81 巻 (2016) 6 号 p. 469-478
Asphaltenes are the components of heavy fraction of oil. Once they deposit and clog the pore throats of reservoir rock, tubing pipes in production wells, and pipelines for oil transportation, it causes a serious problem for oil production. The objective of this paper is to show the applicability of “digital oil”, which is a full molecular model of a crude oil, and Molecular Dynamics (MD) simulation to solve the asphaltene problems. For this purpose, a digital oil of “A-crude oil” produced from an oil field actually confronting asphaltene problems was generated. On the generation of the digital oil, some analytical experiments were conducted and Quantitative Molecular Representation (QMR) method was applied. As a result, the digital oil was successfully generated. Then, asphaltene-asphaltene association energies in 7 different solvents: n-heptane, toluene, benzene, xylene, isopropyl alcohol (IPA), an azeotrope mixture of toluene and IPA, and a ternary azeotrope mixture of water, toluene, and IPA, were evaluated by calculating the Potential of Mean Force (PMF) using MD simulations and umbrella sampling technique. As expected, aromatics (toluene, benzene, and xylene) presents much lower association energies of asphaltene in absolute value than that in n-heptane. Among the 7 solvents, the ternary mixture provides the lowest association energy in absolute value. Therefore it can be the most effective solvent for the asphaltene of “A-crude oil”. Furthermore, it can be supposed from the result of this study that the association energy becomes lower in absolute value when the aromatic asphaltene solvents tend to distribute near around the edge of asphaltene molecule much more. Thus the digital oil and MD simulation would be a useful tool to solve the asphaltene problems in oil production.