Qualitative and semi-relative quantitative changes in flavor profiles associated with the storage of raw sardine (Sardina pilchardus) were investigated. A sensory panel generated a list of 20 odorant descriptors of raw sardine. Forty-seven volatile components were identified by gas chromatography-mass spectrometry and were quantified by gas chromatography-flame ionization detector (GC-FID). Among them, 34 were highlighted as potent odorants using an olfactometric method. (E,E)-2,4-octadienal, E-2-penten-1-ol and 2,3-butanedione are the most potent odorants of raw sardine. The odor-active compounds responsible for oxidized flavors increased during storage, whereas sulfur-containing compounds associated with marine odors decreased. These results could be related to the increase in rancidity aroma and the decrease in marine/iodized aroma identified by the sensory panelists in stored raw fish.

Volatile compounds of cooked fillets of Silurus glanis reared under two conditions occurring in France were studied. They were extracted by dynamic headspace, identified by gas chromatography/mass spectrometry, and quantified by gas chromatography-flame ionization detection. Odor active volatile compounds were characterized by gas chromatography-olfactometry. Sixty volatile compounds were detected in dynamic headspace extracts, among which 33 were odor active. Rearing conditions affected their estimated concentrations and their odor intensities, but very few qualitative differences were exhibited (only seven volatile compounds were concerned). A good correlation between quantitative and olfactometric results is shown. 2-Methylisoborneol and (E)-2-hexenal were less represented in OUTDOOR extracts, while 2-butanone was less represented in INDOOR extracts. In addition, olfactometric results can be closely related to those previously obtained by sensory analysis. Boiled potato sensory odor of the silurus cooked fillets can be related to (Z)-4-heptenal and methional, and buttery odor can be related to 2,3-butanedione, an unknown compound (RI = 1010), and 2,3-pentadione.

The colour of raw fillets and the texture of cooked fillets of European catfish (Silurus glanis) were studied. The catfish fillets resulted from two types of farming conditions used in France. Sensory and instrumental analyses were performed on the fillets. Two farming parameters were found to affect the colour and texture of European catfish fillets: water temperature and farming time. The shorter the farming time was, the lighter the fillets were, while a higher water temperature produced yellower and greener fillets. A higher water temperature and a longer farming time produced softer cooked fillets, while the higher the water temperature was, the less juicy the cooked fillets were. A good correlation and complementarity were observed between sensory and instrumental analyses of the colour and texture of European catfish fillets.

Over the past few years, the use of essential oils in breeding has been gaining ground because of their wide range of application, especially in terms of antimicrobial activity, and the current emphasis on the use of natural compounds, particularly to replace antibiotics. Among the essential oils and their active components, 4 molecules are of great interest and are studied in this paper: carvacrol, thymol, cinnamaldehyde, and diallyl disulfide. In particular, we examined the transfer of essential oil components from animal feed to milk. Because these components could confer an undesirable odor or taste to milk and derived products or interfere with milk processing, it is necessary to quantify their residues in milk. Milk is a complex biological matrix containing large amounts of water, fat, and protein. As a consequence, the extraction of volatile compounds, which are mainly lipophilic, is often critical. In this work, a method based on solid-phase microextraction coupled with gas chromatography-mass spectrometry was developed and optimized. We evaluated the effect of several parameters on the sensitivity of the method: fiber type, fiber to sample contact, extraction temperature, extraction time, agitation, and salt addition. The best results were obtained using a divinylbenzene-carboxen-polydimethylsiloxane (50/30 µm) fiber in headspace mode at 34.6°C for 32.6 min without agitation or salt addition. Standard calibration curves were constructed, and the limits of detection and quantification and ranges of linear response and repeatability were evaluated. Detection limits ranged from 25 to 50 µg/L, quantification limits from 50 to 75 µg/L, and variation coefficients from 23.5 to 36.6%. Therefore, the optimized methodology was able to evaluate whether significant transfer of essential oil components from feed to milk occurred, and the method is simple and solvent-free. We applied the method to milk samples from cows whose feed included thymol, carvacrol, cinnamaldehyde, or diallyl disulfide as feed additives at the recommended dose and at twice the usual dose. None of the 4 essential oil components were found in milk samples, regardless of the dose administered.