br Model parameterization br Results and discussion br Concl
Results and discussion
Conclusion Following an approach previously proposed for SAFT-VR model, we have used PR-CPA EoS without explicitly accounting for the reactions for the modeling of CO2 and H2S-water-alkanolamine solutions. The reactions between acid gases and water are treated as strong associations by adding dedicated associations sites on water and acid gases. Adjustable parameters are obtained from some binary and ternary data and are in some cases temperature dependent. It can be concluded that PR-CPA EoS can accurately represent the solubility of acid gas in aqueous alkanolamine solutions over a wide range of conditions (temperatures and loadings). Less satisfactory results are obtained at high loadings. Furthermore, vapor phase compositions of importance to the estimation of solvent loss and the speciation in the liquid phase have been successfully predicted using the same adjustable parameters fitted to phase equilibria data. On the other hand, the estimation of enthalpy of Paclitaxel was only partially satisfactory, indicating that the temperature dependency of the model requires improvement. Finally, PR-CPA EoS has been further validated by correctly estimating multi-component systems containing acid gas mixture, alkanolamine, water, and methane. In conclusion, PR-CPA EoS provides a convenient platform for performing calculations for mixtures with acid gases, water and alkanolamine with few adjustable parameters and without use of extensive experimental data. The PR-CPA EoS could be integrated into simulation software such as Prosim® (Toulouse, France) or Aspen Plus™, in order to solve the challenges in process design in the context of acid gas treatment from natural gas.
Introduction Phase behavior models for mixtures of bitumens (or heavy oils) and n-alkanes are required for the simulation of solvent-assisted in situ heavy oil recovery processes, solvent based oil sand extraction processes, and solvent deasphalting processes [1,2]. These mixtures can form multiple phases including vapor-liquid (VL), liquid-liquid (LL), VLL, and possibly VLLL regions [, , , ]. Depending on the solvent and the conditions, the second liquid phase can be a light solvent-rich phase or a heavy asphaltene-rich phase. Asphaltene-rich phase formation, often termed precipitation, and subsequent deposition is also a flow assurance issue for some conventional oil applications including undersaturated live oil depressurization and carbon dioxide flooding [, , , ]. It has proven challenging to model the full range of crude oil/solvent phase behavior, particularly asphaltene-rich phase formation, with a single model and a consistent fluid characterization. Asphaltenes are defined as the fraction of a crude oil that is insoluble in a paraffinic solvent (usually n-pentane or n-heptane) and soluble in an aromatic solvent (usually toluene). They are the heaviest and most aromatic fraction of a crude oil with the highest density, molecular weight and heteroatom content [, , ]. Asphaltenes are known to self-associate into nano-aggregates consisting of 5–10 molecules on average [, , ]. The changes in this self-association at different temperature, pressures, and compositions and the effect of self-association on crude oil phase behavior are currently ill-defined. Most crude oil phase behavior modeling approaches do not explicitly consider asphaltene self-association. To facilitate the discussion of the modeling approaches, the following terms are defined: To date, the two most successful approaches for modeling asphaltene precipitation are regular solution models and equations of state (EoS). Regular solution models are particularly useful in matching the onset and yield of precipitated asphaltenes from oils diluted with different solvents or in blends [, , , ]. However, these models are limited to liquid-liquid or liquid-solid equilibria. Equations of state, on the other hand, can describe both vapor-liquid and liquid-liquid equilibria. They have been widely applied in the petroleum industry and are the cornerstone of modern chemical process simulators [21,22].