Varietal discrimination of extra virgin olive oils by near and mid infrared spectroscopy Nicoletta Sinelli a, ⁎, Monica Casale b , Valentina Di Egidio a , Paolo Oliveri b , Daniele Bassi c , Debora Tura c , Ernestina Casiraghi a a Università degli Studi di Milano, Department of Food Science and Technology, Via Celoria 2, 20133 Milan, Italy b Università di Genova, Department of Drug and Food Chemistry and Technology, Via Brigata Salerno 13, I-16147 Genoa, Italy c Università degli Studi di Milano, Department of Crop Production, Via Celoria 2, 20133 Milan, Italy abstract article info Article history: Received 10 March 2010 Accepted 21 July 2010 Keywords: Extra virgin olive oils Near infrared spectroscopy Mid infrared spectroscopy Cultivar The use of near and mid infrared spectroscopy, combined with chemometric analysis, was explored as a tool to classify samples of Italian extra virgin olive oil on the basis of the cultivar. A total of 82 monovarietal samples (‘Casaliva’, ‘Leccino’ and ‘Frantoio’) of extra virgin olive oils were analysed. Several variables were measured: the free acidity, the peroxide value, spectrophotometric indices, the fatty acid composition, carotenoids, chlorophylls and tocopherol content. The same samples were also scanned by using NIR and MIR spectroscopy. The classification methods (LDA and SIMCA) were applied on chemical data and on the spectral data after having used the algorithm SELECT, as feature selection technique. The results showed that NIR and MIR spectroscopy is an interesting technique compared with traditional chemical index in classifying olive oil samples on the basis of the varietal origin. The spectroscopic methods could represent a reliable, cheap and fast classification tool, not requiring chemical analyses for discrimination among cultivars. © 2010 Elsevier Ltd. All rights reserved. 1. Introduction The extra virgin olive oil, owing to its high nutritional value and significant health benefits, is one of the most valuable ingredients of the Mediterranean diet. Extra virgin olive oil composition determines its intrinsic quality and could be influenced by several factors. Cultivar, environment and horticultural techniques affect the fruit physiology (Tura et al., 2007). Other factors as latitude, climatic conditions, irrigation regime, fruit ripening, harvesting and extraction technologies influence the distribu- tions of the fatty acids (D'Imperio, Dugo, Alfa, Mannina, & Segre, 2007; Stefanoudaki, Kotsifaki, & Koutsaftakis, 1999; Torres & Maestri, 2006) and triglycerides (Stefanoudaki, Kotsifaki, & Koutsaftakis, 1997). The effect of cultivar and of its interaction with the environment on the qualitative profile and the oxidative stability of extra virgin olive oil have been studied by determining the concentration profiles of saturated and unsaturated fatty acids, triglycerides, diacylglycerols and triacylglycerols, sterols, phenolic compounds, hydrocarbons, pigments and volatile components. These compounds differ according to the fruit variety (Lerma-Garcıá, Herrero-Martínez, Ramis-Ramos, & Simó-Alfonso, 2008). Traditionally, these parameters have been estimated by classical analytical methods, most of which are based on gas chromatography (GC) and high-performance liquid chromatography (HPLC). In the last few years, attention has been focused on authentication for genetic varieties of olive oils using nuclear magnetic resonance (NMR) fingerprinting (Camin et al., 2010; Mannina, Patumi, Proietti, Bassi, & Segre, 2001). Nevertheless, all these methods have several drawbacks, the most significant of which are low speed, high cost, and the necessity of sample pre-treatments and of highly-skilled personnel. Infrared spectroscopy in both the near (NIR) and mid (MIR) regions, combined with multivariate data analysis, has proven to be a successful analytical method for quantitative and qualitative model- ling of a wide variety of food and food process. These techniques facilitate real-time measurements at all stages of production, and they offer a fast, non-destructive and cost effective method of food analysis (Fagan & O'Donnell, 2008; Woodcock, Downey, & O'Donnell, 2008). Recent applications of NIR and MIR spectroscopy in edible oil analysis, reported in literature, include quality parameter determina- tion (Ahmed, Daun, & Prybylski, 2005; Azizian & Kramer, 2005) and adulteration detection (Banu & Mauer, 2002; Maggio, Cerretani, Chiavaro, Kaufman, & Bendini, 2010). These methods have been successfully applied in authentication studies of olive oil on the basis of geographical origin (Bendini et al., 2007; Casale, Casolino, Ferrari, & Forina, 2008; Galtier et al., 2008; Sinelli, Casiraghi, Tura, & Downey, 2008), but there is not any study regarding the application of these Food Research International 43 (2010) 2126–2131 ⁎ Corresponding author. Dipartimento di Scienze e Tecnologie Alimentari e Micro- biologiche, Università di Milano, Via Celoria 2, 20133 Milan, Italy. Tel.: +39 02 50319179; fax: +39 02 50319190. E-mail address: nicoletta.sinelli@unimi.it (N. Sinelli). 0963-9969/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodres.2010.07.019 Contents lists available at ScienceDirect Food Research International journal homepage: www.elsevier.com/locate/foodres