Autentikasi Cepat Kopi Spesialti Arabika Java Preanger Kultivar Typica, Sigarar Utang dan Yellow Bourbon Menggunakan Spektroskopi UV dan Metode PLS-DA

Abstract

Java Preanger arabica specialty coffee of Typica, Sigarar Utang, and Yellow Bourbon is currently available in the premium coffee market at high prices and is currently one of the targets for coffee fraud. This study aims to evaluate the use of the UV spectroscopy method and the PLS-DA (partial least squares-discriminant analysis) method for rapid authentication of Java Preanger arabica specialty coffee with three different cultivars. Samples of arabica specialty ground roasted coffee and robusta were prepared for 550 samples with a weight of 1 g for each sample. All samples were extracted by standard procedures and diluted in a ratio of 1:70 (volume/volume). The diluted coffee sample was then taken as much as 3 mL, inserted into a quartz cuvette cell, and the spectra were taken using a UV-visible spectrometer in the wavelength range of 190-400 nm with 1 nm intervals. The results of the PCA transformation spectra at a wavelength of 250-400 nm showed that three cultivars of arabica specialty and robusta coffee could be distinguished and were in different clusters. However, Typica and Sigarar Utang arabica specialty coffees are in clusters that are close to each other and some of them overlap each other, indicating that the two arabica specialty coffees have similarities. The results of PLS-DA transformation spectra at a wavelength of 250-400 nm show that the developed classification model can be used for the classification of robusta and arabica coffee with three different cultivars with classification accuracy reaching 97.8 to 100%.