Determination of the impact of bottle colour and phenolic concentration on pigment development in white wine stored under external conditions
Introduction
Wine has traditionally been stored in glass bottles of different colours and shapes. The selection of the colour and shape is often dictated by market forces in an attempt to make the wine readily identifiable and more attractive to the consumer. Bottle colours from clear to dark brown to dark green are widely used, the selection to some extent being dictated by the expected shelf life of the wine. That is, there is anecdotal evidence that darker coloured bottles tend to give greater protection to wine from the influences of light exposure on the assumption that dark colours do not allow the transmission of UV radiation. Lighter colour or clear bottles tend to be used for wines with short shelf life.
More recently, environmental groups have proposed that the wine industry should move away from dark green to Flint or amber glass bottles as a contribution towards reducing recycling costs and energy demands [1]. Such a move would significantly change the amount and wavelengths of light (ultra-violet and low wavelength visible) impacting on the stored wine.
Early research on the effects of light on Champagne wines [2] suggested that detected off-odours came from the degradation of sulfur-containing amino acids. The term ‘goût de lumiere’ or ‘goût de soleil’ (light sickness) was coined by these researchers. Dozon and Noble [3] examined the sensory effects of white still and sparkling wines that had been exposed to fluorescent light (35 cm from two 40 W fluorescent tubes). Off-odour production was observed as was a decrease in positive aroma descriptors. Wine stored in green bottles showed a significant sensory difference after 31.1 h (still) and 18 h (sparkling). These times were reduced to 3.3 h (still) and 3.4 h (sparkling) when clear glass bottles were used [3].
White wines generally darken with age, but enhanced rates of colour development are indicative of non-enzymatic oxidative processes. The wine industry has traditionally used the absorbance measurement at 420 nm as an indicator of the so-called ‘browning’ of white wine [4], [5], although more recent research suggests that the visual observance of ‘brown’ in white wine consists of a combination of yellow and red colours [6], [7].
Simpson [5], in a detailed study of the relationship between phenolic composition and ‘browning propensity’ of white wines identified catechin-type phenolic compounds as having the most positive correlation with observed colour development, as measured by the absorbance at 420 nm. Clark et al. [8] have shown that tartaric acid can degrade when model wine systems are exposed to sunlight in clear glass bottles, leading to the production of glyoxylic acid. The glyoxylic acid can bridge two catechin molecules to generate a yellow xanthylium pigment that absorbs at 440 nm [9], [10]. When this reaction occurs in the presence of caffeic acid, development of a brown colour is observed [7], [11]. Further, it is also known that xanthylium pigments derived from (−)-epicatechin have double the absorbance intensity of the equivalent (+)-catechin-based xanthylium pigment [12].
Trace amounts of iron can enhance colour development of model wine systems [8]. Iron(III) citrate is photoactive at wavelengths around 360 nm [13] and it has been suggested that similar photoactivity by iron(III) tartrate may lead to the production of the hydroxyl radical that in turn initiates oxidative spoilage processes in the model wine system [8]. Such a process may be critical for the initiation of oxidative colour development reactions.
There are limited reports describing the relationship between bottle colour and oxidative development of white wines. One study on fino wines exposed to a 1500 W lamp showed that losses of phenolic compounds were greater in clear (transparent) glass, but greater colour development (420 absorbance) occurred for wines in topaz bottles [14]. An alcohol beverage made from oranges was stored in clear, green and brown bottles [15]. A higher browning index was observed with clear bottles, but a phenolic analysis was not performed.
This work was undertaken to examine the link between bottle colour and oxidative pigment development of white wines. Both model wines and a commercially sourced white wine were used in the study. Four different coloured bottles were used, representing the range commonly available in the Australian wine industry. Outdoor storage was used which, although somewhat extreme, reflects the fact that the exposure of bottled wine to light tends to occur more in retail outlets or during storage by consumers than in wineries. The outdoor exposure also allowed ready identification of factors critical to colour development including the relative effects of the initial phenolic concentration, glass bottle colour, and light exposure.
Section snippets
Reagents and chemicals
Water purified through a Milli-Q (Millipore) water system (ISO 9001) was used for all solution preparations and dilutions. All glassware was soaked overnight in a 10% nitric acid (BDH, AnalR) and then rinsed with copious amounts of water. l-(+)-tartaric acid (>99.5%), (+)-catechin hydrate (98%), (−)-epicatechin hydrate (98%), potassium bitartrate (99%) were purchased from Sigma–Aldrich. Ethanol (AR grade, >99.5%, Ajax Fine Chemicals), methanol (AR grade, >99.9%, Mallinckrodt Chemicals), glacial
Characterisation of bottle types
To assess the ability of each bottle type to limit or stop UV/visible light reaching the sample within the bottle, a section of glass from the main body section of each bottle type was removed using a glass cutter and the transmission spectrum recorded. These sections were representative of the main exposure area of the wine in the bottle. Fig. 1 compares the transmission spectrum for the four bottle types. It is apparent that Flint is capable of transmitting all visible and some UV light and
Acknowledgement
This work was supported by the Australian Grape and Wine Research and Development Corporation (project CSU 01/02).
References (29)
- et al.
Food Chem.
(2007) - et al.
Tetrahedron Lett.
(1999) - et al.
Food Chem.
(2007) - et al.
Inorg. Chim. Acta
(1994) - et al.
Food Chem.
(2007) - et al.
J. Chromatogr. A
(2000) - http://www.wrap.org.uk/retail/case_studies_research/glassrite_wine.html, 2009 (accessed 8...
- et al.
Sci. Alim.
(1983) - et al.
Am. J. Enol. Vitic.
(1989) - et al.
Chemical Analysis of Grapes and Wine: Techniques and Concepts
(2004)
Vitis
Aust. J. Grape Wine Res.
Aust. J. Grape Wine Res.
Aust. J. Grape Wine Res.
Cited by (48)
Non-enzymatic browning of wine induced by monomeric flavan-3-ols: A review
2023, Food ChemistryDevelopment of an irradiation equipment to accelerate the degradation of rosé wine in antique green and flint bottles
2023, Current Research in Food ScienceA colourimetric fluorescent probe for the sensitive detection of total iron in wine
2022, Food ChemistryCitation Excerpt :When wine is stored in a sealed anaerobic state, iron mainly exists in the form of ferrous ions. However, when wine comes into contact with air, ferrous ions are oxidized to iron ions (Kunsági-Máté, Stampel, Kollár, & Nikfardjam, 2008; Maury, Clark, & Scollary, 2010). Iron ions in wine form unstable soluble coordination compounds with organic acids such as tartaric acid, malic acid and citric acid, which can result in iron breakdown precipitation (Grant-Preece, Barril, Schmidtke, & Clark, 2017).