This aim of this work was to identify the odorant compounds responsible for the typical sensory descriptors attributed to freshly distilled Cognac spirits, not matured in barrels. Panelists were first selected and trained for gas chromatography-olfactometry. Among the 150 volatile compounds identified by gas chromatography-mass spectrometry analysis, only 34 are mainly responsible for the odors detected in the spirits. The "butter" descriptor is explained by the presence of diacetyl, the "hay" descriptor by nerolidol, the "grass" descriptor mainly by Z-3-hexen-1-ol, but also by other compounds, the "pear" and "banana" descriptors by 2- and 3-methylbutyl acetates, the "rose" descriptor by 2-phenylethyl acetate, and the "lime tree" descriptor by linalool. This study demonstrated that many odorant molecules are already present in freshly distilled Cognac, thereby giving the spirit its specific aroma.

Eight samples of freshly distilled Calvados were extracted using pentane. Gas chromatography with either a mass spectrometer or flame ionization detector was used to determine the volatile compounds composition of the extracts. More than 120 molecules were identified in Calvados and then correlated with results obtained by olfactometric analysis in our earlier work [Guichard, H.; Lemesle, S.; Ledauphin, J.; Barillier, D.; Picoche, B. Chemical and Sensorial Aroma Characterization of Freshly Distilled Calvados. 1. Evaluation of Quality and Defects on the Basis of Key Odorants by Olfactometry and Sensory Analysis. J. Agric. Food Chem. 2002, 50, 424-432 (preceding paper in this issue)]. Of these, 16 of the 19 molecules that constitute the "aroma skeleton" were identified, including 5 esters, 2 ketones, 5 phenolic derivatives, 2 alcohols, and 2 carboxylic acids. Numerous compounds were also associated with odors found in part 1. These molecules can be considered as being responsible for the good quality of Calvados or, in contrast, for defects. Relative levels of some major olfactive compounds were also estimated and tentatively compared with olfactometric indices found in part 1. A good correlation was found in many cases. Two important markers of defects in Calvados were also identified. 3-Methylbut-2-en-1-ol leads to an "herbaceous" defect, and 1,1,3-triethoxypropane seems to give an "acrolein" defect in the product. "Floral" notes of the aroma of freshly distilled Calvados seem to be due to the presence of phenolic derivatives such as 2-phenylethanol and 2-phenylethyl acetate. Low-molecular-weight esters such as ethyl 2-methylpropanoate, ethyl 2-methylbutanoate, and 3-methylbutyl acetate give, in general, the "fruity" notes. However, the overall aroma of Calvados seems likely to be a subtle balance of various functionalized compounds.

No abstract

Gas chromatography coupled with mass spectrometry (GC-MS) using both electron impact and chemical ionization detection modes led to the determination of the volatile composition of two samples of freshly distilled Cognac and two samples of freshly distilled Calvados. A total of 169 volatile compounds were directly identified in dichloromethane extracts obtained by liquid-liquid extraction. Trace compounds present in both spirits were characterized with the help of preparative separations. In a first step, groups of compounds were separated by preparative GC, and the fractions were analyzed on a polar stationary phase by GC-MS. In a second step, silica gel fractionation was used to separate them by polarity. In this study, 331 compounds, of which 162 can be considered as trace compounds, were characterized in both freshly distilled Cognac and Calvados. Of these, 39 are common to both spirits; 30 are specific to Cognac with numerous hexenyl esters and norisoprenoidic derivatives, whereas 93 are specific to Calvados with compounds such as unsaturated alcohols, phenolic derivatives, and unsaturated aldehydes.

Eight freshly distilled samples of Calvados, a fermented and distilled apple juice, were analyzed by sensory evaluation and direct injection GC to determine the composition of higher alcohols, esters, and aldehydes. The composition determined by direct injection was tentatively related to sensory descriptors. Esters have a probable maximum level around 500 g/hl of pure alcohol (PA). This level also corresponds to the threshold of the main ester constituent, ethyl acetate. A high ratio of esters to ethyl acetate seems to be of prime importance for good quality. Total aldehydes, with a maximum level between 8 and 11 g/hl of PA and mainly comprising acetal (maximum between 5 and 9 g/hl of PA), were related to a "green" descriptor. Higher alcohols do not have a direct impact on quality, but other volatile compounds with a positive impact on flavor should probably be present at a high level. As overall quality was not well related to sensory quality, it was necessary to perform more a precise analysis to determine the key odorants. The Calvados samples were thus extracted using pentane. Gas chromatography, employing both a flame ionization detector and an olfactometry port, was used to analyze the obtained extracts. Seventy-one odors were detected and distributed according to Calvados quality determined by sensory evaluation. Nineteen odors common to all Calvados samples constituted the "skeleton" of the aroma. Twenty-eight odors were specific to a quality class: 6 for good quality, 4 for neutral, and 18 for defective. Twenty-four other odors had either too low an odor impact or no evident specificity.

There is no universal method that can be applied to extract bound extracellular polymeric substances (EPS) from benthic diatoms of intertidal sediments without causing cell lysis. Six extraction methods were tested on a diatom culture of Navicula jeffreyi to establish the best compromise between high yields of carbohydrate extraction and minimum cell lysis. Extraction with distilled water provoked cell lysis (as already known). The five other extraction methods (dowex resin, artificial seawater of half salinity and extractions after pretreatment with gluteraldehyde by three methods: water, dowex water and dowex buffer) did not provoke cell lysis as shown by transmission electronic microscopy. This result was confirmed by the minimum release of internal compounds (protein, ATP) and by the low proportions of glucose in dowex-extracted EPS compared with the water-extracted EPS, from which the high glucose content must be inferred as contamination by the chrysolaminaran. The extraction with dowex resin resulted in the second-highest concentration of carbohydrate after the water extraction and the EPS were especially rich in deoxy sugars, hence increasing the hydrophobic feature of these substances. For these reasons, we recommend extraction with dowex, which is also the best method for extracting bound EPS from other biofilms such as in activated sludges.

A total of 207 volatile compounds were identified in extracts of four French labeled brandies: Armagnac, Cognac, Calvados, and Mirabelle. Relative levels of all components were determined using GC-MS after integration of a selected peak of the mass spectrum of each. Each type of brandy could be clearly discriminated using PLS-DA statistical analyses based on these levels. French Mirabelle spirit, which was studied for the first time, was characterized by higher levels of many aldehydes and acetals and by the presence of compounds having an odd number of carbons together with benzaldehyde and some of its derivatives. Many possible derivatives of acrolein and high amounts of butan-2-ol were rather specific for the volatile composition of Calvados. The most important difference between the two wine-based samples seemed to be directly linked to the distillation system used. Many furanic compounds are specific to Cognac, whereas two or three compounds such as 1-(ethoxyethoxy)-2-methylbutane and gamma-eudesmol were specific to Armagnac. These two brandies presented rather high distributions of isobutanol and isopentanols, whereas Mirabelle and Calvados compositions offer more concentrated aliphatic linear alcohols.

a b s t r a c tUnderstanding the mechanisms involved in the perception of food aroma is one of the major objectives of flavour studies. Yet, it remains difficult to explain this perception due to the diversity of cross-modal interactions that occur between aroma, taste and texture during food consumption. Various sensory and instrumental methods have been developed to describe such interactions and to highlight their origins. Each of them has its own advantages and drawbacks. This paper describes the methods used over the past decade in order to help researchers choose the right approach to study aroma-related interactions. Their objectives, weaknesses and strengths are reported and contrasted. Their ability to reveal interactions in model and real food matrices is also discussed. Finally, innovative original approaches are presented.