Extraction of mandelic acid with tri-octyl-phosphine oxide (TOPO) in different solvents: Equilibrium and neural network analysis

KİRİŞ B., Uslu H., Santos D., BÜYÜKSAATÇI KİRİŞ S., AŞÇI Y. S.

Journal of the Indian Chemical Society, vol.101, no.11, 2024 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 101 Issue: 11
  • Publication Date: 2024
  • Doi Number: 10.1016/j.jics.2024.101412
  • Journal Name: Journal of the Indian Chemical Society
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Analytical Abstracts, Chemical Abstracts Core
  • Keywords: Mandelic acid, Neural network, Solvent, Tri-octyl-phosphine oxide
  • Istanbul University Affiliated: Yes

Abstract

Mandelic acid is an important carboxylic acid used in pharmaceutical industries. It is also important to use it as a purified form. In this study, selective extraction of mandelic acid was done by tri-octyl-phosphine oxide (TOPO) diluted in different solvents such as methyl isobutyl ketone, 1-octanol, octyl acetate, dimethyl phthalate, 2-octanone, cyclohexane and toluene. The high selectivity of mandelic acid from aqueous solution was supported by thermodynamic parameters (loading factor, distribution coefficient, and extraction efficiency). The obtained values for each solvent were applied to the Neural Network Analysis to predict phase equilibrium behaviour in ternary systems. The results showed the highest mandelic acid extraction efficiency (93.65 %) and distribution coefficient (14.74) were attained with the organic phase mixture prepared with MIBK and TOPO. Also, it was found that extraction efficiencies increased with increasing TOPO amount in the medium for all studied solvents.