论文部分内容阅读
将偏最小二乘法(PLS)用于紫外分光光度数据的解析,建立了同时测定甲基苯甲醛3种同分异构体的模型。在230～304 nm范围内,将测得的48个样品的吸光度值作为校正集,另18个样品的吸光度值作为预测集用于建模。所建立的邻、间、对甲基苯甲醛模型的平均回收率分别为101.2%、100.2%和98.9%;均方根误差(RMSE)分别为0.2667、0.3853和0.2118;预测浓度范围分别为4.6～16.2μg/mL、5.8～17.4μg/mL和6.5～20.6μg/mL。讨论了混合物中3种同分异构体浓度比例对测定结果的影响,并确定了最佳的浓度比例范围。对模拟样品进行加标回收率试验。并通过与顺、反丁烯二酸两种同分异构体测定结果的比较,得出了有意义的结论。
Partial Least Squares (PLS) was used for the analysis of UV spectrophotometric data, and the simultaneous determination of three isomers of methylbenzaldehyde was established. In the range of 230 ~ 304 nm, the absorbance values of the 48 samples were taken as the calibration set, and the absorbance values of the other 18 samples were used as the prediction set for the modeling. The average recoveries of the o, p, p-methylbenzaldehyde models were 101.2%, 100.2% and 98.9%, respectively. The root mean square error (RMSE) were 0.2667,0.3853 and 0.2118 respectively. The predicted concentrations ranged from 4.6 ~ 16.2 μg / mL, 5.8-17.4 μg / mL, and 6.5-20.6 μg / mL. The effect of the ratio of the three isomers in the mixture on the determination results was discussed and the optimum concentration ratio range was determined. Spike recovery test on simulated samples. Through the comparison with the determination results of two isomers of cis and fumaric acid, the conclusion is reached.