声光可调-NIR法预测烟草中的总烟碱、总糖、还原糖和总氮

采用声光可调(AOTF)-近红外光谱(NIR)仪和Unscrambler化学计量学光谱分析软件,通过旋转扫描的方式测定了534个不同年份、不同地区、不同等级国产烤烟烟叶样品的NIR,采用PLS1建立了测定烟草中总烟碱、总糖、还原糖和总氮的校正模型,并用这些模型对41个未知烤烟烟叶样品进行了预测。结果表明:总烟碱、总糖、还原糖和总氮含量的NIR预测模型的决定系数R2分别为0.9869、0.9851、0.9733和0.9721,预测的平均相对偏差分别为4.21%、2.23%、2.91%和3.60%。AOTF-NIR技术适用于烟草中总烟碱、总糖、还原糖和总氮含量的快速分析。

Prediction of Total Nicotine,Total Sugar, Reducing Sugar and Total Nitrogen in Tobacco with AOTF-NIR

The near infrared (NIR) spectra from 534 domestic flue-cured tobacco samples of different crop years, habitats and grades were determined by rotating scanning with acousto-optic tunable filter-near infrared (AOTF-NIR) spectroscope and Unscrambler stoichiometry software. The calibration models for predicting total nicotine, total sugar, reducing sugar and total nitrogen in tobacco leaves were established by partial least square, and checked with 41 unknown flue-cured tobacco leaf samples. The results showed that the determination coefficients (R2) of the models for total nicotine, total sugar, reducing sugar and total nitrogen were 0.9869, 0.9851, 0.9733 and 0.9721, respectively; their average relative standard deviations were 4.21%, 2.23%, 2.91% and 3.60%, respectively; the method was suitable for rapid analysis of total nicotine, total sugar, reducing sugar and total nitrogen in tobacco.