Phase equilibria studies on chemical mixtures encountered in the natural gas industry.

Abstract

Natural gas processing involves removing impurities from the gas streams. These impurities include carbon dioxide, nitrogen, hydrogen sulphide, water vapour, mercury, and others. These impurities must be eliminated from the gas streams, often using solvents, to meet sales specifications, enhance calorific value, lessen corrosion and blockages in pipelines due to hydrate formation and to allow for cryogenic gas processing. Solvents such as methanol and the lower molecular weight glycols have the most suitable characteristics to be employed as hydrate inhibitors, whilst 2,2′-[Ethane-1,2-diylbis(oxy)] di(ethan-1-ol) (triethylene glycol (TEG)) is mostly used in dehydration plants. In this study, phase equilibria data for mixtures of six chemical species commonly encountered in the processing of natural gas were studied. Phase equilibrium measurements were performed using a combined static (synthetic or analytic) apparatus. The apparatus comprises a horizontal cylindrical sapphire tube fitted with a movable piston that can be used to adjust the cell volume, thereby fixing/controlling the pressure in the process. A mobile Rapid Online Sampler Injector (ROLSI™) was fitted to the equilibrium cell for sampling both the vapour and the liquid phases. Vapour- liquid equilibrium (TPxy) data were measured and modelled for the following test systems, carbon dioxide + n-hexane and carbon dioxide + n-decane over a temperature range of 313.15 to 319.23 K. Bubble point (TPx) data were measured and modelled for the following test systems: carbon dioxide + methanol; carbon dioxide + TEG; methane + methanol; methane + TEG; carbon dioxide + aqueous TEG systems over a temperature range of 298.10 to 323.15 K. Generally, good agreement was observed between the reported literature data and the experimental data measured in this work, thus validating the experimental techniques used. New TPx data were measured and modelled for seven novel systems of this study, namely: methane + propane + methanol; methane + propane + TEG; methane + methanol + TEG; carbon dioxide + methanol + TEG; methane + propane + methanol + TEG; methane + propane + methanol + water + TEG; methane + propane + carbon dioxide + methanol + water + TEG over a temperature range of 283.15 to 323.15 K and in selected composition regions. The composition ranges and conditions are typical of those found in gas pipelines and gas dehydration units. The experimental data were modelled in Aspen Plus V11-12 using appropriate thermodynamic models, i.e., Peng Robinson (PR), Soave-Redlich-Kwong (SRK), Peng Robinson Wong Sandler (PRWS), Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT), and the Cubic Plus Association (CPA) models. The maximum absolute average relative deviation (AARD) in pressure on all the modelled data were 4.89%, 8.67%, 7.39%, 9.63% and 19.7% for the PRWS, SRK, CPA, PR, and PCSAFT models, respectively, indicating that the PRWS model best described most of the systems. The measured data contributes to the information required for the process design, control and monitoring of methanol and/or TEG in gas conditioning systems. Furthermore, the data helps refine thermodynamic models that can predict phase behaviour in multicomponent systems in applications mentioned earlier, including gas hydrate inhibition, subsea gas processing, carbon capture, and storage.