Identification of wool, cashmere, yak, and angora rabbit fibers and quantitative determination of wool and cashmere in blend: a near infrared spectroscopy study

Abstract

Near infrared spectroscopy coupled with chemiometric analysis was investigated as a fast and non destructive method for the identification of wool, cashmere, yak, and angora rabbit fibers in the raw and combed sliver state and for the quantitative determination of cashmere in cashmere/wool blends. The main differences among spectra of different animal hair arise from physical charateristics rather than chemical characteristics (mainly pigmentation and mean diameter) of animal fibers. The Soft Independent Modelling by Class Analogy method allows the classification of distinct fibers into separate groups with interclass distances ranging from 12.64 for the nearest classes (white cashmere and wool) to above 1000 for the most distant classes of white and pigmented fibers. Percentages of recognition and rejection of 100 % were found with the exception of a yak sample that was not rejected from the pigmented cashmere class (98 % of rejection). The prediction capacity of the model was also evaluated. Quantitative analysis was carried out using samples obtained by carefully mixing known amounts of wool and white cashmere. A standard error of the estimate of 8.5, a standard error of prediction of 13.10 and a coefficient of determination of 0.95 were calculated. From the results obtained, it can be concluded that near infrared spectroscopy can be used as a tool for an initial and rapid screening of unknown animal fiber samples in the raw and combed sliver states and for a fast and coarse estimate of the amount of cashmere in wool/cashmere blends.