A New High-Pressure High-Temperature Apparatus for Phase Behaviour Measurements on Multicomponent Mixtures
Saif Zahir Al Ghafri, Emmanuel C Efika and J. P. Martin Trusler
Qatar Carbonates and Carbon Storage Research Centre, Department of Chemical Engineering, Imperial College London, South Kensington Campus, London SW7 2AZ
* Corresponding author e-mail: firstname.lastname@example.org.
Understanding of the phase behavior of CO2-hydrocarbon mixtures at reservoir conditions is essential for the design, construction and operation of carbon capture and storage (CCS) and enhanced oil recovery (EOR) processes. In order to model these processes quantitatively, it is necessary to know the phase behavior and physical properties of the mixtures formed between CO2 and reservoir fluids under the conditions prevailing in the reservoir and, for this purpose, compositional thermodynamic models are required. There is a great deal of interest in developing improved models and computational packages to predict the phase behavior and thermodynamic properties of such mixtures with the least inputs of experimental data. In pursuit of this objective, a wide variety of approaches have been developed, ranging from empirical and semi-empirical equation of state (EoS), such as cubic equations, to molecular-based models such as the Statistical Associating Fluid Theory (SAFT). Nevertheless, experimental data are still required to tune the interaction parameters of such models, to help in developing predictive approaches, and to assess the predictive capabilities of all models. While equilibrium data for binary CO2-hydrocarbon mixtures are plentiful, equilibrium data for ternary and multi-component CO2-hydrocarbon mixtures are limited, especially for systems containing heavy hydrocarbons and/or hydrocarbons other than alkanes. Therefore providing new experimental data for multi-component mixtures at reservoir conditions is of great value.
2. Apparatus Design
Various techniques have been employed to determine the phase behaviour of multicomponent mixtures under high-pressure and high-temperature conditions. The classification of these techniques depends mainly on how composition is determined. In the analytical method, the compositions of the coexisting bulk phases are determined whereas, in the synthetic method, only the overall composition is determined experimentally. Multi-component mixtures, especially those containing heavy components, can be difficult to analyse and so the proposed apparatus implements a synthetic method in which mixtures of precisely known composition are prepared and their phase behaviour observed visually in a variable-volume cell. This mode of operation permits the determination of various types of phase boundary including vapour-liquid, liquid-liquid and vapour-liquid-liquid loci, critical curves of mixtures, solid-fluid equilibria, and cloud curves. However, in addition, the apparatus was fitted with phase sampling devices and on-line gas chromatography so that the analytic method may also be adopted.
The apparatus was designed for a maximum working pressure and temperature of 50 MPa and 473.15 K respectively permitting both synthetic and analytic approaches. The apparatus comprises a high-pressure high-temperature variable-volume view cell, with wetted parts fabricated from titanium, driven by a computer-controlled servo motor system, and equipped with a sapphire window for visual observation. The equilibrium cell is a horizontally-orientated cylindrical vessel with volume variable from approximately (11 to 67) cm3. The cell was equipped with five high pressure ports accommodating gas and liquid inlet line, an outlet line, two sampling ports, and temperature sensor. The outlet line was fitted with a rupture-disc safety device. The equilibrium cell was encased in an aluminium heating jacket to control the temperature. The jacket was fitted with axial holes to accommodate cartridge heaters and additional Pt100 temperature sensors. A PID process controller was used to regulate the temperature. Mixing of the cell contents was accomplished by means of a magnetic stirrer bar placed inside the cell. Four calibrated syringe pumps with a maximum service pressure of 70 MPa were used for quantitative fluid injection. One pair of pumps was configured to accept high-pressure gases or liquefied gases, while a second pair was configured to accept components that are liquid under ambient conditions. A low-dead-volume pressure transducer was used for the pressure measurements while the cell temperature was measured by means of a Pt100 sensor. The phase behavior of the mixture in the cell was observed with the aid of a CCD camera operating with both front and rear illumination.
Rolsi sampling devices were fitted to the cell with sampling capillaries designed to allow sampling of the lower and upper phases of a multi-phase fluid system. The sampling valves were connected directly to the gas chromatograph by means of heated transfer lines, and a valve within the GC was used to select between the two samplers. The amount of sample to be withdrawn was determined by the opening time of the Rolsi sampler which was set by means of the process controller provided. The process controller also provides for the automatic injection of a sequence of samples into the GC so that the repeatability of the composition measurements was verified. The gas chromatograph- supplied with hydrogen and zero-air gas generators- was fully customisable packed-column system provided with multiple means of calibration. The system was set-up with a three-column arrangement coupled to parallel thermal conductivity and flame ionisation detectors.
The entire apparatus was fully automated and key aspects of experiment control and data acquisition were operated by means of the custom software.
3. Experimental and Modeling
The apparatus was calibrated and validated by comparison with published isothermal vapour-liquid equilibrium data for different binary systems. The vapour-liquid phase behaviour of the mixture (CO2 + N2 + propylcyclohexane ) and its binaries was measured over the temperature range (323 and 423) K and pressures up to the critical pressure. The molar ratio between carbon dioxide and nitrogen in the ternary system was fixed at different values, and the bubble-curve and dew-curve was measured under carbon dioxide addition along three isotherms. The measurements were obtained using both synthetic and analytical approaches.
In this work, we also explore the predictive capability of SAFT-g-Mie  in relation to the phase equilibria of the mixture. In this approach, the statistical associating fluid theory, stemming originally from the perturbation theory of Wertheim , is implemented such that complex molecules are modelled by fused spherical segments representing the functional groups from which the molecule may be assembled. Generalized Mie potentials are used to represent the segment-segment interactions and all interactions parameters, both like and unlike, as implemented in this work, were previously determined using available experimental data (mainly pure component data) relating to systems comprising the constituent groups.
- V. Papaioannou, T. Lafitte, C. Avendano, C.S. Adjiman, G. Jackson, E.A. Muller, A. Galindo, Group contribution methodology based on the statistical associating fluid theory for heteronuclear molecules formed from Mie segments, J. Chem. Phys, 140 (2014).
- M. S. Wertheim, Fluids with highly directional attractive forces. II. Thermodynamic perturbation theory and integral equations, J. Stat. Phys, 35 (1984).
We gratefully acknowledge the funding of QCCSRC provided jointly by Qatar Petroleum, Shell, and the Qatar Science and Technology Park, and their permission to publish this research.
Keywords: Phase behavior; modeling; CO2; hydrocarbon;; variable volume cell, SAFT, propylcyclohexane, + 1-phenylhexane