Abstract
A new approach to red wine aging was examined that uses a current passed continuously through glassy carbon electrodes in a process of electrochemical microoxidation. A Cabernet Sauvignon wine in 300-L tanks was subject to treatments run in triplicate over 12 weeks at 16°C consisting of a control, oxygenation to 0.67 mg/L twice weekly (equivalent to 4 mL O2 per liter wine per month), and electrochemical oxidation of 6144 μA using 250 x 6-mm glassy carbon rods. The potential at the electrode was observed to exceed 2 V, where both oxidation of wine polyphenols and direct oxidation of ethanol can occur. While there was considerable loss of bound SO2 in the oxygenated wines, the concentration of bound SO2 in the electrochemical microoxidation wines actually increased as the trial proceeded, consistent with the rise in acetaldehyde observed in these wines during the final weeks of the trial. There was little change in concentrations of individual polyphenols, although the rate of decline in monomeric anthocyanins and free quercetin was greater in the oxygenated and electrochemical microoxidated wines than in the controls. Spectrophotometric measures showed an increase in SO2-resistant pigments and a larger modified hue value (420/520 nm) in both oxygenated and electrochemical microoxidated wines. Electrochemical microoxidation provides an alternative technology for aging red wines with precise control of the rate of oxidation and treatment cut-off.
Oxygen is currently introduced into red wines using a variety of means as part of its maturation following fermentation and before bottling. It occurs in the barrel as air permeates the oak staves (Vivas et al. 2003) or is introduced when the bung is removed, in large doses during racking operations or slowly using the modern technique of microoxygenation (Moutounet et al. 1996). Various effects of oxygen have been identified, largely involving the polyphenol content of the wine, including changes in wine astringency through modification of tannin size and structure, lowering of sulfidic and vegetative odors, and stabilization of color through linking of anthocyanins with wine flavanols (Castellari et al. 2000, Atanasova et al. 2002, McCord 2003, Llaudy et al. 2006, Cano-López et al. 2006).
A major compound involved in these processes is acetaldehyde, itself thought to be formed from oxidation of ethanol by peroxide or hydroxyl radicals released during the initial oxidation of catechol-containing polyphenols and in the presence of metal catalysts (Singleton 2000, Danilewicz 2003, Waterhouse and Laurie 2006). Acetaldehyde derived from ethanol oxidation, in addition to any already present following yeast fermentation, can then link anthocyanins with flavanols to form unstable (purple) ethyl-linked pigments and also more stable (orange) pyranoanthocyanin pigments (Fulcrand et al. 2004, Eglinton et al. 2004, Alcalde-Eon et al. 2006).
An alternative to molecular oxidation, and one that is used in a wide range of industrial processes, is to pass a current through an electrode. Industrial applications of electrolysis include large-scale production of chlorine, sodium hydroxide, and various metals such as aluminum and copper. A process for wine oxidation in titanium tanks, controlled by means of the passage of small currents, has been patented (Guglielmi and Simoncelli 2002). In this patent, the treatment of 50-L wine samples is described at two current levels in titanium tanks over 60 days, with the observation that wines treated with microcurrents had less free anthocyanin but greater color intensity, more anthocyanins present in more stable structures, and decreased astringency and bitterness as evaluated by wine experts.
In previous research into development of analytical methods to characterize wine polyphenols, we showed that at a glassy carbon electrode, the compounds most easily oxidized are polyphenols, particularly those containing catechol or pyrogallol groups. These groups oxidize under wine conditions at ~400 mV (versus a Ag/ AgCl reference electrode), while oxidation of ethanol at the carbon surface will occur once the electrode potential exceeds ~1000 mV (Kilmartin et al. 2001, 2002). At the same time, formation of hydrogen peroxide or other reactive oxygen species, as occurs during reduction of molecular oxygen, is not expected with the electrochemical approach. The potential advantages of electrochemical microoxidation include precise control of the applied potential or current to target specific groups of oxidizable compounds in wine, a precise cutoff when the process needs to be terminated, and the absence of air bubbles in the wine which could potentially remove volatiles from the wine should the bubbles be allowed to reach the wine surface (although not often an issue with larger tanks).
In this study, the performance of glassy carbon rods in the microoxidation of Cabernet Sauvignon wine in 300-L stainless-steel tanks over a period of 12 weeks was examined. In addition to a control wine receiving no oxidation, wines were subject to oxygenation at a rate of 4 mL O2 per liter of wine per month, and electrochemical microoxidation using a current of 6144 μA passed continuously through glassy carbon rods (an equivalent level of oxidation to 3.1 mL of O2 per liter of wine per month), with all treatments run in triplicate. This electrolytic process is commonly referred to as galvanostatic or intentiostatic electrolysis and is easier to set up than a controlled potential (potentiostatic) process which requires inclusion of an additional reference electrode. The development of the wines was monitored for changes in polyphenol concentration, sulfur dioxide, color, and buildup of acetaldehyde.
Materials and Methods
Electrode characterization.
250-mm lengths of 6-mm-diam glassy carbon rods (GC-20S, Tokai Carbon Co., Tokyo, Japan), freshly cleaned by abrasion with moistened emery paper, were placed in contact with a red wine in a 15-L enclosed 316 stainless-steel tank, which was continuously stirred using a magnetic stirrer. Currents of 38, 77, 1536, 3072, and 6144 μA were applied using an EG&G model 363 Potentiostat (Oakridge, TN). The change in potential versus an Ag/AgCl reference electrode (+207 mV vs. she), in conjunction with a stainless-steel counterelectrode, was monitored over a 24-hr period.
Wine.
Cabernet Sauvignon grapes were grown at Te Mata’s Woodthorpe Vineyard situated in the Dartmoor Valley of Hawkes Bay, New Zealand, using vertical shoot-positioning. The grapes were harvested on 10 Apr 2005 at 21.2 Brix, crushed, destemmed, and inoculated with Malvin 245D at Mission Estate Winery. An addition of 20 mg SO2/kg fruit was used at the crusher prior to inoculation, but no further SO2 additions were made until the start of aging. The must was transferred to a 20,000-L stainless-steel tank for alcoholic fermentation. Fermentation temperatures ranged between 19 and 26°C, requiring 12 days to reach dryness, and 200 mg/L diam-monium phosphate was added on day four. The cap was managed by daily pumpovers to prevent it from drying out and to maintain cap temperature. A postfermentation maceration of 4 days was also used. The wine was pressed off the skins on 26 Apr 2005 and allowed to settle for a few days before the clarified wine was racked off the gross solids into a stainless-steel tank for storage. Spontaneous malolactic fermentation occurred during the 8-week storage. The finished wine had 13% v/v alcohol, pH 3.75, 4.9 g/L titratable acidity (to pH 8.2), and 2.2 g/L residual sugar measured using the Rebelein method (Iland et al. 2000).
Tanks and electrochemical microoxidation.
Nine sealable 300-L stainless-steel tanks, 650 mm diam and 1000 mm high, were purchased from Marchisio S.p.A. Pieve Di Teco, Italy. The 316 stainless steel used to construct these tanks had the following elemental specifications (in addition to Fe): <0.08% C, 16 to 18% Cr, 10 to 14% Ni, 2 to 3% Mo, <2% Mn, <0.75% Si, <0.045% P, and <0.03% S.
The tanks were subsequently modified by the University of Auckland engineering department to include a sealable lid with various fittings for the electrode assembly, stirring rods, inert gas entry, and sampling ports. After the tanks were filled with wine, they were left to stand for one week before the start of treatments to allow any dissolved oxygen in the wines to dissipate. Each wine was continually agitated using an overhead variable speed stirrer attached to a food-grade stainless-steel rod with a 10- x 2-cm folded blade rotating at 100 rpm to ensure homogeneity and maintained at 16°C within a thermostatically controlled cool room.
A series of three treatments were performed in triplicate. The first was oxygenation at a rate of 4 mL/L per month. In the absence of a suitable unit to deliver the O2 continuously, oxygenation was achieved by a 150-mL pulse of food-grade O2 delivered twice weekly at a rate of 25 mL/min through a hollow stainless-steel rod attached to a sinter sparge. During this process, dissolved oxygen concentrations were monitored in the wine using a model 3650 O2 Micro Logger (Orbisphere Laboratories, Neuchâtel, Switzerland), and O2 additions were continued until the desired concentration of 0.67 mg/L was achieved (8.5 applications a month sums to 5.7 mg/L, or 4.0 mL O2 per liter wine, using an O2 density of 1.43 g/L obtained from the ideal gas law, given that one mole [32 g] O2 occupies 22.4 L at STP). This oxygenation treatment, being periodic and involving the buildup of some dissolved oxygen, differs from the modern commercial practice of continuous microoxygenation. The headspace of each tank was subsequently sparged with argon to ensure any oxygen present was displaced. For the second treatment, electrochemical microoxidation was applied using a current of 6144 μA (130 μA/cm2 at the 47.4 cm2 electrode surface) at a glassy carbon rod suspended in the wine (equivalent to 3.1 mL O2 per liter wine per month, with conversion details given below). Control treatments without oxidation were also set up. The 316 stainless-steel tank served as the counterelectrode to complete the electrochemical cell.
Samples of 250-mL were collected every 14 days over a 12-week period for chemical analysis under a positive pressure of argon gas. Analysis of SO2 was undertaken at the Mission Estate Winery in Hawkes Bay, while samples were transported to the University of Auckland under argon for polyphenol and spectrophotometric analyses within 24 hr of sampling.
Sulfur dioxide.
Free and bound sulfur dioxide concentrations were monitored every two weeks using the aspiration method (Iland et al. 2000). The detection limit of this method is about a drop (0.04 mL) of the 0.01 M NaOH solution used in the titration, equivalent to about 0.6 mg/L SO2. To maintain free SO2 above 10 mg/L and avoid microbial spoilage, 10 mg/L SO2 was added, as 10% potassium metabisulfite in water, when the wine was transferred to the nine 300-L tanks, while further additions of 15 mg/L were made at week two and 20 mg/L at week four.
Spectrophotometric analyses.
The following spectrophotometric measures were adapted from the literature (Iland et al. 2000) to compare wine color components. 2 mL wine was transferred to a test tube and 30 μL 25% (w/v) potassium metabisulfite in distilled water was added, mixed thoroughly, and allowed to stand for 45 min. The absorbance reading at 520 nm (ASO2520) provided a measure of pigments resistant to SO2 bleaching. A further 2 mL wine was transferred to a test tube and treated with 20 μL acetaldehyde. After 45 min, the absorbance was measured at 420 nm and at 520 nm, and the ratios were compared to give the modified hue (Aacet420/Aacet520).
Total phenols.
Total phenols was determined in duplicate using a scaled-down Folin-Ciocalteu method (Waterhouse 2002). Gallic acid was used as standard and results are expressed as mg gallic acid equivalents/L.
HPLC analyses.
Concentrations of individual phenolic compounds were determined by reversed-phase HPLC using a mobile phase comprising aqueous acetic acid (analytical grade, Scharlau, Spain) and acetonitrile (190 grade, Ajax Finechem, Auckland, New Zealand) at a flow rate of 0.8 mL/min (Kilmartin et al. 2002). Analyses were performed on a Phenomenex (Torrance, CA) Luna C18 column (4.6 x 205 mm, 5-μm particle size) at 25°C on an Agilent Technologies (Waldbronn, Germany) 1100-series instrument coupled to a diode array detector. Flavan-3-ols and benzoic acids were monitored at 280 nm, hydroxycinnamic acids at 320 nm, flavonols at 365 nm, and anthocyanins at 520 nm. Samples were filtered through a 0.45-μm regenerated cellulose filter before direct injection of 25 μL onto the column. The following standards were used: gallic acid, (+)-catechin, (−)-epicatechin, caffeic acid, quercetin (Sigma, St Louis, MO), and malvidin-3-glucoside (Extra-synthèse, Genay, France).
The mean degree of polymerization of the proanthocyanidin fraction was determined by thiolysis with benzyl mercaptan (Fluka Chemika, Buchs, Switzerland). The proanthocyanidin fraction was obtained according as described (Kantz and Singleton 1991). Small glass columns (5-mm diam and 150-mm long) were slowly packed ~75% full with a mixture of Sephadex LH-20 (Amersham Biosciences, Sweden) swelled in 50% analytical grade methanol (Ajax Finechem). The column was equilibrated with water containing 0.2% acetic acid before sample loading. Small 5-mL volumes of wine were then introduced at a slow loading rate of 0.1 to 0.2 mL/min. The column was developed with ~15 void volumes of 60% methanol (0.2% acetic acid) in Milli-Q water at a flow rate of 0.3 to 0.5 mL/min to elute nonpolymeric phenolics. Proanthocyanidins were then recovered by continuing the column development with 15 void volumes of 50% acetone (0.2% acetic acid) in Milli-Q water at a flow rate of 0.3 to 0.5 mL/min. Each fraction obtained was reduced to less than 5 mL using a rotary vacuum evaporator at 37°C and made up exactly to a volume of 5 mL using 50% methanol in Milli-Q water. The proanthocyanidin fractions were then hydrolyzed using a thiolytic reagent in a process adapted from previous work (Prieur et al. 1994). Equal volumes (400 μL) of the polyphenol fraction and 5% benzyl mercaptan in 0.2 M HCl in methanol were mixed. After sealing in small glass bulbs, the hydrolytic reaction was carried out at 60°C for 5 hr. Once cooled, characterization of the hydrolyzed polyphenol fraction and calculation of the mean degree of polymerization was performed similar to the published method (Rigaud et al. 1991), using aqueous acetic acid and acetonitrile at a flow rate of 1.0 mL/min and the HPLC column and detector (at 280 nm) as described above for determination of individual phenolics.
Acetaldehyde by gas chromatography.
Quantification of acetaldehyde in both free and bound forms was adapted from published work (Peinado et al. 2004). A Shimadzu (Kyoto, Japan) GC-17A series gas chromatograph (GC) was used, coupled to a flame ionization detector running at 220°C The column used was a capillary DB Wax 12–7012 column (15 m x 0.25 mm x 0.25 μm) (Alltech Associates, Deerfield, IL). The column carrier gas (N2) was delivered at 1.4 mL/min with a split ratio of 43:1 and an injector temperature of 220°C. The initial loading temperature was 35°C for 5 min, increasing by 4°C/min to 55°C, then ramped at 20°C/min to 200°C and held at 200°C for 5 min, for a total run time of 22.3 min. 50 μL of the internal standard 4-methylpentan-2-ol (98%; Aldrich), prepared as 50 g/L in 40% ethanol, was added to 5-mL samples of wine or to standard solutions of acetaldehyde (>99%, Merck Schuchardt, Hohenbrunn, Germany). After homogenization, 1.0-μL aliquots were manually injected onto the GC column. The limit of detection for the GC method was 0.2 mg/L and the limit of quantification was 0.5 mg/L.
Statistical analysis.
Mean values and standard deviations were calculated using one-way analysis of variance (ANOVA). The significance of differences between mean values obtained was determined using a student t-test at the 95% significance level, p < 0.05.
Results and Discussion
Glassy carbon electrodes.
0.77 mL O2 (VO2 = 0.77 x 10−3 L) represents 1.1 mg O2 at 0°C (based upon the density of 1.43 g/L O2 gas), or 3.44 x 10−5 mol (nO2). As four electrons are passed in the reduction of each oxygen molecule to water, this would be equivalent to 1.38 x 10−4 mol electrons (ne).
(eq 1)
The equivalent oxidation achieved by applying a current at an electrode and the number of moles of oxygen can be determined using the following conversions:
(eq 2)
Using the Faraday constant (F) of 96,485 C/mol, the charge (Q) passed here is 13.3 C (C = A × s), for 0.77 mL O2 in a liter of wine. To pass this charge through an electrode in a liter of wine over 30 days (2,592,000 s) would require a current (I) of 5.1 μA, which translates to 77 μA in 15 L wine or 1536 μA in 300 L wine.
Currents of 38 and 77 μA were first applied to 250-mm lengths of glassy carbon rod immersed in a red wine, equivalent to 0.38 and 0.77 mL O2 per month for 15 L wine (the size of smaller research tanks in use at Auckland). The potential began in the 400 to 500 mV range, and then climbed to values over 1 V (Ag/AgCl) after 40 to 120 min and reached values between 1.2 and 1.5 V after 24 hr (Figure 1⇓). At potentials close to 400 mV, a range of catechol-containing polyphenols found in red wine can be readily oxidized at the glassy carbon electrode (Kilmartin et al. 2001). However, the products of polyphenol oxidation are known to build up on the electrode surface, requiring a higher electrode potential to continue the oxidation (Kilmartin 2001), while the adsorption of organic molecules also leads to inhibition catechol group oxidation at a glassy carbon electrode (DuVall and McCreery 2000). On the other hand, the charge (0.28 C) passed during 1 hour of oxidation at 77 μA (2.87 x 10−6 mol electrons), would only be sufficient to oxidize about 0.4 mg catechin (1.44 x 10−6 mol, given that two electrons are required to oxidize each molecule of catechin [290 g/mol] [Kilmartin et al. 2001]) and thus the rise in potential will not be due to depletion of oxidizable polyphenols in the wine.
Electrical potential versus time curves for 250-mm length glassy carbon rods immersed in a red wine with various applied currents. Potentials recorded at 1-min intervals.
Higher currents were then applied as needed to produce oxidation equivalent to 0.77, 1.54, and 3.1 mL O2 per month for a 300-L wine tank (i.e., currents of 1536, 3072, and 6144 μA). In each case, the voltage exceeded 2.5 V within 5 min, the maximum range of the potentiostat (Figure 1⇑). A combination of rapid buildup of polyphenol oxidation products on the electrode surface, providing a resistive layer, and the need for the electrode to oxidize other wine components to sustain the higher currents (e.g., ethanol or water) likely led to the rise in potential. When the potential across the two terminals of the tanks during the course of the electrochemical microoxidation trial was measured using a digital multimeter (Voltcraft model, VC 555), a value of around 3 V was typically obtained.
To complete the electrochemical cell, reduction processes occurred at the 316 stainless-steel tank which served as the counterelectrode. These included reduction of protons from wine acids and the reduction of any dissolved oxygen left in the wine, processes in which equal quantities of protons and electrons were removed and which largely balanced the protons and electrons released from oxidations at the carbon electrode, so the total acidity of the wine changed very little over time. As the wine was also being continuously stirred, significant pH gradients were not set up in the wine, although in an unstirred tank these may well develop.
The glassy carbon rods used in this trial are appealing electrode materials because the purity of the glassy carbon leads to low background currents in various solvents. Glassy carbon is a costly material, however, and is difficult to produce in thicker forms. Alternative cheaper carbon electrodes could be considered in future, in which larger and more porous surface areas are in contact with the wine (e.g., vitreous carbon foams, carbon cloths, or graphite rods). That would allow higher currents to be passed, but the presence of impurities in the materials must be checked to ensure that unwanted electrochemical processes do not occur.
Sulfur dioxide and acetaldehyde.
Total SO2 can be derived from the concentrations of free and bound SO2 in the treatment wines during the final 6 weeks of the trial (Figure 2⇓). Following SO2 additions of 10 mg/L when the wine was transferred to the 300-L tanks, 15 mg/L at week 2, and 20 mg/L at week 4, a total SO2 concentration of 32 ± 4 mg/L was recorded in the control wines at week 6, of which ~70% was in the free form. A concentration of 23 mg/L free SO2 at a wine pH of 3.75 is equivalent to a molecular SO2 concentration of ~0.25 mg/L. A higher concentration of molecular SO2 and more effective protection against microorganisms would be obtained by acidifying the wine. The total SO2 was not significantly different by week 12 in the control wines, when a total concentration of 30 ± 4 mg/L was measured. By contrast, wines undergoing oxygenation were already down to a total SO2 concentration of 23 ± 2 mg/L by week 6, declining further to 11 ± 3 mg/L at week 12, with a very low bound SO2 component. The impact of wine oxidation following introduction of molecular O2 was seen in the decline of both free and bound SO2 concentrations.
Concentrations of (A) free SO2, (B) bound SO2, and (C) acetaldehyde over the final 6 weeks of a red wine trial (ELMOX: electrochemical microoxidation).
Wines subjected to electrochemical microoxidation retained a similar amount of total SO2 to the control wine (30 ± 4 mg/L at week 6, which remained at 29 ± 5 mg/ L by week 12), but in this case there was considerable buildup in the bound SO2 fraction, reaching 63% of the total by week 12 (Figure 2⇑). This increased bound fraction can be related to the increased acetaldehyde concentration in the electrochemical microoxidation wines later in the trial. While the measured acetaldehyde remained 1 mg/L in all wines up to week 8, the concentration climbed to 10 mg/L in the electrochemical microoxidation wines by week 12. Given the strong binding that occurs between acetaldehyde and free SO2, this concentration of acetaldehyde could account for 14 mg/L bound SO2. It was thus apparent that ethanol was directly oxidized to acetaldehyde in the electrochemical microoxidation wines toward the end of the trial. This acetaldehyde became bound with SO2, but without a significant loss in total SO2.
The application of 6144 μA is equivalent to 0.0385 mol electrons per week, or 0.0193 mol acetaldehyde in 300 L wine if all of the current was used in its production:
(eq 4)
This in turn equates to 2.8 mg/L acetaldehyde per week, or 34 mg/L over 12 weeks, which could potentially bind with 4.1 mg/L SO2 per week. The corresponding change in ethanol concentration over 12 weeks (~10 g in 300 L wine) would only be about 0.03% of its initial concentration. Conversely, if all of the current were applied to oxidation of a catechol-containing polyphenol such as catechin (i.e., 0.0193 mol), then up to 19 mg/L catechin per week would be consumed. If all the current had been applied to oxidation of water that forms dissolved oxygen in the wine (equation 1 in reverse), then 1.0 mg/L per week would have been produced, which would have been taken up by the red wine without a detectable increase in dissolved oxygen.
The 7 mg/L increase in bound SO2 during the final 6 weeks of the electrochemical microoxidation trail is associated with uptake of ~5 mg/L acetaldehyde, without taking into account any acetaldehyde produced during the 12-week trial which had reacted or degraded in some other manner. It thus appears that a significant fraction of the applied current at the glassy carbon electrodes went into direct oxidation of ethanol to acetaldehyde.
Polyphenol development.
In contrast to changing concentrations of SO2 and acetaldehyde, there was much less change evident in polyphenolic content. The total phenols, originally at 2794 ± 20 mg/L, ended several percent higher, but similar in all wines, by the end of the 12-week trial (Figure 3⇓). While more phenols were not added to the wine, changes in their structure as new compounds are formed can lead to a different response in the Folin-Ciocalteu assay. The size of the tannin fraction, given by the mean degree of polymerization, started at an average of 19.7 ± 0.5 and stayed within two units of this value throughout the trial, with no significant difference among the three treatments. Any change in the astringency of these wines will thus rely upon changes in tannin chemistry other than the average size of the proanthocyanidins (Vidal et al. 2002).
Changes in concentration of total phenols, mean degree of polymerization (MDP), and selected monomeric phenolics in red wines over 12 weeks.
Among the individual polyphenols monitored by HPLC, concentrations of several compounds such as caftaric acid and catechin declined by only a few percent across the 12-week trial, but with no significant difference among the treatments (Figure 3⇑). Adequate concentrations of free SO2 were likely maintained to limit oxidation of catechol-containing compounds. On the other hand, the expected decline in monomeric anthocyanins was greater in wines subject to oxidation. A greater lowering in the concentration of malvidin-3-glucoside was seen in the oxygenated wine (by 21.2 ± 1.4%) than in the electrochemical microoxidation wine (17.9 ± 1.6%), with a smaller drop observed in the control wine (by 12.5 ± 3.0%) over 12 weeks. Similar trends were seen with the glucosides of delphinidin, petunidin, and peonidin, with the mean values being statistically different from one another ( p < 0.05). A greater loss of monomeric anthocyanins in microoxygenated wines has been reported previously (Cano-López et al. 2006). In this trial, concentrations of free quercetin declined considerably during the trial, although the flavonol glycosides remained relatively steady, with a greater loss of quercetin in the oxygenated wines over the other two treatments (Figure 3⇑). Losses of quercetin have been reported in studies of red wines undergoing barrel aging (Fang et al. 2007), while quercetin is an excellent scavenger of reactive oxygen species such as H2O2 (Park et al. 2003). H2O2 is an expected by-product of the initial reaction of O2 with polyphenols, but is not expected to form during the direct electrochemical oxidation of polyphenols or ethanol, which may explain the greater loss of quercetin in the oxygenated wines during this trial.
Spectrophotometric measures.
Some development in wine chemistry with the oxidative treatments was evident through results obtained using spectrophotometric measures. Toward the end of the trial, there was an increase in the concentration of pigments resistant to sulfite bleaching, with similar increases in both treated wines above that seen in the control wines (Figure 4A⇓). These increases are related to losses in monomeric anthocyanins also observed by the end of the trial. In the electrochemical microoxidation wines, the buildup of acetaldehyde could have contributed to formation of bonds between anthocyanins and flavan-3-ols to produce pigments more resistant to SO2 bleaching. However, the similar increase in SO2-resistant pigments in oxygenated wines points to the formation of such compounds even with less acetaldehyde present.
Spectrophotometric measures of wine composition: (A) SO2 resistant pigments (ASO2520) and (B) modified hue (Aacet420/Aacet520).
The development of wine hue (ratio of 420 to 520 nm abs) was also examined. It was necessary to first treat the wines with acetaldehyde due to the variable SO2 concentrations in the treatments and its bleaching effect on anthocyanins. The control wines, with higher concentrations of free SO2 by the end of the trial, had much lower 520 nm absorbances than the oxidated wines, an effect which was removed upon treatment with acetaldehyde to bind up the SO2 present. Over the final 6 weeks of the trial, there was a steady increase in the modified hue for oxygenated and electrochemical microoxidation wines, with the highest concentrations being reached in the oxygenated wines (Figure 4B⇑). This increase in hue could be due to the formation of brown polyphenol oxidation products (Singleton 2000) and to the formation of polymeric pigments such as the pyranoanthocyanins, which absorb relatively more light at 420 nm than at 520 nm compared with the original monomeric anthocyanins. By contrast, a similar hue (or tint) was seen with microoxygenated and control wines, in which case an increase in purple, ethyl-linked anthocyanins may have compensated for the increase in orange pyranoanthocyanins (Cano-López et al. 2006).
Conclusions
The effect of electrochemical microoxidation using glassy carbon rods on red wine was similar in several respects to that achieved by oxygenation using molecular O2, including the increase in SO2-resistant pigments that develop as anthocyanins bind with other polyphenol components in the wine and a greater decline in monomeric anthocyanins. There was also a similar increase in modified hue (420 to 520 nm abs) for oxygenated and electrochemical microoxidation wines. On the other hand, there was a greater buildup of acetaldehyde and bound SO2 in the electrochemical microoxidation wines for the conditions used in this trial, largely owing to the moderately large currents and high associated electrode potentials required at this level of oxidation. The use of higher surface area electrodes will allow a lower electrode potential to be used and more of the current will go into oxidation of polyphenols, thereby limiting production of acetaldehyde from ethanol oxidation, should that be required.
Footnotes
Acknowledgments: This research was supported by Mission Estate Winery, Hawkes Bay, New Zealand, and winemaker Paul Mooney (hosting Stuart Dykes on an Agmardt PhD Scholarship and Alec Fell on a Technology for Industry MSc Fellowship).
- Received November 2006.
- Revision received April 2007.
- Copyright © 2007 by the American Society for Enology and Viticulture
Literature Cited
Vol 58 Issue 4
