THE INFLUENCES OF WINEMAKING TECHNOLOGIES AND GRAPES WINEGROWING AREA ON VOLATILE PROFILE OF FETEASCA NEAGRA WINES

Transcription

1 THE INFLUENCES OF WINEMAKING TECHNOLOGIES AND GRAPES WINEGROWING AREA ON VOLATILE PROFILE OF FETEASCA NEAGRA WINES Ph.D. Stud. Emanuela-Filofteia Peltea University of Agronomical Sciences and Veterinary Medicine, Bucharest, Romania Abstract Feteasca neagra is a romanian red wine variety obtained from grapes originated in 4 different terroirs (, Bucuresti, Odobesti şi Valea Calugareasca). Grapes were vinified in the autumn of 2009 by classical winemaking technology, excepting fermentation which was made in the absence or in the presence of extraction s. There were obtained six technological variants, one control (to which will be reported the results obtained after analyzing all the established variants) and five adding different oenological materials (V = control, V2 = Lalvin BM 45 +, V3 = Lalvin BM 45 +, V4 = Lalvin BM 45 + Lallzyme OE +, V5 = Lalvin BM 45+ Lallzyme OE + e, V6 = Lalvin BM 45 + Lallzyme OE + ), with three repetition for each variant. For Bucharest center were obtained only four technological variants (V3 V6), because there were no samples produced without s from Bucharest grapes. The results were reported as means of all repetitions. The present study was made to evaluate the sensory profile in order to determinate the effect of, and liophylised s addition. The differences presented in volatile profile were evaluated by performing Principal Component Analysis (PCA) and Discriminant Factor Analysis (DFA) on the basis of chromatograms obtained with a dual-column gas-chromatograph working on the principle of an electronic nose based on flash-chromatography. The results showed that both the winemaking technology and terroir had a big influence on volatile profile of analyzed wines. Keywords: Feteasca neagra, red wine, winemaking technology, sensory profile. INTRODUCTION Feteasca neagra is an autochtonous variety increasingly recognized as an important representative for our red wine assortment. As such, it has witnessed a rapid spread to various viticultural regions of the country. This variety leads to wines of very good colour and a complex red fruit and plum aroma which, however, depends on vintage year and cultivation region. In order to discriminate the analyzed wines was used the Heracles analyzer (performed by Alpha MOS, France), an instrument working on the principle of an electronic nose based on flashchromatography. The data, obtained as chromatograms, are processed with the apparatus software (AlphaSoft ver..0), by statistical methods (PCA Principal Component Analysis and DFA Discriminant Factor Analysis), resulting diagrams which permit the examination of samples discrimination and identification. The method s parameters are presented in table. Table. Method parameters Parameter Value Sample volume 4 ml wine Incubation 60 o C Incubation 600s Injection volume 2500µl Sampling 20s Trap 40 o C Trap prepurge 5s Other trap Trap desorbtion 250 o C; trap parameters preheat 20s; incubation 60s Initial 40 o C (initial hold GC program 2s), final 200 o C (final hold 5s), column heating rate 5 o /s Data acquisition 40s Injector 200 o C Detector 220 o C 2. MATERIALS AND METHODS Feteasca neagra grapes were harvested in the autumn of 2009 from four different Romanian vineyards (Bucharest, Odobesti, Valea Calugareasca and ) and vinified in 6, 8

2 respectively 4 technological variants in our laboratory. The winemaking was performed in vats by classical on cap in the absence or in the presence of extraction s. One month after the fermentation was completed, and lyophilized s were added in both treated wines and non- treated wines. Each technological variant was obtained in triplicate. In brief, all the variants are presented in table 2. The oenological BM45 used for the fermentation of all samples was provided by Lallemand and is mostly recommended for red wines, in which it enhances the varietal character, with a moderate fermentation speed at a between 8-28 C. According to the producer s description BM45 produces high levels of polyphenol-reactive polysaccharides, resulting in wines with great mouth feel and improved color stability. The extraction used, Lallezyme OE from Lallemand, is a high concentration pectinase from Aspergillus niger recommended for red grapes. It also displays medium cellulase and hemicellulase secondary activities, allowing for a smooth extraction effect. The were Pronektar medium toast from American oak provided by Tonnellerie Radoux USA. Generally the American release in wines less s than French, but they confer slightly more aroma of vanilla and cocoa. The s used were both provided by Enologica Vason. The Premium is a granulated obtained from oak wood from the region of France through a special hydro-alcoholic extraction that allows the extraction of substances similar to those extracted during aging in wooden casks. This contributes to the stabilization of the colour fractions of red wines, improving the aging potential and its resistance to oxidation. The Premium is a granulated obtained from toasted French cask oak through a special extraction and desiccation process. This product is a very soft, supposed to increase the woody toasty hints tending towards cocoa. The wines were analyzed 4 months after the completion of fermentation. To differentiate the variants based on their volatile profiles a Heracles electronic nose based on dualcolumn flash gas-chromatography from Alpha Mos company, was used (Antoce and Namolosanu, 2009; Antoce et al., 2009). The analytical methods and parameter used for wines were developed in our laboratory (Antoce et al., 2008; Antoce et al., 2009 a, b). Table. The technological variants wines produced from Feteasca neagra (FN) grapes Type of Technological Dosage and Terroir variant of addition Odobesti Bucuresti FNO FNO2 FNO3 FNO4 FNO5 FNO6 FNP FNP2 FNP3 FNP4 FNP5 FNP6 FNB3 oenological product oak oak OE 250 OE 2 g/hl OE 2 g/hl OE 250 OE 2 g/hl OE 2 g/hl beginning of 9

3 Valea Calugareasca FNB4 FNB5 FNB6 FNV FNV2 FNV3 FNV4 FNV5 FNV6 oak OE 250 OE 2 g/hl OE 2 g/hl OE 250 OE 2 g/hl OE 2 g/hl in general, each type of wine (and technological treatment) from the others. In the first part of this study I tried to discriminate the samples by their wine-growing area of grapes. The first method applied was Principal Component analysis. The 54 value of discrimination index is very good, especially knowing that the analyzed wines are obtained by many different winemaking technologies. As it can be seen in figure the software has identified a first principal component (PC) which explain % from the variability of initial data and which is a linear combination of the 9 sensors (peaks) used in the analysis. The second principal component (PC2), perpendicular on the first one, explain % from the variability of initial data. Valea Călugărească Odobeşti Bucureşti Figure PCA diagram with Feteasca neagra wines disrimination by the wine-growing grapes area. PC- 68,643%; PC2-22,625%. Discrimination index: RESULTS AND DISCUSSIONS Aside of the enzymic treatment, the treatment with and soluble s was expected to influence the volatile profile as well. The electronic nose Heracles was used to analyze and compare the volatile profiles of all the wines studied. Because of the influence of the place of origin of the grapes as well of all the variations in technology (addition of, and two types of oenological s) the results vary from one case to another, but what can be said in all cases is that the apparatus was able to distinguish, The second applied method, Discriminant Analysis Function (DFA), identifies the linear combination of initial variables, combinations named discrimination functions; simplifying the initial set of data is assured a better discrimination of analyzed groups of samples. The DFA diagram corresponding to the same data and sensors which provided the PCA diagram is showen in figure 2. The four groups of wine are very good discriminated, despide that each group contain different variants. The percentages shown on each of the DFA axes indicate the amount of information provided by those axes (or Discriminant Functions), useful to discriminate of the groups of samples. 20

4 Odobeşti Bucureşti For wines from (Figure 4) and Valea Calugareasca (Figure 5) centers the variant V and V2 had a similar sensory profile, showing that the addition of had not a significant influence. By the contrary, the two types of oenological s added seems to have a significant effect on sensory profile, the variants V5 and V6 being always very good diferentiated from the others wine variants. Valea Călugărească 3 r2, Figure 2 DFA diagram with Feteasca neagra wines disrimination by the wine-growing grapes area. DF- 80,342%; DF2-5,05%. It was taken into account a number of 05 peaks for samples discrimination The second experiment regarded the discrimination by technological variant for each wine-growing grapes area. The applied method was Discriminant Analysis Function (DFA). For the wines obtained from grapes originate in Odobesti center on the diagram it can be seen that the variants with addition, V2 respectively V4, are grouped in closed area. In the same way are grouped the variants with oenological s addition, V5 respectively V6. We may say that these wines have similar sensory profile. 2 r2, r2, Figure 4 DFA diagram with disrimination of Feteasca neagra wines form center. DF-94,808%; DF2-2,968%. It was taken into account a number of 2 sensors 2 r2, 5 r2, 3 r2, 4 r2, 6 r2, 4 r2, 3 r2, 3) r2, 4 r2, 2 r2, 3) (r r2 6 r2, 5 r2, 6 (r r2 5 (r r2 Figura 5 DFA diagram with disrimination of Feteasca neagra wines form Valea Calugareasca center. DF- 99,572%; DF2-0,2956%. It was taken into account a number of 8 sensors Figure 3 DFA diagram with disrimination of Feteasca neagra wines form Odobesti center. DF-93,539%; DF2-4,73%. It was taken into account a number of 9 sensors For wines obtained from grapes originated in Bucharest we may say that variant 3 and 4 were very good discriminated on dieagram and, variants 5 and 6 with addition of the two types of oenological s are grouped in the same area on 2

5 diagram. To discriminate the samples it was taken into consideration a number of 6 peaks. 6 r2, Figura 6 DFA diagram with disrimination of Feteasca neagra wines form Bucharest center. DF-84,683%; DF2-5,53%. It was taken into account a number of 6 sensors The last part of this experiment regarded the discrimination of tested samples, grouped by the technological variant for all the wine-growing grapes areas. Figure 7 shows the obtained diagram after processing the results by Discriminant Analysis Function method. In this case too the apparatus has made a very good discrimination of the six groups of samples by the winemaking technology applied, despite that every goup contain samples originated from three different winegrowing area. The variants V5 and V6 are grouped in the same area on diagram; so we may say that the addition of oenological s had a significant influence on sensory profile of the two wines. As it can be seen on the diagram the addition of oak (variant 2) had no significant influence on sensory profile of analized wines. a 4 r2, a 5 r2, 5 r2, a 6 r2, a 3 r2, 4 r2, a r2, a 2 3 r2, r2, Figura 7. DFA diagram with disrimination of Feteasca neagra wines from all four wine-growing area. DF- 64,876%; DF2-7,646%. It was taken into account a number of 85 sensors 4. CONCLUSIONS Althought there are differences between the technological variants amoung every wine-growing area and variety the differences induced by terroir are dominants. As for the effect of on the volatile profiles of the resulting wine the results obtained with the electronic nose have shown that a good discrimination is possible among the technological variants of Feteasca neagra variety. This means that the addition, coupled or not with oenological s addition, had a significant effect of the Feteasca neagra wines volatile profiles. The experiences and analysis made with electronic nose had shown that the apparatus is able, generaly, to distinguish the wine samples by the applied treatments the discrimination degree being very good. An exception represent variant V2, indicating that the type of used for this study had no significat effect on sensory profile of wine. By contrary, the two types of oenological s added seems to have a significant influence on sensory profile, the variants V5 respectively V6 being always very good diferentiated by the other variants. 5. REFERENCES [] Bartlett, P.N., Elliott, J.M. and Gardner, Electronic noses and their application in the food industry, J.W. Food Technology, Vol 5 p44. [2] Bertrand A., 98 - Influence de la clarification de mout sur la production des aromes, Actualitees Oenologioues et Viticoles, , Dunod, Paris [3] Bertrand, A., Beloqui, M.A.A., Les enzimes a activite secondaire appliquees a l expression aromatique des cepages, Bordeaux Cite mondiale. [4] Bertrand, A., Mesure de la couleur, F.V. Nº 04, O.I.V., Paris. [5] Darriet, P., Etat de nos connaissances sur l arome et les precurseurs d arome du Sauvignon, Lallemand vol. 4 Les aromes du vin. Caracterisation et genese., Toulouse. [6] J. Bakker and G. M. Arnold, Analysis of Sensory and Chemical Data for Color Evaluation of a Range of Red Port Wines, Am. J. Enol. Vitic. 44::27-34 [7] James A. Kennedy, Jordan Ferrier, James F. Harbertson and Catherine Peyrot des Gachons, Analysis of Tannins in Red Wine Using Multiple Methods: Correlation with Perceived Astringency, Am. J. Enol. Vitic. 57:4: [8] Soares, D., Spectrocolorimetry in wines, F.V. Nº 07, O.I.V., Paris. [9] Zambonelli C., Tini V., Castellari L., Guida all uso dei Lieviti selezionati in oenologia, Calderini edagricole, Bologna *** Catalog produse oenologice Lallemand 22