近红外光谱法对茶叶中游离氨基酸含量的快速测定

为建立一种快速检测茶叶中游离氨基酸含量的方法,应用近红外光谱分析技术对贵州省不同产地的162个绿茶样品中的游离氨基酸含量进行检测,并结合偏最小二乘法(PLS),建立茶叶中氨基酸含量的预测模型。结果表明:通过标准归一化(SNV)光谱预处理,光谱范围选择6 657.74～6 178.16、5 793.11～5 688.98、4 493.33～4 350.62cm-1,主因子数为11,得到模型的内部交互验证相关系数(R)为0.981 56,交互验证均方差(RMSECV)为0.501;模型的预测值与实测值的相关系数为0.998,预测均方差(RMSEP)为0.312,稳定性试验得到的相对标准偏差(RSD)小于0.4%。综合分析,傅里叶变换红外光谱法与化学计量学方法相结合可以实现茶叶游离氨基酸的快速检测,满足检测要求。

Rapid Determination of Free Amino Acids in Tea by NIR Spectroscopy

In order to establish a method for rapidly detecting free amino acids of tea,162 green tea samples of different origins in Guizhou province were detected by near infrared spectroscopy(NIR).Combining with PLS,a mathematical model of amino acids in green tea was set up.According to the standard normal variate processing method,the selected spectra ranged between 6 657.74-6 178.16,5 793.11-5 688.98,4 493.33-4 350.62 cm-1.The number of principal factor was 11.The correlation coefficient of the model and the root mean square error of crossvalidation(RMSECV) were 0.981 56 and 0.501,respectively.Moreover the correlation coefficient and the root mean square error of prediction(RMSECV) between predicted value of the model and chemical determination value were 0.998 and 0.312,respectively,and the RSD was below 0.4% in the stability test.The result showed that the combination of FTIR spectroscopy and chemometrics was proved to be quick and reliable for free amino acid detection,which would satisfy the requirements of detection.