Solubility model of palm oil extraction from palm fruit using sub-critical R134a

Abstract

The exploration of alternative solvents for supercritical fluid extraction (SFE) technology has been attributed to the high capital investment due to higher pressure required by using supercritical carbon dioxide (SC-CO2) as a solvent. One of the potential alternative solvent is the sub-critical R134a, which can be operated at lower pressure than SC-CO2. This research investigate the use of dense gas approach, density based approach and solubility parameter to predict the solubility model of palm oil extraction from palm fruit using sub-critical R134a in SFE systems. Firstly, the dense gas approach involves the estimation of pure component vapour pressures, critical properties and binary interaction was implemented. This is followed by the development of thermodynamic model by using the equation of states (EOS) which are Peng-Robinson (PR) and Soave-Redlich-Kwong (SRK) combined with four mixing rules that includes excess Gibbs energy model. The density-based as the second approach requires the information of density, pressure and temperature. The performance of seven density based models were analysed in this research. Lastly, the solvent and solute solubility parameters were calculated using regular solution theory. Solvent specific coefficient for R134a was then determined using experimental data published. The proposed solvent specific coefficient for R134a is 11.8138 and this coefficient can be used for universal calculation of solubility which involves R134a as a solvent. Based on comparison of all correlation method, dense gas approach using the combination of PR EOS and Solute-Solute Interaction (SSI) mixing rule shows the lowest Average Absolute Relative Deviation (AARD), 0.08% compared to other methods. Due to complex calculation involved, T-P model regressed by Design Expert software is suggested as the best method to model the solubility behaviour of palm oil extraction from palm fruit using sub-critical R134a in the SFE systems.