Abstract:
Mixed matrix membranes are of great importance for gas separation. Commonly used fillers for
MMMs are carbon molecular sieves and zeolites. Due to great technical importance of separation
by MMM, estimation of permeability is needed. Purpose of this study is to select and modify a
model for the permeation prediction of gas through mixed matrix membranes Permeation
estimation models are discussed in this work. Original models represent two phases, polymer
phase and an inorganic phase. By incorporating factors representing defects of mixed matrix
membranes results can be made closer to the experimental data. Based upon analogy between
parameters, models of thermal/electrical conductivities are applied to membrane permeation. By
incorporating factors representing membrane defects results of permeation are made closer to the
experimental values. Models are run by using MATLAB and results are compared. Hatta and
Taya model(1) is selected for permeation prediction. Possible defects in MMMs might be voids
formation, rigidification of polymer chains and the pore blockage. is Chain immobilization
factor also known as ratio of interface permeability to the matrix permeability and is a factor
representing ratio of interphase thickness to particle radius are introduced to the present model.
Chain immobilization is a factor which causes a reduction in permeability. Models are validated
by using experimental data of Matrimid-CMS, Ultem1000-CMS, BAPB-BPADDA (2,2 -bis(4-
aminophenoxy)biphenyl and bisphenol A tetracarboxylic dianhydride)/ Zeolie 4A. Maxwell
model, Lewis Nielson model, Baschirow model, Russell model, Hatta and Taya model and
present model are evaluated for different MMMs and a comparison is done for different gases at
different filler loadings. Average absolute relative error is calculated, present model shows the
relative error in the range of 3.25%-7.80%. Results showed that the present model is in good
agreement with the experimental.
