Abstract
Muconic acid is a six-carbon dicarboxylic acid with conjugated double bonds that finds extensive use in the food (additive), chemical (production of adipic acid, monomer for functional resins and bio-plastics), and pharmaceutical sectors. The biosynthesis of muconic acid has been the subject of recent industrial and scientific attention. However, because of its low concentration in aqueous solutions and high purity requirement, downstream separation presents a significant problem. Artificial Neural Networks and Differential Evolution were used to optimize process parameters for the recovery of muconic acid from aqueous streams in a system with n-heptane as an organic diluent and ionic liquids as extractants. The system using 120 g/L tri-hexyl-tetra-decyl-phosphonium decanoate dissolved in n-heptane, pH of the aqueous phase 3, 20 min contact time, and 45 °C temperature assured a muconic acid extraction efficiency of 99,24 %. Low stripping efficiency compared to extraction efficiency was observed for the optimum conditions on the extraction step (120 g/L ionic liquids dissolved in heptane). However, re-extraction efficiencies obtained for the recycled organic phase in three consecutive stages were close to the first extraction stage. The mechanism analysis proved that the analysed phosphonium ionic liquids (PILSs) extracts only undissociated molecules of muconic acid through H-bonding.