Production of monoglycerides by glycerolysis of olive oil with immobilized lipases: effect of the water activity

The production of monoglycerides by glycerolysis of olive oil catalyzed by lipases from Candida rugosa, Chromobacterium viscosum and Rhizopus sp. immobilized in a hydrophylic polyurethane foam was investigated. The effect of the amount of aqueous phase used for foam polymerization on the competing reactions of glycerolysis and hydrolysis was studied. The highest monoglyceride production was achieved with the C. rugosa lipase which was thus selected for subsequent studies.The extent to which hydrolysis and glycerolysis occur and the influence of the initial a _w of the system on both reactions were also investigated. In glycerolytic reaction systems, initial rates of fatty acid release were always higher than in hydrolytic systems. At a _w values lower than 0.43, hydrolysis was completely repressed, although glycerolysis still occurred. This suggests that hydrolysis of the ester bond in the glyceride, promoted by glycerol, is the first reaction step.In glycerolysis, initial rates of FFA and DG production increased exponentially with the initial a _w of the system.The lowest total conversion (in terms of % TG consumed) at 48 hours was obtained at intermediate a _w values; higher conversions at extreme a _w indicated an increase in hydrolytic and glycerolytic rates, at high and low a _w respectively.The yield of MG increased with decreasing a _w . The highest yield of MG (70%, w/w) was obtained at the lowest a _w used (0.23). The initial a _w of the reaction system is an important parameter in glycerolysis.List of Symbols a _w thermodynamic activity of water - C ₁₂₀ lauric acid - C ₁₄₀ myristic acid - C ₁₈₁ oleic acid - DG dediglyceride (s) - FAME fatty acid methyl ester (s) - FFA free fatty acid (s) - FID fire ionization detector - Gly glycerol - n number of replicates - MG monoglyceride (s) - PCA principal component analysis - PU polyurethane - r correlation coefficient - TG triglyceride (s) - TO triolein - significance level The authors are grateful to Prof. L. Beirão da Costa and to Prof. L. Campos, Inst. Sup. de Agronomia (ISA), and to Prof. J.M.S. Cabral, Inst. Sup. Técnico, Lisbon, Portugal, for inspiring discussions and advice, and to Mrs. Marlene Dionísio, ISA, for invaluable help with some of the experimental work.