作　　者： ; ; ; ; ; ;

机构地区： 广东工业大学轻工化工学院

出　　处： 《广西大学学报（自然科学版）》 2011年第3期385-388,416,共5页

摘　　要： 隧道状锰氧催化剂(cryptomelane)对二甲醚(DME)的催化燃烧有良好的促进作用,优化cryptomelane型锰氧催化剂的工艺条件可以提高催化燃烧的活性。在溶胶凝胶法合成锰氧催化剂的试验中,反应时间、焙烧温度和反应物的质量比对cryptomelane型催化剂的合成影响较大。为了得到合成的最佳工艺条件,以上述3要素为神经网络的输出变量,以XRD三强线的相对强度值为神经网络的输入变量,构建BP网络,优化锰氧催化剂的合成工艺条件。预测结果表明,焙烧温度为445.35℃,葡萄糖与高锰酸钾的反应时间为58.4 s,葡萄糖与高锰酸钾的质量比为1.1时,能合成较好的晶型的cryptomelane型锰氧催化剂,活性评价实验结果表明其催化效率良好。与实验结果相比较,模型具有较高的预测精度,预测误差在合理的范围内。 Tunnel-type cryptomelane has a good effect in the DME catalytic combustion reaction.Optimization of tunnel-type cryptomelane conditions can increase the activity of catalytic combustion.In the sol-gel synthesis of cryptomelane catalyst test,reaction time,calcination temperature and the mass ratio of reactants had great influences on the reaction of the cryptomelane.In order to obtain optimal conditions of the synthesis,the BP neural network was constructed with the above three elements as output variable,the XRD relative intensity of the three strongest lines as input variable,to optimize the conditions of the synthesis.Forecast results showed that when the temperature was 445.35 ℃,reaction time was 58.4 s and the mass ratio of glucose and potassium permanganate is 1.1,a better crystal eryptomelane with higher catalytic efficiency was synthesized.Compared with the experimental results,the model had a higher prediction accuracy,and the prediction error was within a reasonable range.

关 键 词： 型锰氧催化剂 人工神经网络 模拟计算 催化燃烧 溶胶凝胶法

领　　域： [理学] [理学]