酮羟胺法合成。80年代中期,意大利Montedipe S.P.A成功开发了TS-1催化环己酮氨肟化的新工艺,克服了环己酮羟胺工艺的缺点,使得己内酰胺生产更加绿色环保,基本实现了“零排放”,合乎绿色化工发展的要求,为己内酰胺生产带来了新的前景,具有重要意义。然而环己酮氨肟化反应过程中催化剂TS-1分子筛存在失活问题。本文在大量实验的基础上,结合TS-1催化氨肟化反应过程的特点,根据推测的反应机理和合理简化,建立了三种反应机理下的本征动力学模型,经模型筛选及参数估值,结果表明双机理模型的计算值与实验值吻合较好。同时,利用幂函数形式的速率方程拟合了实验数据,并对方程的参数进行确定,获得了不同实验条件下的反应速率,然后再用高斯-牛顿法对动力学模型进行了参数估计,得出了氨肟化反应以及该反应体系中H2O2分解的动力学方程,并对该模型进行了残差分析和F检验,对氨肟化反应和H2O2分解的动力学模型的计算值与实验值进行了比较,平均相对误差绝对值分别为9.45%和6.21%。表明幂函数模型能合理、可靠地描述TS-1催化环己酮氨肟化反应的速率,且为进一步的过程开发与工程设计提供理论上的指导和设计上的依据;而H2O2分解副反应动力学模型仅适用于该反应体系。在研究TS-1分子筛催化环己酮氨肟化反应动力学的基础上,考察了TS-1分子筛催化剂失活前后活性的变化,分析出失活主要是因沉积物堵塞孔道所致,这些沉积物多为反应体系中生成的副产物。同时建立了失活动力学模型,并对模型参数进行估算,通过残差检验和统计检验说明失活动力学模型残差分布合理,参数回归显著,平均相对误差绝对值为6.23%,模型能够反映TS-1分子筛催化环己酮氨肟化的失活反应过程,是合理、可靠的。失活模型能很好的预测TS-1分子筛催化环己酮氨肟化反应失活过程。最后,对失活后的TS-1分子筛进行再生实验,结果表明,其催化剂结构没有发生明显变化,再生后催化剂的比表面积、孔体积及催化活性均得到较好地恢复。更多还原

【Abstract】 Cyclohexanone oxime, one of the most important precursors of caprolactam for nylon production, was conventionally produced by cyclohexanone-hydrocylamine process, which suffers of several drawbacks such as many reaction steps involved and large amount of byproduct of ammonium salts. A novel process based on the direct ammoximation of cyclohexanone with ammonia and hydrogen peroxide over TS-1 catalyst was developed by Montedipe S.P.A of Italy in the middle of 1980’s. This process does not suffer of the disadvantages above and meets the request of“Green Chemical Engineering”with zero emission. The dangerous reagents was not used and ammonium salts did not form in the process and cyclohexanone oxime is easily to separate by extraction. It is an environment-friendly process and is considered to be the most promising way for cyclohexanone oxime production.Notwithstanding the large amount of studies on this system recently, there is not a general agreement about the catalytic reaction mechanism. The deactivation of TS-1 for this reaction is usually considered to be the bottleneck for the industrialization of this process and the mechanism of deactivation is still an ambiguity.In this work, the ammoximation of cyclohexanone with ammonia and hydrogen peroxide over TS-1 catalyst in the liquid phase was investigated. The intrinsic kinetics and deactivation kinetics of ammoximation of cyclohexanone over TS-1 was derived. By analyzing the reaction system, we proposed three kinds of reaction mechanisms. The experimental and simulation results show that the reaction scheme based on the two-mechanisms may be the most possible reaction pathway. The proposed reaction mechanism agrees well with the experimental results.The rate equations for ammoximation of cyclohexanone and decomposition of hydrogen peroxide as well as catalyst deactivation were obtained and the parameters were estimated by Gauss-Newtonian algorithm. The comparison of predicted values from the proposed models with experimental results showed that the absolute average relative error were 9.45% for ammoximation, 6. 21% for decomposition of H2O2 and 6.23% for catalyst deactivation, which showed the reliability of the obtained rate equation and were expected to be used for process design.The structure properties of the catalyst before and after deactivation was analyzed and the results show that some by-products of the reaction covered the surface of the catalyst and blocked the pores in the catalyst, which may be the main reason forcatalyst deactivation. The deactivated titanium silicalite-1 zeolite can be regenerated by heating in air. The activity, the special surface area and also the pore volume of the regenerated catalyst were almost the same as the fresh one, indicating that the structure of the catalyst did not change during deactivation and the deactivation was mainly caused by reversible pore blocking.更多还原