The application of D-optimal design for modelling the red wine ageing process

The aim of this study was to investigate effects of grape seed content in pomace and ageing time on phenolic compounds and colour of red wine using a statistical experimental design. In accordance with the results, it was planned to evaluate a potential use of obtained model for determination of wine shelf life, in terms of its chromatic characteristics. Response surface methodology was used and experiments were carried out according to the two-factor D-optimal design with cubic model. According to the design, wines were obtained from pomace supplemented with 30-120 g/L of seeds and left for ageing from 3 to 48 months. Parameters (responses) monitored in this study were: total phenols content, total anthocyanins content, flavan-3-ols content, colour intensity and hue. Modified third order polynomial equations, which represent fitted models for investigated responses, are shown as adequate (R2 > 0.90 and P < 0.001). The validation of models obtained showed good agreement between modelled and independent experimental results. This study has shown that addition of supplementary quantities of grape seeds in pomace led to stabilization of red wine colour during longer storage periods. Despite a sharp fall of anthocyanins content during first year (decrease even by 80%), colour intensity of red wines did not suffer greater changes (decrease by 10%) and it was still acceptable after four years of storage. The use of obtained models for determination of wine's shelf life, as well as, the necessity of further examination, is shown.

► We use D-optimal design for modelling the red wine ageing process. ► We examine effects of seed content and ageing time on wine phenolics and colour. ► Obtained modified third order polynomial equations are shown as adequate. ► Addition of grape seeds in pomace led to stabilization of wine colour. ► Potential use of the models for determination of wine's shelf life was shown.