Understanding the equilibrium behaviour of a novel phase splitting solvent in the presence of high pressure CO2
Summary
The main objective of this research was to increase the knowledge of carbon capture solvents which form two phases when absorbing high pressure CO2, e.g. from natural gas fields. Based on the results of previous experiments, a literature study and already existing knowledge of carbon capture, a number of possible blends have been identified. During the screening experiments, the MAC was used to rapidly determine whether or not a blend forms two phases or not. The results indicate that besides water sulfolane, MDEA and a primary/secondary amine have to be present for the two phases to form. However, too little information is available to draw any solid conclusions.
After the screening experiments, high pressure test were conducted in the VLE to determine the loading and phase composition at various partial pressures. At random pressure intervals, samples were taken of either phase, which were analysed for e.g. composition and loading. The tested blend consisted of 10 wt% AEP, 25 wt% water, 30% MDEA and 35 wt% sulfolane. It was observed that the heavy (or physical) phase is predominantly sulfolane (>85 wt%), with small amounts of water and MDEA. Meanwhile, the light (or chemical) phase consists of all four components being present with fractions between 11 and 35 wt%. At a CO2 partial pressure of 1.500 kPa, over 4,1 mole of CO2 can be absorbed per litre of blend. The chemical (top) phase absorbed the majority of the CO2 (4,5 mole/L), while the physical (bottom) phase has a loading of ca. 1,56 mole/L.
Heating the chemical phase to 120 0C causes the majority of the absorbed CO2 to be stripped, as at 20 kPa the loading reduces to 0,16 mole L-1. Hence, the chemical phase has a cyclic loading of ca. 3,84 mole L-1; assuming a 700 kPa CO2 partial pressure and 30 kPa partial pressure in the stripper. The physical phase has – again 700 kPa – a cyclic loading of ca. 0,62 mole L-1. Combining these loadings gives a cyclic loadings of 3,19 mole L-1, which is a 10 % increase compared to an aMDEA blend being used at the same conditions.
Furthermore, the energy associated with this process is ca. 55 % less, compared to the aMDEA process, which implies that a significant reduction in energy consumption can be achieved. However, it should be noted that this statement is based on VLE figures which are at the limit of the apparatus; which means that some deviation might have been caused by this.
Also, the equipment which was used to determine the phase composition at various partial pressures (i.e. the GC and Karl Fischer titrations) proved to be unreliable. The sulfolane concentration in the physical phase and the water content of either of the phases could – for unknown reasons – not be determined accurately
Ultimately, the result of this thesis is that more is known about the composition of the two phases when CO2 is absorbed at high pressure and some estimates are made for the (energetic) performances of the blend. The knowledge of this blend can be further expanded by future research, focussing not only on improving the analytical methods used but also optimizing the blend.