Algorithm for estimating SARA fraction properties and its application in predicting the densities of heavy oils, bitumen and diluted bitumen with solvents

Authors:

Abstract
Using aromatic ring index (ARI) and the densities of saturates, aromatics, resins and asphaltenes (SARA) at any temperature, an algorithm was developed to iteratively estimate the molecular weights and specific gravities (60 °F) of the SARA fractions. The algorithm circumvents molecular weight measurements in reporting SARA fractions properties. Then, using the corresponding state principles the critical properties and acentric factor of the fractions were estimated using the correlations available in literature. Further, applying regular solution theory the densities of heavy oils, bitumen, and five diluted Athabasca bitumen with 5 solvents were predicted from Hankinson-Thomson and a compressed liquid density correlation. The average absolute deviation (AAD) in predicting the densities was found to be 1.33% for 1524 points with temperature range from 288 K to 463 K and pressure from 0.1 to 10 MPa. The method exhibits predictive ability and can be used satisfactorily for any heavy oil/bitumen/solvent system. In order to improve the accuracy of the calculated densities, a constant density translation method can also be applied to the bitumen/oil mixtures. When the translation method was used, the AAD of mixtures was improved to 0.69%. The proposed method was found to be better or comparable to other models in predicting densities.

This publication has 39 references indexed in Scilit: