Diffusion coefficients of C18 unsaturated fatty acid methyl esters in supercritical carbon dioxide containing 10% mole fraction ethanol as modifier

• We determined the diffusion coefficients in 10 mol% ethanol modified CO2.
• Four C18 unsaturated fatty acid methyl esters were selected as solutes.
• The diffusion coefficient decreased with increasing the number of C-C double bonds.
• We investigated several available models for pure CO2 with the experimental data.
• The free volume model of Dymond gave the least average absolute deviations.

To determine the molecular diffusion coefficients of C18 unsaturated fatty acid methyl esters in supercritical carbon dioxide (scCO2) containing 10 mol% ethanol as a modifier, four methyl esters of C18 fatty acids, i.e., methyl oleate, methyl ricinoleate, methyl linoleate and methyl linolenate were selected as the typical solutes. The diffusion coefficients were measured at temperatures from 313.15 to 333.15 K and pressures from 15 to 27 MPa using the Taylor–Aris chromatographic peak broadening (CPB) technique. The influences of temperature, pressure, density and viscosity of the solvent mixture on the diffusion coefficients were examined. The results show that methyl oleate always diffuses faster than methyl ricinoleate at the same operating condition. Moreover, the D12 values in ethanol-modified scCO2 decrease with the increase of the number of C-C double bonds in C18-methyl ester, which is consistent with the trend reported in pure scCO2. The diffusivity data are compared with the estimation of eleven predictive models. The modified Wilke–Chang equation is the best purely predictive model and the free volume model of Dymond with two adjustable parameter gives the least errors with average absolute deviations lower than 2.5%.

Figure optionsDownload as PowerPoint slide