基于近红外光谱定量分析植物水解蛋白掺假牛奶

Quantitative Analysis of Hydrolyzed Vegetable Protein(HVP) Adulterated Milk by Near Infrared Spectroscopy

植物水解蛋白是一种典型的蛋白质含氮掺假物质,被不法商贩作为廉价的蛋白质添加到牛奶中以提高其蛋白质含量。本研究自行配置不同浓度的植物水解蛋白掺假牛奶,应用近红外光谱分析仪漫反射模块直接扫描掺假样品,同时使用透射分析模块扫描三氯乙酸前处理的掺假样品;应用TQ软件分别建模并比较两种方法差异。结果表明:使用一阶导数&Norris Derivative滤波预处理,在8000~5000 cm^(-1),主成分数10,使用PCR构建了最优植物水解蛋白漫反射定量分析模型,RMSEC、RMSEP和R分别为0.224、0.214和0.98414,平均回收率为93.2479%。使用一阶导数&Norris Derivative滤波预处理,在10000~7000 cm^(-1),主成分数10,使用PCR构建了最优植物水解蛋白透射定量分析模型,RMSEC、RMSEP和R分别为0.175、0.138和0.99036,平均回收率为98.7351%。经过三氯乙酸处理的透射模型更加精确和稳定,可以用于牛奶中植物水解蛋白的检测,可以为牛奶品质控制和快速定量提供参考。

Hydrolyzed Vegetable Protein （HVP） was a typical adulterated compound that was used as an inexpensive protein by criminals to increase the nitrogen content in milk. Different concentrations of HVP adulterated samples were prepared in this study and were scanned directly with the near-infrared spectrometer （MRs）. Samples pretreated with trichloroacetic acid were analyzed by Transmission Analyzer, and the quantitative analysis models were established by TQ software to compare the difference between the two methods. Results showed that, the optimized quantitative analysis model by Fiber Optic Module was established with 10 principle components factors through preprocessing method of first derivative ＆ Norris derivative filter in the spectral region of 8000-5000 cm1. The root mean square error of cross （RMSEC）, root mean square error of prediction （RMSEP） and correlation coefficient （R） of the model were 0.224, 0.214 and 0.98414 respectively with a recovery of 93.2479%., The optimized quantitative analysis model by transmission analyzer was established with 10 principle components factors through preprocessing method of first derivative ＆ Norris Derivative Filter in the specWal region of 10000-7000 cm^-1. RMSEC, RMSEP and R of the model were 0.175, 0.138 and 0.99036 respectively with a recovery of 98.7351%. The quantitative analysis model pretreated with tricholoroacetic acid was accurate and stable, which could be used in rapid quantitative analysis of HVP in milk and provide reference for milk quality control.