Multi-response optimisation of the extraction solvent system for phenolics and antioxidant activities from fermented soy flour using a simplex-centroid design

• Total phenolics were efficiently extracted using only water as solvent.
• Pure ethanol and methanol were not good enough for isoflavones extraction.
• 50 wt% aqueous ethanol can be used to obtain extracts with high phenolics content and antioxidant activity.
• The extractability of isoflavones is not completely consistent with the polarity of solvent.
• Optimisation of the solvent system should be investigated for each sample type.

A simplex-centroid design comprising three solvents (water, ethanol and methanol) was used to optimise the extraction mixture for phenolics and antioxidant activities from defatted soy flour fermented with Monascus purpureus or Aspergillus oryzae. Total phenolics were more efficiently extracted using only water for both samples. The highest antioxidant activities by the DPPH and ABTS methods were obtained using extraction mixtures containing at least 75 wt% water. Specific water:ethanol:methanol ratios promoted the joint optimisation of the total phenolic and isoflavone contents as well as antioxidant activities: 0.5:0.375:0.125 (wt/wt/wt) and 0.5:0.3:0.2 (wt/wt/wt) from defatted soy flour fermented with M. purpureus or A. oryzae, respectively. However, a water:ethanol ratio of 0.5:0.5 (wt/wt) was deemed optimal because it is comprised of green solvents and yielded results that were greater than 90% of the multi-response maximum values. Both the solvents and the sample matrix strongly influenced the extractability of total phenolics and isoflavones.