Chemical analysis of extracts of New Zealand woods in wine

Background: Reliance on a restricted range of grape varieties and flavour profiles is potentially a risky situation for sustainability of the New Zealand wine industry. Competitors in other countries may offer very similar products at lower cost. Whereas geographical exclusivity may help to minimize this risk, New Zealand wines with few exceptions tend to follow copyable European models. There is a need to broaden the production base into other wine styles that could compete on distinctiveness and overall quality rather than on price. Oak, usually in the form of barrels, is the traditional way of flavouring wines with distinctive caramel, smoke- like and vanillin notes. Use of woods other than oak to flavour wine may be the way to introduce new wine styles. Aim: This project examines the chemistry of the application of New Zealand woods instead of oak to flavour wine. Methods: A range of woods was selected based on existing use and botanical similarity (oak, and cherry and silver beech), or on their association with the New Zealand ethos and botanical similarity (matai, feijoa, macrocarpa, pohutukawa, radiata pine, totara, kahikatea, rimu, and manuka). Wood chips cut to 2 x 1 x 0.25 cm – instead of barrels – were toasted at two levels, 200°C for two hours, deemed light toast, and at 210°C for three hours, deemed a dark toast. The parameters investigated were moisture content after drying to 110°C, wood weight losses resulting from the different toasting levels, colour measurements in Hunter colour space, ultraviolet spectrophotometric analysis of extractables, and their gas chromatographic analysis. Results: Of the woods measured, oak had the lowest moisture content. The weight loss of oak chips at 200°C was much greater than that of other woods, but the colour change did not indicate losses due to severe charring. Other woods that showed severe weight loss on dark toasting (rimu, macrocarpa) did char severely. Colour measurements showed that toasting did not greatly affect the hue angle (the basic colour) of the chips, but the colour intensity (saturation) was strongly reduced, as was the overall reflectance of light (L* value). Light toasting yielded higher concentrations of extractables as determined by spectrophotometry between 200 and 400 nm. Model and real wines treated with six of the woods at both toasting levels were analysed by gas chromatography. At both toasting levels, American oak yielded the greatest number of extractables. American oak, manuka, and matai added similar number of compounds in the model and real wine light toast treatments, whereas pohutukawa, silver beech and totara yielded more compounds in the real wine than in the model wine. 2-(Methoxymethyl)-5-methoxyphenol was unique to both toasting levels of American oak in real wine. Similarly, 3,4-dimethyl phenol, 5-methyl-2-furaldehyde and 5-butyldihydro-4-methyl-2(3H)-furanone, were detected only in real wine treated with American oak toasted at both levels. In all wood treatments, 4-hydroxy-3-methoxy cinnamaldehyde and 3,5-dimethoxy-4-hydroxy cinnamaldehyde were present in model wine but not in real wine. This was attributed to the presence of SO2 in real wine. The greatest expression of 4-hydroxy-3-methoxy cinnamaldehyde was in the manuka dark toast treatment, where its relative concentration was approximately 10 fold higher than in other wood treatments. 4-(2-Hydroxyethyl)phenol was detected only in manuka dark toast in model wine, whereas it appeared in all wood and toast treatments in real wine. Its greatest expression was in matai light toast and silver beech dark. 4-(Ethoxymethyl)-2-methoxy phenol was detected only in the matai and totara dark toast treatments in both the model and real wine. Furfural and vanillin were present in both real and model wines for all wood treatments. American oak, manuka, and totara showed the greatest expression. Conclusion: As determined by ultraviolet absorption and gas chromatography, there were patterns of extraction common to all woods and both toast levels, as well as several unique and near unique patterns. This research was limited to physical and chemical changes, but preliminary sensory trials – unreported here – suggest that manuka may be a good commercial flavour prospect. It certainly had an extractable profile resembling that of American oak