Assumptions for Membrane Separation Simulation

In order to simulate an HMD system, the following assumptions on membrane separation 

Membrane is assumed to be of carbon molecular sieve (CMS) type due to its ability to handle high pressures and large feed flowrates without any plasticization, making it viable for use in an ethylene plant (Vu et al, 2002). Membrane performance is assumed to not deteriorate within its expected life of a conservative 3 years, compared with 5 years in the study of Ploegmakers et al. (2013). Compressor and heat exchangers are assumed to have a life of 10 years 

Membrane permeance and selectivity data are based on those of polymeric membranes obtained from Faiz and Li (2012). This is due to a lack of studies on CMS membranes for the separation of olefins/parafifins. The assumption of the use of polymeric membrane data is conservative since CMS membranes have been shown to give better permeance and selectivity than polymeric membranes for olefln/paraffin separation (Xu et al., 2012) 

In general, when selectivity is plotted against the permeance for a given olefln/paraffin set on a log-log plot, a linear upper bound with negative slope can be observed, as mentioned in Burns and Koros (2003) and Rungta et al. (2013). This leads to the following correlation between selectivity and permeance [Equation (10.8)]  

Q is the permeance of fast-permeating component A, is the selectivity of A shown by the membrane as compared with slow-permeating component B, and and are, respectively, front factor and slope of the linear upper bound. The parameter 

---