Study of binary adsorption and diffusion in microporous materials with single crystal membrane technique

Single Crystal (100 x 100 x 300 μm) Membrane (SCM) technique has been extended to wider temperature and pressure ranges to study multicomponent micropore diffusion mechanisms in microporous materials. The permeation through the membranes was monitored by an online mass spectrometer, which is capable of monitoring flow rates as low as 10−12 mol/s. Fluxes of both pure components and binary mixtures of methane/n-butane at different temperatures (35°C and 100°C), concentrations (10%, 50%, 70% mole n-butane) and feed side pressures (10–740 torr) were measured.; The experimental system includes the mass spectrometer, single crystal membranes capable of stopping TIPB (triisopropanebenzene 9.3 Å) from permeating, diffusion assembly and accessories. The mass spectrometer abundance was calibrated by gas flow through a capillary of 110 μm in diameter and 5 meter in length. The corrected intracrystalline diffusivities extracted from the pure component permeation data are consistent with the reported results measured by the same SCM technique for both components. Simulation programs for the multicomponent diffusion through single crystal membranes were written in Matlab using the phenomenological adsorption diffusion model and shooting method. The simulation programs allow us to compare the model predicted permeation results with the experimental results.; The phenomenological adsorption diffusion model can explain the pure component permeation results, but not the binary permeation results. The pure component intracrystalline diffusivities for methane and n-butane calculated from pure component flux are 1.22 × 10−8 and 3.77 × 10−11 m2/s respectively at 35°C and 40 torr feed side pressure. The phenomenological adsorption diffusion model has been the model used by many researchers to explain their membrane permeation results. Both consistent and inconsistent results have been reported in literature. It is generally expected that the strongly adsorbed component (n-butane in this work) should reduce the permeation of the weakly adsorbed component (methane in this work) when permeating together based on preferential adsorption of the membrane surface. Our experimental results show that the two components (n-butane and methane) nearly do not influence each other when they permeate together through the single silicalite crystal membrane. To bridge the inconsistency between the experimental results and the predictions from the phenomenological adsorption diffusion model one will need to develop a more advanced technique, or develop a method that can monitor the local adsorbate concentration. No explanation for the inconsistency between the experimental results and the phenomenological predictions is offered in this dissertation.