High-value products, such as hazelnuts, are particularly vulnerable to fraud due to their price dependence on geographical origin. Guaranteeing hazelnuts’ authenticity is essential for consumer trust and safety. Stable isotope analysis has become a reference method for origin authentication as it is reliable, robust, and easily transferable across laboratories. However, multiple isotopic markers coupled with chemometric techniques are often needed to authenticate food provenance accurately. In this study, we focused on assessing the potential of bulk δ18O, along with δ2H and δ13C of the main fatty acids, as hazelnut-origin authenticity markers. PLS-DA classification models were developed to differentiate samples (n = 207) according to their region of origin. This multi-isotopic approach provided promising external validation results, achieving a 94% global correct classification rate in discriminating hazelnuts from regions with distinct geographical and environmental conditions. This study lays the groundwork for further model development and evaluation across additional production areas and harvest years.

This work discusses food authentication tools based on the volatile metabolome assessed by gas chromatography-mass spectrometry (GC-MS). Volatile organic compounds (VOCs) serve as markers for food flavour and aroma and, in some cases, they hold potential for food authentication. On top of this, advanced data analysis approaches enhance analytical data extraction and understanding. Untargeted methods provide comprehensive authenticity insights, surpassing traditional targeted approaches, and automated processing improves robustness and reduces user dependency. Chemometric tools, particularly classification methods, are extensively used for authentication based on both GC-MS fingerprinting and untargeted profiling of VOCs. Despite challenges in transferability, untargeted methods perform well in authentication models, making them valuable for internal quality control and official control guidance for fraud detection. This work highlights the need for further research to establish untargeted authentication tools as official methods in the near future.

Official control methods to detect olive oil (OO) adulteration fail to provide satisfactory consumer protection. Thus, faster and more sensitive screening tools are needed to increase their effectiveness. Here, the official method for adulterant detection in OO was compared with three untargeted screening methods based on tri-acylglycerol analysis using high-throughput (FIA-HESI-HRMS; HT-GC-MS; HPLC-RID) and pattern recognition techniques (PLS-DA). They were assayed on a set of genuine and adulterated samples with a high natural variability (n = 143). The sensitivity of the official method was 1 for high linoleic (HL) blends at >= 2 % but only 0.39 for high oleic (HO) blends at >= 5 %, while specificity was 0.96. The sensitivity of the screening methods in external validation was 0.90-0.99 for the detection of HL and 0.82-0.88 for HO blends. Among them, HT-GC-MS offered the highest sensitivity (0.94) and specificity (0.76), proving to be the most suitable screening tool for OO authentication.

Microbiological, chemical, and sensory analyses were coupled to understand the origins of kombucha organoleptic compounds and their implication in the flavor of the kombucha beverage. By isolating microorganisms from an original kombucha and comparing it to monocultures and cocultures of two yeasts ( Brettanomyces bruxellensis and Hanseniaspora valbyensis ) and an acetic acid bacterium ( Acetobacter indonesiensis ), interaction effects were investigated during the two phases of production. 32 volatile compounds identified and quantified by Headspace-Solid Phase-MicroExtraction-Gas Chromatography/Mass Spectrometry (HS-SPME-GC/MS) were classified according to their origin from tea or microorganisms. Many esters were associated to H. valbyensis , while alcohols were associated to both yeasts, acetic acid to A. indonesiensis , and saturated fatty acids to all microorganisms. Concentration of metabolites were dependent on microbial activity, yeast composition, and phase of production. Sensory analysis showed that tea type influenced the olfactive perception, although microbial composition remained the strongest factor. Association of B. bruxellensis and A. indonesiensis induced characteristic apple juice aroma.

According to the last report from the European Union (EU) Food Fraud Network, olive oil tops the list of the most notified products. Current EU regulation states geographical origin as mandatory for virgin olive oils, even though an official analytical method is still lacking. Verifying the compliance of label-declared EU oils should be addressed with the highest priority level. Hence, the present work tackles this issue by developing a classification model (PLS-DA) based on the sesquiterpene hydrocarbon fingerprint of 400 samples obtained by HS-SPME-GC-MS to discriminate between EU and non-EU olive oils, obtaining an 89.6% of correct classification for the external validation (three iterations), with a sensitivity of 0.81 and a specificity of 0.95. Subsequently, multi-class discrimination models for EU and non-EU countries were developed and externally validated (with three different validation sets) with successful results (average of 92.2% of correct classification for EU and 96.0% for non-EU countries).

Nowadays, 144 Protected Designations of Origin (PDO) and Protected Geographical Indications (PGI) under the EU quality schemes correspond to extra virgin olive oil (EVOO). As endorsed by the EU Food Fraud Network, PDO/PGI EVOOs are particularly vulnerable to fraudulent practices because of their high economic value. For this reason, the present study aims to develop an instrumental tool to assess the compliance of EVOO with PDO label-declaration using a large sample set (n = 350). As a case study, PDOs from Catalonia were used. Therefore, discriminant analysis based on the sesquiterpene hydrocarbon fingerprint determined by HS-SPME-GC-MS achieved to correctly classify an average of 93.6% of samples among the four Catalan PDOs, leaving unassigned the 6% of the total sampling (external validation results for 3 iterations). On the other hand, the proposed strategy allowed discriminating each Catalan PDO from non-PDO samples produced in different geographical areas with an efficiency between 95% and 99%.

Virgin olive oil (VOO) is a highly appreciated product fundamental in the Mediterranean diet. Since its sensory attributes are greatly influenced by the olive cultivar, the varietal authentication of VOOs is needed to protect consumers from misleading information. The present study aims to evaluate the suitability of sesquiterpene hydrocarbon (SH) fingerprint as VOO cultivar marker beyond geographical, agronomical and processing conditions. The study was mainly focused on Mediterranean Arbequina oils. SH profile of more than 400 VOOs from 6 counties and 38 different cultivars and coupages was analysed by Headspace Solid Phase Microextraction-Gas Chromatography-Mass Spectrometry (HS-SPME-GC-MS). Partial Least Square-Discriminant Analysis (PLS-DA) classification models were built with the aligned chromatograms. A binary PLS-DA model was built to distinguish 'Arbequina' oils from those of other cultivars (non-'Arbequina' class) and it was externally validated. The results of the external validation showed a 95.1% of overall correct classification confirming the suitability of SH fingerprint as a screening method for the authentication of Arbequina VOO. Also, the discrimination capacity of SH fingerprinting to authenticate VOOs from other cultivars was preliminary explored and promising results were obtained.

Fast and effective analytical screening tools providing new suitable authenticity markers and applicable to a large number of samples are required to efficiently control the global olive oil (OO) production, and allow the rapid detection of low levels of adulterants even with fatty acid composition similar to OO. The present study aims to develop authentication models for the comprehensive detection of illegal blends of OO with adulterants including different types of high linoleic (HL) and high oleic (HO) vegetable oils at low concentrations (2-10%) based on shotgun triacylglyceml (TAG) profile obtained by Flow Injection Analysis-Heated Electrospray Ionisation-High Resolution Mass Spectrometry (FIA-HESI-HRMS) at a large-scale experimental design. The sample set covers a large natural variability of both OO and adulterants, resulting in more than one thousand samples analysed. A combined PLS-DA binary modelling based on shotgun TAG profiling proved to be a fit for purpose screening tool in terms of efficiency and applicability. The external validation resulted in the correct classification of the 86.8% of the adulterated samples (diagnostic sensitivity = 0.87), and the 81.1% of the genuine samples (diagnostic specificity = 0.81), with an 85.1% overall correct classification (efficiency = 0.85).