Changes in Non-Volatile Compounds and Extracts of Wines Due to Yeast Species and Fermentation Temperature

Transcription

1 Changes in Non-Volatile Compounds and Extracts of Wines Due to Species and Fermentation C. S. Ough and C. L. Winger Department of Viticulture & Enology, Uniersity of California, Dais, CA U.S.A. Receied April 1982 The real reducing sugar-free extracts and the calculated extracts of wines ary significantly with fermentation temperature within the range of l2-2l C. Releant compositional changes are shown and discussed. The three yeasts used did not cause any significant difference in the calculated extracts. The conclusions were that any critical assessment of wine or juice amelioration based only on these forumulas would be subject to criticism. In recent years the true nature of some California wines has been questioned by certain European countries. They hae suggested that because the calculated extracts did not meet a certain set of criteria, water had been added to either the grapes or the wine. Most of the alues used to establish these criteria were obtained from European grapes, which were probably ameliorated with sugar prior to fermentation and, in addition, fermented with a high leel of insoluble solids and at warmer temperatures than now used in California wineries. The calculated extract is obtained by subtracting from the sugar-free extract measurements all th~ compounds which are ariable. The list of compounds used and some of the formulas commonly applied, are gien in Table 1. These, in fact, are not all the compounds which can affect the extract and that are ariable. Other compounds include nitrogen compounds which can ary in amount depending on the fermentation conditions (Ough et al., 1969). The phenol content can also ary depending on fermentation conditions such as temperature and skin contact (Ough et al., 1969). TABLE 1 Calculated Extract Formulas Used. Rebelein's short form calculated extract (Rebelein, 1971): CE=.92NE-.9T-.5E Gilbert's long form calculated extract (Gilbert, 1976): CE= NE-1.18TA- M-.61L- C-.95S-.37V-.61G Gilbert's calculated extract (Anon., 1977): CE= NE-1.18TA-M-.61L-C-.37V-.6E Breitbach's calculated extract (Breitbach, 1978): CE= NE-1.18TA- M-.61L-C-.37V-.1E-.61G Where: CE = calculated extract NE = non-reducing extract L = lactic acid V = olatile acid T = titratable acid (as tartaric) E =ethanol M = malic acid G = glycerol and 2,3-butanediol TA = tartaric acid C = citric acid S = succinic acid and all are reported in g/ C. The purpose of this experiment was to ferment a grape juice of known purity and ariety in order to determine if yeast species or temperature of fermentation affected the composition and the calculation of the extracts. Preiously, Hannemann and Radler (198) indicated that different species or strains of yeast caused ery significant changes in the calculated residual extracts. MATERIALS AND METHODS Grapes and juice: Chenin blanc grapes were picked at mid-season from the Oakille ineyard of the Uniersity of California. They were at 22.5 Brix, 8.5 g/ total acidity (as tartaric acid) with a ph of 3.35 and were crushed and pressed in a hydraulic basket press. The juice obtained would be equialent to commercial free run juice. Approximately 6 litres of juice were collected. Sixty mg/ of sulphur dioxide was added and mixed in. The juice was settled at 15 C oernight with added pectinase, and racked the following morning into three containers, each holding 15 litres. : Dry commercial yeast was used to inoculate the juices. The source of yeast was fresh samples from Uniersal Co. The three yeasts used were Montrachet (Saccharomyces cereisiae), California Champagne (Saccharomyces bayanus) and Flor (Saccharomyces fermentati). These yeasts were rehydrated in 4 C water and then added to the respectie containers. Each container receied sufficient yeast culture to bring the iable cell count to about 1 6 cells/me. Two hundred mg/ of diatomaceous earth was also added. : Three fermentation temperatures were selected for the experiment: l2 C, l6 C, and 21 C. This ariation is not great, but coers the general range now used in California to ferment white juice. From preious work (Ough and Amerine, 1966, 1967 and Ough et al., 1969, 1972), it has been well documented that fermentation temperature affects the wine composition oer these ranges. Experimental design: The contents of the filled containers, with the indiidual yeast suspended in them, were each diided into 15 containers. Fie of each were placed into each of the three temperature controlled rooms. Each of the fie replicate samples was tested for Brix changes during fermentation. When the fermentations were complete (or as nearly so as they would go), they were racked seeral times and placed into a -2 C room and allowed to stabilize for about two months, then carefully filtered into full containers and analysed. S. Afr. J. Eno!. Vitic., Vol. 3. No. I

2 18 Changes in non-olatile compounds and extracts of wines Analytical methods: The methods mainly used were those of Amerine and Ough (198). The dried extract was done by drying under acuum at 7 C until constant weight was achieed. The indiidual organic acids, except malic, were measured by the HPLC system 1.3 described by Heymann (198). The malic acid was measured enzymatically because of interferences in the HPLC system. RESULTS AND DISCUSSION The acid components are presented in Table 2, along with their statistical significance. The three yeasts caused significantly different results with all the acid components measured except citric and tartaric acids. Howeer, the Champagne yeast replicates at 21 C fermentation temperature had started to undergo malolactic fermentation prior to getting them into the -2 C room. If the 21 C Champagne samples are excluded from the results then the malic and lactic acid alues are not significantly different. The same also holds for the ph and titratable acidity data. It seems that the two main factors which these yeasts affect are succinic and acetic acids. Montrachet definitely produced more succinic and less acetic acids in these tests than did the other two yeasts. Looking at the temperature effects, the ones that were not biased by the partial malolactic fermentation of the Champagne at 21 C were citric, tartaric, succinic and acetic acids. There was a decline in the citric and tartaric acids with increasing fermentation temperature and an increase in succinic acid. The acetic acid leels were not significantly different. The titratable acidity, ph, malic and lactic acids were all affected by the partial malolactic fermentation of the Champagne yeast samples at 21 C. Both with the Montrachet and Flor samples and the 12 C and l6 C samples, there were no significant differences other than those already identified - that is, succinic and acetic for the yeast, tartrate and succinic for temperature effects. The interaction data of yeast with temperature are shown in Figure 1. The Montrachet's succinic production is more enhanced by increase in temperature than that of the other two yeasts. '--- CJ) "'ti <( c 'i:i :;) "' 1.15 LO LSDcc.139 "'** /:y O~A.8 5 / A.1-..., I 8 21 Fermentation Temp c F1G. l Succinic acid production by Montrachet ( ), Flor ( ), and Champagne (.6) yeasts at three fermentation temperatures. Some of the major fermentation products are gien in Table 3 along with the indicated significant differences in the means. The ethanol alues are not significantly different if they are adjusted for the residual sugar alues. The Champagne yeast did not quite ferment to dryness, and if the equialent alcohol yields are considered, the differences are within experimental ariation. Montrachet and Champagne produced less glycerol than the Flor yeast, and Montrachet produced only half as much 2,3-butanediols as the other two. The effects of the residual sugar in the Champagne yeast samples were spread oer all the replicates fairly TABLE2 Acid Components in the Wines. Total acidity Citric Tartaric Treatment g tartaric acid/ ph acid g/e acid g/e Montrachet 1.48, 3.26,.48, 2.lL Champagne 9.97b 3.34b.45, 2.17, Flor 1.83, 3.28,.46, 2.9, LSDd.36***.33*** 12 C 1.59, 3.28,.48, C 1.65, 3.29a.b.47, 2.9h 21 C 1.8h 3.32b.44b 2.7b LSDd.36***.24**.29*.16** ' b.' Those alues with different subscripts are significantly different. d LSD means least significant difference between mean alues at the leel indicated. *, **, *** Significant at 5, 1and.1 % leels. Malic acid g!e 6.56, 5.12b 6.42,.836*** 6.56, 6.48, 5.6h.836*** Succinic Lactic Acetic acid acid g/ acid gle g!e 1.1 L.28,.26,.86b.7b.54b.94b.25,.46b.8***.26***.117***.8,.29,.42,.94b.29,.4L l.17c.65h.43,.8***.26*** S. Afr. J. Enol. Vitic., Vol. 3. No

3 eenly. Therefore, the ethanol decreases with temperature increases could be accepted. The glycerol (Ough et al., 1972) and 2,3-butanediols (Ough and Amerine, 1967) also behae as expected, increasing with increased fermentation temperature. Howeer, the 2,3- butanediols, as indicated in Figure 2, were produced less extensiely by the Montrachet yeast, and the amounts were less affected by temperature. Changes in non-olatile compounds and extracts of wines TABLE3 Fermentation Products in the Wines. LSD= OJ 14Hlf Butanediol Ethanol Glycerol g!c Treatment %/ g!c Leo Me so Montrachet 12.53, 5.58,.28,.o7, Champagne 12.34b 5.58,.58b.14, Flor 12.44,.b 6.3h.54b O.lh LSDd.15**.51***.51 ***.25*** IJ) "... w z < I- ::>!f ~.7 ~o l2 C 12.56, 4.9,.34,.8, 16 C 12.45,_b 5.5b.44b.1, 21 C 12.29b 7.4,.61,.14b LSDd.26***.51***.51 ***.25*** ' b,' Those alues with different subscripts are significantly different. d LSD means least significant differences between mean alues at the leels indicated. *, **, *** Significance at 5, 1 and. 1 % leels. FERMENTATION TEMP. c F1G.2 2,3-Butanediol production by Montrachet( ), Flor (), and Champagne (L'l.) yeasts at three fermentation temperatures. Other analyses concerned with the extract are gien in Table 4. The reducing sugar is significantly different for the Champagne yeast as this yeast failed to ferment to dryness. This occurred at all temperatures. The potassium alues were greatest for Montrachet and the least for the Flor samples. Proline was used to the greatest extent by the Montrachet, about 8 mglf less remaining than for the other two yeasts. Champagne yeast absorbed less phenols from the juice than did the other two yeasts. No significant differences were noted for the ash or alkalinity of the ash. The effects of tern- perature were noted in only two instances. The potassium was lower for the l2 C samples and the proline lower for the 21 C samples. There were some significant interactions in these comparisons. The Flor yeast samples showed a definite effect of temperature on the potassium absorption where the others did not (Figure 3). Proline data graphed (Figure 4), for temperature effects on the indiidual yeasts indicate the ability for Montrachet yeast to assimilate more proline than the other two yeasts. TABLE4 Other Residual Compounds in the Wines. Reducing sugars Potassium Pro line Phenols Treatment g!c g/c g!c g!c Montrachet 2.19,.749,.46,.219, Champagne 4.3h.72h.487b.232h Flor 2.21,.622,.485b.225, LSDd 1.37***.24***.15***.6*** l2 C 3.33,.636,.47L.225, 16 C 2.6L.716h.467,.224, 21 C 2.5,.722h.44h.225, LSDd.24***.15*** Ash' g1e Alkalinity' of the ash meq/f 1.493, 17.9, 1.438, 18.3, l.34l 17.2, , 16.7., 1.45, 18.L , 18.6, ' b,' Those alues with different subscripts are significantly different. d LSD means least significant difference between mean alues at the leel indicated. ' These sets of alues calculated on pooled replicated samples for each yeast at each temperature. *, **, *** Significance at 5, 1and.1 % leels. S. Afr. J. Eno!. Vitic., Vol. 3. No

4 Changes in non-olatile compounds and extracts of wines.8 IL5.7 J1 A=----e.45 -:::::: Cl) :== =- -c:i ~ :::>..,....,.. <t 5 a L s D D.4 i"" 1 ' OI.4 w z _, IX a.. LSD=IL27.,." ,.,., FERMENTATION TEMP. c F1G.3 Change in potassium content during fermentation at three fermentation temperatures by Montrachet( ), Flor (), and Champagne (L.) yeasts B 11 FERMENTATION TEMP c F1G.4 Change in praline content during fermentation at three fermentation temperatures by Montrachet (), Flor (L.) and Champagne( ) yeasts. Table 5 gies the extract data, real and calculated by four formulas. The weighed extract was done on composite pooled samples of the replications of each yeast at each temperature. The oerall alue for the driedweighed extract was 2.23 g/ and the aerage alue for the hydrometer extract was g/. This is close to aerage alues reported preiously (Ough et al., 1969). With the dried-weighed extract, it is difficult to get consistent results. The calculations, therefore, were done using the hydrometer extract. The non-reducing extract showed a significantly lower alue for the Champagne yeast. None of the calculated extracts was significantly different for the arious yeasts. The temperature caused a significant increase in the non-reducing extract at 21 C. This was also seen in three of the four calculated extracts. All the calculated extracts and the real non-reducing sugar extracts are lower at the lower temperatures. This certainly would be expected for the calculated extracts, mainly because of the indicated loss in ethanol and titratable acidity with increasing temperature. TABLES Extracts and Calculated Extracts. Weighed Hydrometer Non-reducing" extract extract extract Rebelein Treatment git g1e g!e short form (11) s Montrachet 19.57, 21.55, 19.36,,,h 3.4L Champagne 2.6L 22.49b 18.45h 3.lL Flor 2.5, 21.89,, 19.68, 3.85, Calculated extracts g/ e Gilbert Gilbert long form (4) equation (2) 4.89, 3.6,, 5.5,, 3.84,, 4.85, 4.8, Breitbach's Equation (3) 4.95,, 4.89,, 4.76,, LSD'.72** 1.4*** 12 C 19.83, 22.9, , 2.75, l6 C 2.15, 21.75, 19.13,, 3.7, 21 C 2.7, 22.1, 19.6h 4.8b LSD'.55*.9*** ' b Those alues with different subscripts are significantly different at the indicated leel. ' LSD means least significant leel. *, **, *** Significance at the 5,1and.1 % leels. ct Non-reducing extract is the hydrometer extract minus the reducing sugar. ( ) Refer to Literature Cited numbers. 4.77, 2.8,, 4.78,, 3.46, 5.24, 5.25h 1.11*** 4.54,, 4.68,, 5.73h.64* S. Afr. J. Eno!. Vitic., Vol. 3. No

5 Changes in non-olatile compounds and extracts of wines 21 Some of the residual compounds not measured in 3. Breitbach, K Restextrakt im Gutachterstreit. these analyses are sulfate, phosphate, chloride and Weinwirtschaft. 114: millerals other than potassium and nitrogen com- 4. Gilbert, E Uberlegungen zur Berechnung und pounds, such as protein, amino acids (other than pro- Beurteilung des Restextraktgehaltes bei Wein.. Weinline ), peptides and others as well as the non-reducing 5. ;i;~~;%~n~~ 2 ~~~~ 7 F. Radler Uber den Einfluss sugars and other carbohydrate residuals. It has been erschiedener Hefestamme auf den Extraktstoffgehalt im shown (Ough et al., 1969) that phosphate and total ni- Wein. Deutsche Lebensmittel-Rundschau. 76: trogen decrease with increasing temperature in the 6. Heymann, H A Study of Succinic Acid Production range used. in Wine. Master of Science Thesis. Uniersity of Califor- It is apparent that the use of these formulas in order nia, Dais. to obtain calculated extract alues, cannot be justified. 7. Ough, C. S., and M.A. Amerine Effects oftem- The earlier expose of Hannemann and Radler (198) perature on Winemaking. Uni. of California Exptl. Staand the results obtained in this study indicate that the tion. Bull. 827: 26p. yeast and temperature effects on the juice composition 8. Ough, C. S., and M. A. Amerine Studies With during fermentation are such that any dependence on Controlled Fermentation. X. Effect of Some Fermenta the calculated extracts for any significant purpose is highly questionable. LITERATURE CITED 1. Amerine, M. A., and C. S. Ough Methods for Analysis of Musts and Wines. John Wiley and Sons, New York, NY. 2. Anonymous Restextrakt als Beurteilungskriterium. Weinwirtschaft. 113: tion on Some Volatile Compounds in Wine. Am. J. Eno!. Vitic. 18: Ough, C. S., M. A. Amerine, and T. C. Sparks Studies With Controlled Fermentations. XI. Fermentation Effects on Acidity and ph. Am. J. Eno!. Vitic. 2: Ough, C. S., D. Fong, and M. A. Amerine Glycerol in Wine: Determination and some Factors Affecting. Am. J. Eno!. Vitic. 23: Rebelein, H Qualitatseinstifung on Weinen, Alig. Deut. Weinfachztg. 17: , S. Afr. J. Enol. Vitic., Vol. 3. No