Eur Food Res Technol (2012) 234:457–465 DOI 10.1007/s00217-011-1652-0 123 ORIGINAL PAPER Application of synchronous Xuorescence spectroscopy for the determination of some chemical parameters in PDO French blue cheeses Khaled Abbas · Romdhane Karoui · Abderrahmane Aït-Kaddour Received: 29 September 2011 / Accepted: 16 December 2011 / Published online: 4 January 2012  Springer-Verlag 2012 Abstract The present study was aimed at investigating the potential of using synchronous Xuorescence spectros- copy (SFS) coupled with multivariate statistical analyses for the determination of some chemical parameters (pH, fat, dry matter, protein and soluble nitrogen) of French blue- veined cheeses belonging to four brands (FA-Fourme d’Ambert, FM-Fourme de Montbrison, BA-Bleu d’Auvergne and BC-Bleu des Causses). Three partial least square regression models with leave-one-out cross-valida- tion technique were considered in the present study. The Wrst one including the “Fourme cheeses” (FA and FM), the second one including the “Blue cheeses” (BA and BC) and the last one including the 4 Blue-veined cheeses (FA, FM, BA and BC). The models qualities were investigated prin- cipally by the R 2 (coeYcient of determination) and the RPD (ratio of standard deviation to root-mean-square error of cross-validation) factors. The results showed that SFS suc- ceeded to predict ash and protein in Blue (ash: R 2 = 0.90, RPD = 3.17; protein R 2 = 0.80, RPD = 2.24) or Fourme cheeses (ash: R 2 = 0.81, RPD = 2.29; protein R 2 = 0.81, RPD = 2.26) when considered individually, while SFS failed to predict all the physicochemical parameters when the two groups were analyzed jointly. Keywords French blue cheeses · Chemical characteristics · Fluorescence Introduction Nowadays, the dairy industry has come under increasing pressure to deliver products of high and constant quality into the market place [1]. Thus, a large number of analyses during the manufacturing and storage of foodstuVs, such as cheeses, are needed. However, food quality can be diYcult to assess, because the complex composition of food com- plicates quantitative and qualitative determination of the individual constituents [2]. For example, the physicochemical characterization of dairy products (e.g. milk, butter, cheese, etc.) is based on determining pH value, fat and mineral contents, nitrogen fractions, etc. When these parameters are determined, isola- tion of the compound of interest is necessary allowing to the destruction of the food. In addition, the destruction step is problematic, since it: (1) is tedious and laborious; and (2) required a huge quantity of chemical reagents as well as skilled operators. Moreover, the information contained in initial matrix can be lost when these components are removed from their context [2]. These methods are not eVective enough to cover the growing demand of listing that industry requires [3]. Taking this into account, there is a continuing demand for new, rapid and relatively cheaper methods for direct rapid determination of the quality measurements in food and food ingredients of food products. Spectroscopic tech- niques such as near infrared, mid-infrared, Xuorescence, etc. are fast, relatively of low cost and provide a great deal of information with only one test measurement. They are con- sidered as sensitive, non-destructive, rapid, environmentally K. Abbas · A. Aït-Kaddour (&) Clermont Université VetAgro Sup, Unité de Recherche CALITYSS, 89 Avenue de l’Europe, 63370 Lempdes, France e-mail: abderrahmane.aitkaddour@vetagro-sup.fr R. Karoui Faculty of Sciences Jean-Perrin, Artois University, Rue Jean Souvraz, 62300 Lens, France