Multiple Liquid Phases in a Natural-Gas System

Abstract

During a recent phase study of a natural gas, two stable equilibrium liquid phases were observed at temperatures below –200F and pressures above 200 psi. This paper reviews the published literature on the occurrence of multiple equilibrium liquid phases and presents analytical data for the vapor and two equilibrium liquid phases of the liquefied natural gas at five experimental conditions. In addition, data for 30 conditions of two-phase equilibria are included. Introduction The low-temperature phase behavior of gases, most of which contained helium, has been investigated in the laboratories of the Helium Activity for many years. Since 1952, experimental studies of these systems have been continuous as part of the research program at Amarillo, Tex. Because of their value to private industries interested in participating in the Helium Conservation Program, several "Open File" reports containing phase equilibria data for helium-bearing natural gases have already been released by the helium Activity. A paper containing information on the general phase behavior, operating criteria and extensive vapor-liquid data for two helium-containing systems was recently published. Additional publications presenting experimental data on the phase relationships of various gas systems are now in process and will be available in the near future. PREVIOUS EXPERIMENTAL WORK Although the formation of multiple liquids has been reported for various systems, to our knowledge this paper is the only substantiated evidence of a vapor-liquid-liquid equilibria in a naturally occurring gas. In 1940, Vink, Ames, and others reported the presence of two liquid phases in a hydrocarbon system consisting of mixtures of crude oils, solvents and natural gas. Eilerts and co-workers published data on the recombined fluids from a gas-condensate well. This condensed gas, containing approximately 76 per cent methane and 24 per cent ethane-plus, exhibited two distinct liquid phases. Weinaug and Bradley observed "unusual" phase behavior in a reservoir mixture. These workers postulated that the anomalous phase behavior was due to the "imminent formation of a second liquid phase". Botkin, Reamer, Sage and Lacey studied two California crude oils that exhibited multiple phases. Kohn and Kurata recently reported two equilibrium liquid phases in the methane-hydrogen sulfide system. Roof and Crawford and Eakin, et al, also have reported experiments with binary systems that formed two stable equilibrium liquid phases. APPARATUS AND PROCEDURE A U. S. Bureau of Mines Phase Equilibrium Apparatus was used in conducting this study. The apparatus and procedures employed in its operation have been previously described and will not be repeated in detail in this report. Briefly, the apparatus consists of a windowed cell which can be maintained within +/−0.5F for temperatures between room temperature and –320F. Pressure within the cell can be maintained within 0.1 per cent of gauge reading up to 800 psig. Equilibrium vapor and liquid samples are obtained in special containers for analysis by a mass spectrometer. Although the accuracy of the analyzer is about +/−0.1 mol per cent, the reproducibility of the phase-equilibrium apparatus is considered relatively Poor. Values reported from methane and nitrogen are considered accurate to +/−1.0 and +/−0.6 mol per cent. Data for ethane-plus in the vapor are accurate within 0.2 mol per cent; liquid-phase data for this aggregate component are accurate within 1.5 mol per cent. For helium in the vapor phase, the analytical data are accurate within 0.2 mol per cent; liquid-phase analyses for this component were obtained by the charcoal adsorption method described by Frost and are accurate within 0.006 mol per cent. All of these references to the accuracy of reported values are conservative estimates based upon a statistical treatment of reproducibility data obtained with the apparatus.

Modeling of the Multiphase Behavior of Nitrogen-Containing Systems at Low Temperatures with Equations of State

“…Finally, to investigate whether the PC-SAFT EoS provides good accuracy for multicomponent systems with only binary information, this model was applied to the modeling of the vapor-liquid equilibrium data for 43 multicomponent mixtures containing hydrocarbon (methane, ethane, propane, n-pentane, n-heptane, and n-decane) and nonhydrocarbon (nitrogen, carbon dioxide, and hydrogen sulfide) components, studied experimentally by Yarborough, 36 and the three-phase vapor-liquid-liquid equilibrium data of a natural gas system rich in nitrogen, reported by Stroud et al 7 The results of the modeling obtained for these systems indicate that the PC-SAFT EoS can be used with confidence to predict the phase behavior of multicomponent systems by using interaction parameters obtained only from experimental binary vapor-liquid equilibrium data.…”

“…1,2 Experimental VLLE data have been reported for the systems nitrogen + ethane, 3,4 nitrogen + propane, 4 and nitrogen + methane + ethane, [3][4][5] including an observation of four fluid phases (one vapor phase and three liquid phases) at T ) 112.2 K and p ) 1.403 MPa for this ternary system. 6 Stroud et al 7 also reported VLLE data for a nitrogen-rich natural gas system from T ) (116 to 130) K and pressures up to 3.447 MPa. Subsequently, data of the two coexisting liquid phases in binary and ternary mixtures of nitrogen with n-alkanes have been published by Kohn and co-workers.…”