Department of Biological Sciences & Chemistry, College of Arts and Sciences, University of Nizwa, Oman.
Department of Chemistry, University of Malakand, KPK, Pakistan.
Food Chem. 2017 Apr 15;221:746-750. doi: 10.1016/j.foodchem.2016.11.109. Epub 2016 Nov 22.
New NIR spectroscopy combined with multivariate analysis for detection and quantification of camel milk adulteration with goat milk was investigated. Camel milk samples were collected from Aldhahira and Sharqia regions of Sultanate of Oman and were measured using NIR spectroscopy in absorption mode in the wavelength range from 700 to 2500nm, at 2cm resolution and using a 0.2mm path length CaF sealed cell. The multivariate methods like PCA, PLS-DA and PLS regression were used for interpretation of NIR spectral data. PLS-DA was used to detect the discrimination between the pure and adulterated milk samples. For PLSDA model the R-square value obtained was 0.974 with 0.08 RMSE. Furthermore, PLS regression model was used to quantify the levels of adulteration from, 0%, 2%, 5%, 10%, 15% and 20%. The PLS model showed the RMSEC=1.10% with R=94%. This method is simple, reproducible, having excellent sensitivity. The limit of detection was found 0.5%, while the limit of quantification was 2%.
本研究采用近红外光谱结合多元分析方法检测和定量骆驼奶掺羊奶。从阿曼苏丹国的 Aldhahira 和 Sharqia 地区采集骆驼奶样品，在吸收模式下使用近红外光谱仪在 700 至 2500nm 的波长范围内进行测量，分辨率为 2cm，采用 0.2mm 光程 CaF 密封池。多元方法如 PCA、PLS-DA 和 PLS 回归用于解释近红外光谱数据。PLS-DA 用于检测纯牛奶和掺假牛奶样品之间的区分。对于 PLSDA 模型，得到的 R-square 值为 0.974，RMSE 为 0.08。此外，还使用 PLS 回归模型定量测定 0%、2%、5%、10%、15%和 20%的掺假水平。PLS 模型的 RMSEC=1.10%，R=94%。该方法简单、可重现，具有良好的灵敏度。检测限为 0.5%，定量限为 2%。
