Abstract
The seed or 'bean' of the coffee plant is an important crop, grown commercially across the world. Two species are commonly cultivated: Coffea arabica and Coffea canephora variant robusta. Analytical techniques for species identification, in particular of coffee products such as ground or 'instant' coffees, are of great importance. In this paper, mid-infrared spectroscopy is proposed as a rapid alternative to existing authentication methods, which are often time-consuming or difficult to implement successfully. A Fourier-transform infrared spectrometer is used for this work, equipped with a diffuse reflectance accessory. Statistical procedures comprising principal components analysis and classical discriminant analysis are applied to spectra of ground roast arabica and robusta beans, and results presented which demonstrate that the species of such samples can readily be identified.
Original language | English |
---|---|
Pages (from-to) | 321-326 |
Number of pages | 6 |
Journal | Food Chemistry |
Volume | 54 |
Issue number | 3 |
DOIs | |
Publication status | Published - 1995 |
Externally published | Yes |