The impact of metal speciation on the development, shelf-life and sensory properties of wine
Abstract
New methods were created for measuring total Cu and Cu fractions in wine, including Cu(II) organic acids (Cu fraction 1), Cu(I)-thiol complexes (Cu fraction 2), and copper sulfides (Cu fraction 3). Cu fraction 3 concentration could be lowered by exposing wine to oxygen, filtration approaches, fining with copolymers or bentonite treatment. Cu fractions 1 and 2 inhibited reductive aromas but during bottle aging their concentrations decreased, and their depletion was accelerated by increasing temperature, light exposure or metallic aluminium. Guidelines for Cu fraction 1 and 2 concentrations at bottling are proposed for minimising reductive aroma accumulation in wines during aging.
Summary
Copper is known to participate in oxidative and reductive reactions in wine. Previous research provided a preliminary understanding of binding agents for Cu in wine and techniques to quantify certain broad categories of Cu using research orientated approaches.
This project advances the past research with the advent of novel measures of both total Cu and Cu fractions in wine, including fractions that have never been reported before. Importantly the techniques are more amenable for use in wineries given they use equipment commonly found in wineries (i.e., spectrophotometer). These measures enabled a better understanding of the role of Cu in oxidation processes, the reductive development of wine and how Cu can be more efficiently removed from wine. As well as providing new insights for these areas, the Cu measures allow a more strategic usage of Cu to improve the outcomes for wine after bottle aging.
The 2 bicinchoninic acid (BCA) colorimetric analysis of total Cu in white wine was an outcome of our previous Wine Australia-funded project. In this current project, a digestion procedure for red wine was devised to destroy red pigmentation and allow total Cu determination by BCA colorimetry. This colorimetric method was further modified allowing the determination of three difference fractions of Cu in white wine: Cu fraction 1 was associated with Cu(II) organic acids, Cu fraction 2 with weakly-bound Cu(I) thiol compounds, and Cu fraction 3 predominantly with copper sulfides but also some strongly-bound Cu(I) thiol compounds. The collection of permeate from wine passed through cellulose-based depth filters incorporated with diatomaceous earth, allowed the measurement of the combined Cu fraction 1 and 2 concentration in white wine, and Cu fraction 1 in red wine. Progress has been made automating the Cu measures but further work is required for full validation.
Cu fractions 1 and 2 were found to have a significant impact on rate of oxygen reaction in white wines containing ascorbic acid, but this was not the case in white wines without ascorbic acid and in red wines. For the latter, the total Cu concentration was more important in determining oxygen consumption rates than the specific distribution of Cu among the different fractions. The Cu fractions did not influence the amount of total sulfur dioxide consumed for a given amount of oxygen in any wine, but the presence of ascorbic acid in white wine could nearly halve this value.
Cu fraction 3 was shown not to have any direct benefit for wine, other than being a potential source of Cu fractions 1 and 2 (e.g., transition afforded with some oxygen exposure). In fact, with light exposure to white wine, Cu fraction 3 could act as a significant precursor for free H2S. Most filtration methods were able to remove some Cu fraction 3. However, membrane filtration media only worked efficiently on small volumes and was found not to be effective at larger volumes. The most effective procedures for removal of Cu fraction 3 was the use of depth filtration with cellulose pads incorporating diatomaceous earth or fining with the copolymer PVI/PVP. Bentonite could also be used for white wines for Cu removal. PVI/PVP was unique for its ability to significantly decrease the concentration of all Cu fractions in white wine.
This project identified the most suitable gas chromatography (GC) methodologies for measuring the forms of H2S and methanethiol that contributed directly to reductive aromas in wine. This was the free H2S and free methanethiol measures by GC with a Sulfur Chemiluminescence Detector (SCD). Gas detection tubes were also suitable for measurement of free H2S, albeit with some overestimation compared to GC-SCD, but still linked well to the sensory assessment of reductive aroma. The link of Cu fractions to the inhibition of reductive aroma compounds was comprehensively established using studies involving 800 bottles of wine. Free H2S did not accumulate in wines with combined Cu fraction 1 and 2 concentrations above 0.02 mg/L and free methanethiol did not accumulate in wines with Cu fraction 1 above 0.03 mg/L. An unfortunate ramification was that, as for the inhibition of free methanethiol (i.e., a thiol compound), Cu(II) addition also decreased the concentration of varietal thiol aroma compounds present in Sauvignon blanc. The effect of Cu(II) on varietal thiol compounds was immediate and lasted until the final timepoint measured in the bottle trial (i.e., 18 months). However, despite the lowered varietal thiol concentrations, the Cu(II) additions at bottling still often led to Sauvignon Blanc wines with more perceived tropical and fruit aromas than the control wines. This was attributed to the added Cu(II) removing masking compounds, such as free H2S and methanethiol.
A key aim of this project was to ultimately provide guidelines for the concentration of Cu fractions at bottling that may limit the onset of reductive development in wines. This ideally required identification of those wines at risk of aging reductively and the establishment of Cu fraction 1 and 2 concentrations that would minimise both the risk of reductive development and the magnitude of the Cu(II) addition. To achieve this, it was necessary to first assess the stability of protective Cu fractions in wine. Cu fraction 1 and 2 were found to decrease during the aging of all wine, with a concomitant increase in Cu fraction 3. The rate of decrease in the protective Cu fractions varied from wine to wine and was accelerated by elevated temperature, light exposure or storage when in contact with metallic aluminium (i.e., cans). This meant that using kinetic studies and average rates of decay were not suitable to predict the time of depletion for the protective Cu fractions in wine. Instead, from the aging study of 43 wines over 12-18 months, it was found that the lag period between the loss of Cu fraction 1 and 2 and the emergence of the reductive aroma compounds was quite large (i.e., ~1-year). Consequently, the presence of Cu fraction 1 and 2 at bottling could still offer protection to wines even after depletion. Based on the aging studies in ideal conditions (i.e., 14-17 °C in darkness), free H2S did not accumulate above 2 µg/L in wines with 0.115 mg/L for the combined Cu fraction 1 and 2 concentration at bottling, and free methanethiol did not accumulate above 5 µg/L when the Cu fraction 1 concentration exceeded 0.04 mg/L at bottling.
Progress was made on assessing a wine for its potential to age reductively. This involved bottling a wine in a similar manner to the final product (i.e., under screw cap and with similar total packed oxygen) and heating the bottled wine for 21-28 days at 40 °C. If the free methanethiol measured in the heated wine was above 5 µg/L then this was also likely to be the case for the wine after 12-18 months of bottle aging. Obviously a more rapid test would be beneficial and the subject of future work.