Institute for Wine Biotechnology

Permanent URI for this community

https://scholar.sun.ac.za/handle/10019.1/5053

Browse

Browsing Institute for Wine Biotechnology by browse.metadata.advisor "Brand, Jeanne"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    South African de-alcoholised sparkling wines: a study focused on sensory and chemical profiles
    (Stellenbosch : Stellenbosch University, 2022-12) Maesela, Lethabo Mologadi; Nieuwoudt, Helene; Brand, Jeanne; Pentz, Chris; Muller, Magdalena; Stellenbosch University. Faculty of AgriSciences. Dept. of Viticulture and Oenology. Institute for Wine Biotechnology.
    ENGLISH ABSTRACT: De-alcoholised sparkling wines contain a maximum of 0.5% v/v ethanol and a minimum carbon dioxide (CO2) pressure of 300 kPa. The wines are produced through yeast-mediated alcoholic fermentation to full ethanol-strength of table wines (circa 12 to 14% v/v), followed by the physical removal of the volatile aroma fraction and reduction of the ethanol concentration. De-alcoholised base wines are reconstituted by re-introducing the volatile fraction, adding permitted chemical compounds, and sparging with external CO2. Although South Africa (SA) only recently joined the global trend of producing de-alcoholised sparkling wines in response to consumer preference for the products, no published information was available on their sensory quality and chemical composition when this study was undertaken. This study used a quantitative and qualitative approach to investigate the sensory quality of nine commercially available South African de-alcoholised white and rosé sparkling wines. A panel of 51 South African wine industry professionals tasted the wines and evaluated their quality using a 20- point scoring system with maximum points for each sensory modality: appearance 3, aroma 7, and palate 10. Qualitative wine sensory profiles were generated using the free listing method, whereby the professionals described their perceptions of the different modalities. Text data mining included standardisation of raw text, lemmatisation to form sensory groups, and categorising of the groups as positive, neutral, or negative wine attributes. Correspondence analysis identified the sensory categories that best described the wine profiles. The wines’ average total quality scores ranged between 11 and 14 points out of 20, with palate quality scoring, on average, from 4.9 to 6.0 out of 10 points. The free listing method produced an information-rich dataset with 2414, 2110, and 3321 word counts for appearance, aroma, and palate, respectively. The lemmatised text data resulted in 10 appearance, 29 aroma and 61 palate sensory categories. Examples of neutral sensory categories included ‘muscat’, ‘citrus’, and ‘fruit’ aroma, whie positive sensory categories included ‘wine-like’ aroma and ‘nose-palate follow-through’ of wine flavours. Negative palate categories (22 in total) included ‘watery’, ‘short finish’, ‘unbalanced’ and ‘acidic’. The volatile aroma profile and basic wine oenological parameters were generated and compared to published data on full-ethanol strength sparkling wines, in the absence of published data on de-alcoholised sparkling wines. Glycerol concentrations were markedly higher in the de- alcoholised wines, ranging from 14.30 to 20.20 g/L. Volatile compounds’ concentration showed lower ranges than in full-strength sparkling wines. For example, ethyl acetate and isoamyl alcohol ranged from 0.427 to 4.677 mg/L and 0.373 to 5.636 mg/L in this study, respectively, versus 8.000 to 45.200 mg/L and 16.317 to 167.080 mg/L in full-strength wines. The results showed that de-alcoholised sparkling wines are unique products with distinct sensory and chemical profiles. Future research topics to be pursued on best de-alcoholisation and re-constitution practices were pointed out. As a first exploratory study into this technically challenging product category, the results generated meaningful and informative feedback for the South African wine industry.