Parametric Study of the Membrane Process for Carbon Dioxide Removal from Natural Gas

Abstract

In this work, the cross-flow model is compared with the cocurrent flow model and the countercurrent flow model for the process of removing CO2 from natural gas. According to the comparison result, the cross-flow model is used for the analysis of the process of CO2 removal from the natural gas. Analysis results indicate that the membrane area is not a monodrome function of the selectivity and both the effects of the fast gas (CO2) permeability and the slow gas (CH4) permeability on membrane area and CH4 recovery are studied. Similar analysis is carried out for the operating pressures. Both the single stage system and the two-stage system with a recycle are studied. In order to reduce the membrane area required, membranes with high permeability and selectivity less than 100 are preferable. With increasing feed side pressure and decreasing permeate side pressure, the membrane area required decreases and the CH4 recovery increases. For the single-stage system using a commercial membrane with selectivity of 20, the CH4 recovery is lower than 90% at a product purity of 98%. Only when the membrane with selectivity higher than 50 is used, the separation of CH4 recovery >98% and product purity >98% target can be fulfilled by the single stage system. Using the two-stage system with a membrane selectivity of 20, the separation target can be achieved. In the two-stage system, the permeate gas of the first stage has to be compressed before it can be sent to the next stage. The compression energy required between stages is estimated to be 107.5 KJ/m3(STP) of feed for the process discussed in this paper. Even considering the conversion factor between heat and electricity of 3−4, the energy consumption of a two-stage membrane system is still much less than that of the amine absorption process which is about 600−900 KJ/(m3(STP) of feed).