MODIFIED BRUGGEMAN MODELS FOR PREDICTION OF CO2 PERMEANCE IN POLYCARBONATE/SILICA NANOCOMPOSITE MEMBRANES

The polycarbonate membranes performances are improved by the incorporation of silica nanoparticles into the polymer matrix. Prepared PC/silica nanocomposite membranes with various silica content are considered and the experimental relative permeance of CO2 gas at 200 kPa (2 bar) feed pressure are used as reference in the predictions using the existing theoretical models such as Maxwell-Wagner-Sillar, Bruggeman, Lewis-Nielsen, Pal, and modified Pal models. TEM and BET analysis were used to characterize the silica nanoparticles; FESEM was used to characterize the morphology of the nanocomposite membranes. The TEM image of the silica nanoparticles reveals that the nanoparticles are mostly spherical. BET results reveal that the surface area and pore diameter of the silica nanoparticles are 618.8 m(2)/g and 0.28 nm, respectively. The existing models have resulted in poor predictions with errors AARE % of 26.52 to 28.02 %. Observation by FESEM image shows that the dispersed particles are surrounded by interfacial voids and rigidified polymer layer. Modified Bruggeman models that consider the interfacial volume show appreciable prediction with AARE 4.59 % being obtained with the pseudo-two-phase Bruggeman model. Moreover, when the model considered pseudo-three-phase morphology, the AARE % value reduced to 3.92 %. Thus, the contribution due to the interfacial rigidified layer was minimal.