Modelling of Fractionated Condensation for Off-Flavours Reduction from Red Wine Fermentation Headspace

Research output: Contribution to journalArticlepeer-review

14 Downloads (Pure)


A mathematical model of fractionated condensation is proposed for predicting the recovery and fractionation of target aromas from red wine fermentation headspaces in order to remove off-flavours. The applicability of the model is assessed for two different alternative processes: fractionated condensation and vapour permeation–fractionated condensation. The aromas of the headspace of red wine fermentation are commonly lost through the fermenter venting system and are enhanced by the stripping effect of the produced CO2. To mimic the operating conditions during the red wine fermentation, all experiments were performed at 30 °C with a red wine model solution containing relevant red wine aromas, the cosolvent ethanol at representative concentrations, and CO2. Both studied processes allow for a good recovery of esters in the 2nd condenser, with over 80% of ethyl acetate and isoamyl acetate recovery when using vapour permeation–fractionated condensation and a recovery of 84–96% of all esters when using fractionated condensation. However, only the integrated process of vapour permeation–fractionated condensation achieves a significant decrease in the amount of ethyl phenols (off-flavours compounds) in the 1st condenser, above 50%, as expected due to the use of an organophilic membrane. The developed model was validated experimentally for the integrated process, proving to be a highly valuable tool for the prediction of aroma fractionation, aiming at the removal of off-flavours.

Original languageEnglish
Article number875
Number of pages12
Issue number9
Publication statusPublished - Sept 2022


  • aroma recovery
  • modelling fractionated condensation
  • off-flavours
  • red wine aroma
  • vapour permeation


Dive into the research topics of 'Modelling of Fractionated Condensation for Off-Flavours Reduction from Red Wine Fermentation Headspace'. Together they form a unique fingerprint.

Cite this