COMPARISON STUDY FOR THE SOLUBILITY PREDICTION OF PHENANTHRENE IN SUPERCRITICAL CO2 ENTRAINED WITH n-PENTANE WHEN USING EQUATION OF STATE AND ARTIFICIAL NEURAL NETWORKS

The present work deals with the comparison study for the solubility prediction of phenanthrene in pure supercritical CO2 and supercritical CO2 entrained with n-pentane (n-C5) as a liquid solvent. The experimental data obtained from literatures for systems above are modeled by using two techniques:1. Peng-Robinson equation of state (PR- EOS). 2. Artificial Neural Networks (ANN). The results of two techniques showed that the ANN technique gives excellent agreement with the experimental data of the systems taken and the percentage of average absolute relative deviation (%AARD) ranges from 5.668×10-5 % to 2.089×10-3 % meanwhile the model when using PR- EOS gives good agreement with (%AARD) ranges from 3.34% to 8.67%.