WINETECH Technical Yearbook 2020

linked to autolysis of the non-Sacch as they die off after Sacch inoculation. • Frootzen fermentations contained the highest concentrations of 2-phenyleth- yl-acetate (floral) and isoamyl acetate (fruity, banana). IN CONCLUSION The study confirms once again the importance of measuring YAN before making any decisions with regard to fermentation. The study also reveals that non-Sacch yeasts do compete with the S. cerevisiae yeasts for nutrients in a species- dependent manner, thereby sometimes affecting the possibility of fermentation to complete and the aromatic outcome. It is therefore very wise to supplement musts with yeast nutrients, whenever non-Sacch yeasts initiated the fermentation. Saccharomyces cerevisiae strains are starved for nitrogen during the final stages of commercial production in order to stop budding formation before drying, since budding yeasts are very susceptible to heat damage. In order to grow in must, com­ mercially-produced yeasts need nitrogen, which will be in the must as a result of the natural YAN. When non-Sacchs start the fermentation, the YAN can be very low

after inoculation leaving the synthetic must with a YAN of 6 mg/L or less (treatment 3). • After 48 hours, Frootzen and Flavia synthetic musts contained YAN’s of 54 and 41 mg/L (treatment 3). • The more natural must-like medium (treatment 3) did not provide better fermentation support compared to treatments 1 and 2. • When EC 1118 was co-inoculated 48 hours after Frootzen and Flavia inoculations, the fermentations went to dryness in the filtered (wild yeasts were removed before S. cerevisiae inoculation) and unfiltered synthetic musts. This indicates the favourable impact of these yeasts on EC 1118 to complete the fermentation (with 54 and 41 mg/L YAN). • In the cases of Concerto and Biodiva/ EC 1118 co-inoculations, the filtered must got stuck, indicating not enough nutrients for EC 1118 to complete the fermentation. In the unfiltered must, the non-Sacch yeasts seemed to release some nutrients back into the medium, thereby allowing fermentations to reach dryness. This is most certainly

ences in residual sugar with or without the presence of ammonia. • All strains consumed glucose faster that fructose. • EC 1118, Concerto and Biodiva had similar sugar consumption rates in the first 48 hours, which were much faster than Frootzen and Flavia. • Depending on the yeast species, ammo- nia consumption occurred at different rates and different amounts. • 18 hours after inoculation, EC 1118, Concerto and Biodiva consumed al- most all of the ammonia (less than 5 mg/L remaining). Frootzen complet- ed ammonia consumption 24 hours after inoculation and Flavia took longer than 48 hours after inoculation. • Amino acid consumption started at about six hours after inoculation for all yeasts. Amino acids were taken up in different order of preference for the different yeasts. • Interestingly, Frootzen and Flavia first released certain amino acids before uptake began. • EC 1118, Concerto and Biodiva took up most amino acids by 48 hours

by the time the commercial S. cerevisiae yeast is inoculated and not supportive of complete fermentations. All yeasts have different requirements for nitrogenous compounds. Suppliers are able to provide guidance on their yeasts’ requirements. The standard addition of DAP to all musts, regardless of YAN and micronutrient content of musts, and yeasts conducting the fermentation, is extremely unscientific and not conducive to optimising wine quality. ACKNOWLEDGEMENTS The study was funded by Winetech and the National Research Foundation. REFERENCE Prior, K.J., Bauer, F.F. & Divol, B., 2019. The utilisation of nitrogenous compounds by commercial non- Saccharomyces yeasts associated with wine. Food Microbiology 79, 75-84. https://doi.org/10.1016/j. fm.2018.12.002.

For more information, contact Karien O’Kennedy at karien@winetech.co.za.

WINETECH TECHNICAL YEARBOOK 2020 122