The Effect of Pre-fermentation Fining & Other Post-harvest Manipulations on Sauvignon blanc Aroma

Abstract

Fining is a common practice used in the wine industry which assists with clarification and/or provides corrective treatment for the aroma, taste or aftertaste of a juice or wine. During this thesis project the fining agents activated carbon (AC), gelatin (G), gelatin with silicon dioxide (GSi), polyvinylpolypyrrolidone (PVPP), and a mixed agent consisting of bentonite, PVPP, and isinglass (M) where investigated for their ability to influence the concentration of several aroma compounds in Marlborough Sauvignon blanc. Sensory trials were also conducted to investigate if any chemical changes impact the sensory aspect of the wines. Although both free run and press fraction juice fining was considered, a stronger emphasis was placed on the production of press fraction wines. The focus of this thesis was conceived in direct collaboration with a winery which noted anecdotally that press fraction juices require severe fining regimes thought to be due to the extended skin contact and oxygen exposure they undergo. Therefore, the project set out to explore if pre-fermentation juice fining could modulate aroma compounds present in Marlborough Sauvignon blanc press fraction wines. The first fining trial conducted was to survey the potential for the chosen commercially available agents (AC, G, PVPP, and M) to induce a change in the aroma chemistry of the experimental wines. Both free run and press fraction juices of two vineyards (Awatere valley and Wairau valley) were fined and the resulting wines analysed. Results showed that several aroma compounds may be influenced by prefermentation fining. For instance, there was a significant (p < 0.05) reduction in linalool and hexan-1-ol across each vineyard and juice combination when AC fining was employed, compared to their respective controls. Moreover, it was also demonstrated that juice origin (vineyard and fraction) was of a high priority when determining the aroma compound content of the experimental wines. Sensory analysis showed that the panel indicated lower counts of the aroma attribute Green vegetables for the Awatere valley wines which had been fined using AC, compared to the respective control wines. Following on from this experiment, another trial using excessive amounts of juice fining was conceived to provide information regarding part batch fining, a strategy used by some winemakers. Wairau valley press fraction juice was fined using AC, G, and PVPP at rates of 5, 6, and 8 g/L, respectively. Again, a significant (p < 0.05) reduction in both linalool and hexan-1-ol was observed for AC fining, as well as a reduction in the varietal thiols 3-mercaptohexanol (3MH) and 3-mercaptohexyl acetate (3MHA). Surprisingly, a significant (p < 0.05) increase in ethyl (dihydro)cinnamate as well as benzyl alcohol as a result of G fining was also noted. Sensory analysis showed a Green veg aroma to help distinguish between wines as well as other attributes such as Floral honey aroma, Floral/honey taste, and Salty aftertaste. The last experiment in this series of fining trials was to determine if results found at the laboratory scale would persist in an industry environment. For this purpose, AC, G, and GSi were used to fine 5,000 L tanks of Wairau valley or a blend of Marlborough press fraction juices. From this, it was shown that several of the observations in aroma compound concentration were applicable at this scale including significant (p < 0.05) reductions in linalool, hexan-1-ol, 3MH, and 3MHA for AC fining. Furthermore, fining with G or GSi generated wines with higher ethyl (dihydro)cinnamate concentrations, a phenomenon not commented upon previously in the literature of press fraction juice fining. An exploratory sensory analysis here indicated that aromas of Floral/talcum powder and Citrus may occur due to the choice of fining agent applied prefermentation. During this thesis project, it was noticed that several external influences occurring post-harvest and pre-winery processing to machine harvested grapes may be influencing the final wines thiol potential. Because of the relationship with the winery, the opportunity arose to conduct small scale pilot studies into these post-harvest factors, namely; (i) grape transport time after harvest, (ii) ultra-violet (UV) light irradiation, and (iii) grape holding temperature before processing. Here, simulated transportation times (0, 1.5, 3, and 4.5 hours) were applied to machine harvested Sauvignon blanc and the levels of 3-S-(hexan-1-ol)-L-cysteine (Cys-3MH) and 3-S-(hexan-1-ol)-L-glutathione (Glut-3MH) analysed. Results indicated that the levels of both thiol precursors increased with longer transportation time. However, fermentation of the juices revealed no subsequent increases in 3MH or 3MHA associated to an increased transportation time. As for the irradiation by UV light, Sauvignon blanc grapes were exposed to a UV light source post-harvest in two forms; either whole bunches or machine harvested. The thiol precursors Cys-3MH and Glut-3MH were quantified in the juices before and after UV treatment. Results showed that irradiation of the grapes with UV light had little to no effect on the thiol precursors. Wines were fermented from the corresponding juices and 18 aroma compounds were quantified. Differences were found between UV treatments of the wines for 3MH, hexan-1-ol, ethyl butanoate, ethyl hexanoate, ethyl octanoate, and phenylethyl alcohol. However, these changes did not occur significantly (p < 0.05) for both grape forms trialled. Grape holding temperature was explored by determining the respective thiol precursor and free varietal thiols of several juices and their wines held for a short skin contact period (2 h) at the temperatures 6, 15, and 24 °C. From this trial, only the 15 and 6 °C juice samples differed significantly in Cys-3MH and Glut-3MH contents. Both thiol precursors were found to be at their highest levels in 15 °C juice samples and their lowest in 6 °C juice samples prior to fermentation. The resulting experimental wines displayed no significant differences (p > 0.05) between the various temperatures explored. Overall the series of fining trials established that pre-fermentation fining does have the ability to alter the concentration of aroma compounds in both free run and press fraction Sauvignon blanc wines. Related sensory work revealed that these changes detected chemically were not always enough to reshape the perception of the wine sensory-wise. From the three pilot studies, it was determined that thiol precursors could be influenced by these external factors, however this was not necessarily true for the free thiol contents of the resulting wines. Chiefly, this thesis project is influential as it contributes new, translatable knowledge for use in the winemaking community.