Mutual Solubility of Water and Hydrocarbons

The accurate modeling of systems containing water and hydrocarbons is important to support key decisions related to the design, simulation, and optimization of a variety of industrial processes, ranging from refineries, to gas plants and liquified natural gas processing facilities. Processes of interest for the production of hydrocarbons now include widely different temperatures and pressure ranges, making the use of empirical models to simulate the behavior of water and hydrocarbon mixtures awkward and prone to inconsistencies. In this work we show that the use of the Peng-Robinson equation of state using the Huron-Vidal mixing rule combined with the nonrandom two-liquid model and temperature-dependent interaction parameters provide an accurate platform to correlate mutual solubility data for a variety of hydrocarbons. Moreover, the interaction parameters were correlated on the basis of simple molecular descriptors such as molecular weight and the paraffin, iso-paraffin, olefin, naphthene, and aromatic (PIONA) chemical family classification and Watson-K factor. The model shows an absolute average error in water mole fraction in the hydrocarbon phase equal to 34% and an absolute average error in the hydrocarbon mole fraction in the aqueous phase equal to 98% using the PIONA-based parameters and an absolute average error in water mole fraction in the hydrocarbon phase equal to 34% and an absolute average error in the hydrocarbon mole fraction in the aqueous phase equal to 148% using the Watson-K factor based parameters. The method can be used for systems defined using pure or pseudocomponents and is easily integrated within the structure of existing process simulators.