Single gas permeation of thin zeolite (MFI) membranes: theory and analysisof experimental observations

Authors
Citation
Aj. Burggraaf, Single gas permeation of thin zeolite (MFI) membranes: theory and analysisof experimental observations, J MEMBR SCI, 155(1), 1999, pp. 45-65
Citations number
36
Categorie Soggetti
Chemistry,"Chemical Engineering
Journal title
JOURNAL OF MEMBRANE SCIENCE
ISSN journal
03767388 → ACNP
Volume
155
Issue
1
Year of publication
1999
Pages
45 - 65
Database
ISI
SICI code
0376-7388(19990331)155:1<45:SGPOTZ>2.0.ZU;2-N
Abstract
Theoretical expressions for single gas permeation are analysed and evaluate d with selected literature and some new experimental data on Silicalite/ZSM -5 membranes. The phenomenological sorption-diffusion (PSD) description (an d its equivalent Maxwell-Stefan equation) covers both the microscopic model s based on configurational diffusion (CD) with delta=1.0 and on surface dif fusion (SD) with delta less than or equal to 1.24. The ratio delta of the p ore diameter over the (kinetic) molecular diameter is important. For 1.24 l ess than or equal to delta less than or equal to 3.0 the gas translational diffusion (GT) or activated Knudsen Model is shown to be the preferred micr oscopic model, which is not covered by the PSD description. The combination of GT and SD mechanisms which is called surface diffusion enhanced micropo re permeation (SEMP) model is the only model able to describe the experimen tally observed minima in the flux of hydrogen, argon and krypton as a funct ion of temperature as well as the experimentally observed combination of a maximum followed by a minimum in the flux versus temperature curve (methane , ethane). Maxima in the flux versus temperature curves were observed for e .g. C-1-C-4 alkanes and carbon dioxide and could be described by PSD as wel l as SEMP models. The observed reversal of the order of the flux magnitudes with increasing temperature and in the activation energies of diffusion in the series Ne>Ar>Kr>Xe at given values of delta close to unity is predicte d by the SEMP model. (C) 1999 Elsevier Science B.V. All rights reserved.