Why is it so extremely difficult to prepare shape-selective Al-rich zeolite membranes like LTA and FAU for gas separation?

Two possible reasons for the difficulty to prepare shape-selective Al-rich zeolite membranes for gas separation are discussed. One reason is the strongly negative electric surface charge of Al-rich zeolites like ETA or FAU in aqueous media as found by zeta potential (ZP) measurements of suspended zeolite powder. This negative surface charge is expected to prevent the negatively charged silica species to enter the slits between the growing crystallites in a zeolite membrane layer. This effect can be avoided by adsorption of positively charged species called intergrowth supporting substance (ISS) neutralizing the electric surface charges of the growing zeolite membrane surface. Another problem for M-rich membranes are extreme mismatches in the expansion coefficients between the zeolite layer and the ceramic or metal support upon drying the as synthesized membranes. The as synthesized FAU or LTA membrane layers contain water. We have simulated a membrane activation in the permeation cell by drying zeolite powder and recording in situ the XRD spectra. We found huge expansion and shrinking effects during the in situ de-watering of LTA and FAU powder samples. This observation is correlated with the experimental finding continuous damage of model membranes upon repeated re-activation cycles in vacuum at 150 degrees C. The model membranes consisted of a few large LTA and FAU crystals embedded into epoxy resin. After each de-watering step, (i) the fluxes of CO(2), CH(4) and N(2) increased and (ii)the ideal selectivities decreased which indicates a continuous membrane damage. However, these semi-damaged LTA and FAU membranes can still separate water from alcohols but do not show any gas separation. (C) 2008 Elsevier B.V. All rights reserved.