增广独立成分分析用于果汁饮料中防腐剂的快速测定

采用正交设计法配制16组包含苯甲酸钠、山梨酸钾、对羟基苯甲酸甲酯钠3种防腐剂水性混合样品,分别测其紫外-可见(UV)光谱.果汁样品通过离心,取上清溶液,稀释一定的倍数测定其UV光谱.将果汁样品紫外光谱数据与模拟水样光谱融合得到增广光谱矩阵,采用独立成分分析(ICA)提取增广光谱矩阵中独立组分(IC)的浓度轮廓,将实际样品中3种防腐剂以外的其他成分作为背景加以扣除,以模拟标准样品3种防腐剂对应的浓度轮廓回归分析得到3种防腐剂测定的独立成分回归(ICR)模型,实际样品中防腐剂的IC值代入ICR模型预测得到相应防腐剂含量.将浓度IC轮廓进行最小二乘回归分析得到与其对应的光谱轮廓,可用作防腐剂与作为其他成分的背景信号的定性分析.方法用于市售某柠檬饮料中防腐剂测试分析,ICR模型的线性相关系数(R)均在0.99以上,测定结果与采用高效液相色谱法相当.

Augmented Independent Component Analysis Used for Determination of Multiple Food Preservatives in Fruit Juice Beverage

Synthetic samples of 16 mixed sample solutions containing three preservatives,sodium benzoate,potassium sorbate,and sodium methylparaben were prepared with orthogonal design method.The UV-Vis spectra(UV)of these prepared samples were measured in the range of 200-400 nm,and the spectra of fruit-vegetable beverage were measured after centrifuge.The measured spectra of the synthetic samples are augmented with that of the beverage sample,then processed by ICA procedure.The concentration profiles,i.e.Independent components(ICs)were first estimated by ICA,and the source spectral profiles can be estimated with the ICs based on least square method.The quantitative models were established by independent component regression(ICR)using the ICs background components removed,and the correlation coefficient of higher are higher than 0.99.As a practical application,three preservatives in lemon juice were determined with the results compared to that of HPLC method.