(576b) Evapoporometry – A Novel Method for Determining the Pore-Size Distribution of MF and UF Membranes

Accurate pore-size characterization of porous media including membranes can be critical for their performance in a wide range of applications.  Several methods for characterizing membrane pore-size distribution are available, but each has significant limitations. Multiple techniques are typically required to provide complete characterization of membrane porosity and pore-size distribution.

This paper describes a novel technique for characterizing the pore-size distribution of membranes.  We refer to this technique as evapoporometry (EP).  EP is based on the principle that the vapor pressure is affected by the curvature of a volatile liquid contained within the pores of a porous material.  This effect is described by the Kelvin equation, which has been shown to be applicable to pores as small as ~8 nm in diameter.  If a wetting volatile liquid is used to saturate the pores, the vapor pressure will be depressed, whereas if a non-wetting volatile liquid is used, the vapor pressure will be enhanced.  The first step requires that a porous sample be saturated with a suitable volatile liquid.  The sample is then placed in a specially designed test cell that is placed on a microbalance that permits measuring sample mass as a function of time.  The slope of the mass versus time curve provides the evaporation rate, which can then be related to the vapor pressure at the interface between the liquid in the porous material and the ambient gas phase.  The vapor pressure in turn can be related to the pore diameter via the Kelvin equation.  If the porous material is pre-saturated with a wetting volatile liquid, the evaporation rate will monotonically decrease as a function of time.  This behavior results because the liquid will evaporate progressively from the largest pores to the smallest pores since the vapor pressure decreases with decreasing pore diameter for a wetting liquid.

EP was utilized to characterize the pore-size distribution of representative flat sheet membranes often used as pore-size standards. These membranes included track-etched polycarbonate (Nuclepore^®) films with nominal pore diameters of 10, 30, 50, and 100 nm and porous bilayer aluminum oxide samples (Anopore^®) with nominal pore diameters of 20 and 100 nm.  For comparison, the membrane samples were also characterized by image analysis of micrographs obtained via scanning electronic microscopy (SEM). EP was also employed to analyze the pore-size characteristics of commercial ultrafiltration membranes. Results will also be presented for the determination of the pore-size distribution of hollow fiber membranes using EP.

The results of comprehensive testing indicated that the pore-size distribution obtained via EP compared favorably with SEM characterization as well as literature-reported values for the tested track-etched and ceramic materials.  EP provided accurate pore-size characterization for pore diameters within the range of ~10-150 nm.  The presentation will also describe the advantages and limitations of this new approach to pore-size characterization.