Classification of drinking water in thailand by near infrared spectroscopy

Abstract

Near-infrared spectroscopy (NIRS) appears as a prominent technique for non-destructive food quality assessment. This research work was focused on evaluating the use of NIRS in classification of drinking water for the benefit of consumers. The water samples were six groups which were 1) Natural Mineral Water 2) Bottled Mountain Mineral Water 3) Bottled Underground Mineral Water 4) Filtered Water (including bottled and household filtered water) 5) Tap Water 6) Distilled Water. The sample was subjected to FT-NIR spectrometer with the pathlength of 2 mm in transflectance mode. The optical data of full spectral range (12500-3600 cm-1) were analyzed by statistic methods of principle component analysis (PCA), soft independent modeling by class analogy (SIMCA), partial least square discriminant analysis (PLS-DA) and cluster analysis (CA). The PLS-DA model confirmed its best performance among the methods used with the average accuracy of 89.22%. It was concluded that the NIRS measurement technique is an attractive and efficient tool for classifying the types of drinking water in Thailand.