Application of cubic equations of state to solid-liquid phase equilibrium of mixtures of hydrocarbons and related materials.

摘要

The application of cubic equations of state (CEOSs) for the prediction of solid-liquid phase equilibria of hydrocarbon mixtures is presented. Crude oil and fuels are examples of mixtures of light and heavy hydrocarbon components. The light hydrocarbon components in oil keep the heavy components soluble in solution. However, the cooling of oil during transportation, processing or use, below the solidification point of the heavy components leads to the formation of a solid wax phase. A number of approaches have been proposed for the prediction of solid phase mixtures in both crude oils and fuels. Some of these models assume ideal liquid solution while others modify the regular solution theory or the Flory-Huggins model, with varying degrees of success. The absence of a usable thermodynamic model in the petroleum industry to describe both the liquid phase as well as solid phase non-idealities of mixtures of hydrocarbons with large size differences is still a major problem.; In this study a new cubic equation of state solid phase alpha function has been developed. The new alpha function when used in the TST/TSTSLE EOS model and together with the Twu vapor-liquid alpha function predicts reasonably good solid vapor pressures and solid molar densities, particularly for light molecular weight compounds. The solid phase thermodynamic properties, H and Cp from the TSTSLE equation agree reasonably well with experimental data for light molecular weight components at the melting point temperature. The TST infinite pressure mixing rules were found to predict and correlate the solubilities of binary alkane mixtures within acceptable tolerances. The optimum SLE value for the NRTL alpha parameter was found to be alpha12 = 0.2.; The new equation of state model was applied to binary mixtures and was shown to predict, with good accuracy, the solid solubility of heavy hydrocarbons in light hydrocarbon solvents over the whole composition range. For ternary and multi-component mixtures composed of library components of the normal paraffins up to nC30, the TST infinite pressure mixing rule (TSTMR)/TSTSLE model predicts reasonably well the solubility, the phase fraction, the wax appearance temperature and the fraction of paraffins crystallized in different solvents. For mixtures with compounds where the heat of fusion, the melting point temperature and the solid transition properties need to be estimated, a simple linear expression in molecular weight was found to estimate the combined transition enthalpies of heavy hydrocarbons reasonably well. In addition to the accurate property estimates provided by this new expression, the use of the new correlation in the TSTMR/TSTSLE model resulted in good temperature predictions at the cloud point (CPT) and the correct trends that are characteristic of the complex bimodal synthetic wax mixtures. The Won-Hansen correlations when used in the TSTMR/TSTSLE model provided good CPT predictions at elevated pressures for dead and live oils. (Abstract shortened by UMI.)