基于实验设计建模优化纳米复合光催化剂ZnO/SnO_2的制备条件

Optimizing the preparatory conditions of nano particle photocatalyst ZnO/SnO_2 coupled oxide based on the methods of experimental design and modeling

采用序贯实验思路,首先基于Plackett-Burman(PB)设计实验数据,经主效应分析和逐步回归筛选出影响制备ZnO/SnO_2粉体性能的主要因素;再以筛选出的Zn/Sn摩尔比及反应时间2个因素设计U_(12)(4×3)均匀设计实验,用PSO-2次回归方法基于该实验数据建立制备ZnO/SnO_2粉体光催化降解甲基橙脱色率DC的预测模型,通过非线性规划求模型最优解获得优化的ZnO/SnO_2制备条件。将在优化条件下制备获得的ZnO/SnO_2粉体用于光催化降解甲基橙,降解30min的DC实验平均值为96.17,模型预测值则为DC=95.66,DC实验值与预测值相对误差仅为0.53%,表明模型可靠。对该粉体采用XRD和SEM进行形貌、粒径及组成分析,表征结果显示该粉体大致呈球形、粒径主要分布在(20～30)rm,粉体中有少量Zn_2SnO_4产生。

Using the train of thought of sequential trial design, the two main effect production factors for preparing nanocrystalline ZnO/SnO2 powders, the molar ratio of Zrl/Sn and reaction time, were chosen to U12（4×3） uniform design experiments based on the data of Plackett-Burman（PB） design experiments by stepwise regression method and main effect analysis. The optimal prediction model for decoloration rate（DC） where the methyl orange（MO） solution was degradated using the prepared nanoparticle photocatalyst ZnO/SnO2, was established by using Quadratic Regression based on the Particle Swarm Optimization（PSO） algorithm and then the process recipes were optimized via nonlinear programming. Under optimised conditions, the prepared ZnO/SnO2 were also used in the photodegradation of methy orange solution, meanvalue 96.17% was achieved in 30 rain, and the predictive value is 95.6622%, so the relative error is just 0.525% which illustrates the best reliability of the optimal model. The coupled oxide prepared under optimised conditions was characterized by X-ray diffraction （XRD） and （SEM）. The resultes show that the nanoparticle is almost presented as sphericity whose grain diameter is distributed between （20-30） nm and that a small amount of nanometer Zn2SnO4 is formed in the mixture of coupled oxides.