Simulating gas–liquid−water partitioning and fluid properties of petroleum under pressure : implications for deep-sea blowouts

© American Chemical Society, 2016. This article is distributed under the terms of the AuthorsChoice License. The definitive version was published in Environmental Science & Technology 50 (2016): 7397–7408, doi:10.1021/acs.est.5b04617.With the expansion of offshore petroleum extraction, validated models are needed to simulate the behaviors of petroleum compounds released in deep (>100 m) waters. We present a thermodynamic model of the densities, viscosities, and gas–liquid−water partitioning of petroleum mixtures with varying pressure, temperature, and composition based on the Peng–Robinson equation-of-state and the modified Henry’s law (Krychevsky−Kasarnovsky equation). The model is applied to Macondo reservoir fluid released during the Deepwater Horizon disaster, represented with 279–280 pseudocomponents, including 131–132 individual compounds. We define >n-C8 pseudocomponents based on comprehensive two-dimensional gas chromatography (GC × GC) measurements, which enable the modeling of aqueous partitioning for n-C8 to n-C26 fractions not quantified individually. Thermodynamic model predictions are tested against available laboratory data on petroleum liquid densities, gas/liquid volume fractions, and liquid viscosities. We find that the emitted petroleum mixture was ∼29–44% gas and ∼56–71% liquid, after cooling to local conditions near the broken Macondo riser stub (∼153 atm and 4.3 °C). High pressure conditions dramatically favor the aqueous dissolution of C1−C4 hydrocarbons and also influence the buoyancies of bubbles and droplets. Additionally, the simulated densities of emitted petroleum fluids affect previous estimates of the volumetric flow rate of dead oil from the emission source.This research was made possible by grants from the NSF (OCE- 0960841, OCE-1043976, and EAR-0950600) and the Gulf of Mexico Research Initiative (GoMRI) to the C-IMAGE and DEEP-C consortia