Document Type : VSI: Honoring NH
Department of Chemical Engineering, College of Engineering, University of Tehran, Tehran, Iran
University of Ottawa, Canada
The bubble point gas transport model is often used to determine the membrane pore size distribution due to its easiness to apply. The conventional approach, in which the experimentally obtained flow rate pressure curve (hereafter called Q-P curve) is analyzed using the methods proposed by Martínez et al. and Khayat et al., is effective but becomes less accurate for the distribution of small pores. Particularly, when there are two pore size distributions, the distribution calculated for the smaller pore sizes is not very reliable. The present work challenges this problem by splitting the overall Q-P curve into two components, one from the distribution of small pores and the other from the larger pores, while modeling the Q-P curve with the modified logistic sigmoid function (MLSF).
Once the overall Q-P curve is split into two components, the distribution parameters, i.e., mean pore size and standard deviations, can be obtained for each component distribution. The overall pore size distribution curve is then drawn by combining both distributions. The method is thoroughly described in this work and applied to the experimental Q-P data reported in the literature.