Deasphalting of crude oils: a molecule-based model to explore an optimization approach and alternative solvents

Actual motivation to reduce environmental impact leads to an increase in the study of green technologies, even those used in processes with fossil fuels. However, far from the proposed reduction of such energies, heavier crude oils are being processed to fulfill the energy demand. Heavy crude oils require an upgrading process to mitigate the effect of noxious components or contaminants such as asphaltenes and heavy metals. In this work, a deasphalting process was studied by simulation as a pretreatment to eliminate the heavy compounds of crude oil. The simulation of such a deasphalting process was carried out by Aspen Plus® software by combining a model for crude oil, formed by 33 molecules and able to describe its complex composition, mainly in the asphaltene fraction, and a model for the phase equilibria that use liquid-liquid equilibrium equations and the modified UNIFAC (Dortmund) model.A new objective function, including extraction yield, asphaltene reduction, and the ratio solvent/crude oil, was proposed and used to optimize extraction conditions and compare the performance of several solvents. Up to 12 pure solvents of different nature (tert-butanol, n-butanol, iso-propanol, methanol, ethylene glycol, acetone, methyl-isobutyl ketone, per-fluoro heptane, anisole, 1-metil-2-pyrrolidone, sulfolane, and n-heptane) were checked as well as some binary mixtures of them to explore more sustainable alternatives against conventional paraffin as precipitating agents. Satisfactory results were obtained when alcohols or ketones were used. A deeper study was developed to understand the effects of the alcohols' molecular structure (branching, chain length, and molecular weight) as solvents. Higher branching reduces the solvent/oil ratio, while higher molecular weight leads to higher extraction yield but poorer asphaltene reduction.