Issue link: http://winesandvines.uberflip.com/i/235959
w i n e M A K IN G 1.4 4500 4000 1.2 3500 Concentration (ppb) Absorbance (465nm) 1 0.8 0.6 0.4 2500 2000 1500 1000 0.2 0 3000 500 0 0 20 40 Days 60 80 100 Figure 4: Time course of optical absorbance of model wine solution (465nm) during oak extraction (Experiment 2). ∆ = absorbance of the solution exposed to parallel-grain oak, X = absorbance of the solution exposed to cross-cut oak. rates of extraction and final concentrations for a wider variety of molecules. Materials and methods Oak samples: Tight-grain American white oak (Quercus alba) heartwood was obtained from the Staggemeyer Stave Mill (Caledonia, Minn.). The oak was quartersawn to the conventional 1.25-inch-thick barrel stave and seasoned for two years. Sample preparation: Oak staves were toasted throughout using an extended toast protocol, then cut into cubes (1.25 x 1.25 x 4.0 cm for Experiment 1 or 1.90 x 1.50 x 8.0 cm for Experiment 2). Cross-cut exposed cubes were covered with epoxy on each parallel axis (four sides), while parallel-exposed cubes were covered with epoxy on each cross-cut axis (two sides). The resultant surface area exposed was 40 cm2 for the cross-cut exposed sample and 52 cm2 for the parallel-exposed sample (Experiment 1) and 30 cm2 per block for both grain exposures (Experiment 2). The cubes were weighed before the epoxy was applied to ensure an equal mass of oak was used in each extraction. Model wine preparation and oak extraction: Model wine solutions were chosen for this study to eliminate the removal of oak flavor molecules due to biological actions and reactions with red wine pigments. Experiment 1: Two extraction flasks were prepared with 13.26 g of oak (one with all cross-cut exposed pieces, the other with all parallel-exposed pieces) in each flask and 1,326 mL of 13% ethanol in deionized water solution. (Note that 10 g/L of extracted oak flavor is more than 0 20 40 60 Time (days) 80 100 Figure 6: Concentration of gallic acid to time (Experiment 2). ∆ = concentrations in solution exposed to parallel-grain oak, X = cross-cut-exposed oak solutions. used in typical winemaking blends. Usually a tank with such a concentration is blended with unoaked wine to achieve the wine's intended style for bottling.) After each week, five 10-mL aliquots were removed from each flask and prepared for SPME-GC-MS analysis. The cross-cut exposed sample solution reached a maximum furfural concentration after two weeks; the solution was removed and replaced with a fresh extraction solution after the week-five sample was collected. Experiment 2: Six different 3-gallon (10.14 L) glass jugs were filled by mixing 1.32 L of 95% ethanol in 8.68 L of deionized water, and 70 g (7 g/L concentration) of tartaric acid (for pH control) was added to each. A 50% sodium hydroxide solution was added drop-wise to the mixture to adjust the pH to 3.3. Six cubes of cross-cut-exposed oak samples were placed in each of three vessels, while six cubes of the parallelexposed oak samples were placed in each of the other three vessels for a total mass of 89 g and surface area of 182 cm2 per container. Three 10 mL aliquots were removed from each sample on days one, three and seven, and then each consecutive week for 14 weeks. Each sample was tested for color by measuring its absorbance at 465 nm as an indicator of extracted phenolics. Quantification: In Experiment 1, furfural was quantified by SPME-GC-MS against an internal standard with comparison to a calibration curve. Vanillin was quantified in the same way using a second internal standard for comparison. In our method, the lower limit of detec- tion (LOD) of furfural and vanillin was 100 μg/L. HPLC quantitative analysis (experiment 2) was performed using the absorbance at 280 nm for furfural, 5-methylfurfural, vanillin, vanillic acid and gallic acid. Syringaldehyde was quantified using its absorbance at 310 nm. Retention times of known standards were compared to the extracts to determine identities of the peaks. External standard calibration was conducted using known standards dissolved in 13% ethanol/water solution. Results Furfural (Experiment 1) Figure 2 is a graph of the concentration of furfural in the ethanol/wine mixture over time for both parallel-exposed and crosscut exposed samples. The concentration of the solution containing the cross-cut exposed cubes reached a maximum after only two weeks of exposure (1,139±40 μg/L). The extraction solution was then removed after week 5 in order to test whether additional furfural would be extracted. The graph demonstrates that the furfural concentration did not grow to the same level as the first extraction, achieving only 245±17 μg/L after an additional five weeks of extraction. The parallel-exposed sample solution however continued to increase in furfural concentration over the entire 10-week period. Because it was not clear if the concentration had reached a maximum level, one additional set of aliquots was withdrawn from the cross-cut exposed flask after the 12th week (included on the graph). This sample showed a concentrapr actica l win ery & vin eya r d JANUARY 20 14 23