Permeance ratio

For a defect-free ideal membrane, the selectivity is independent of thickness, and either permeability ratios or permeance ratios can be used for comparison of selectivi-ties of different materials. Nonideal module flow patterns, defective separating layers, impurities in feeds, and other factors can lower the actual selectivity of a membrane compared to tabulated values based on ideal conditions (Koros and Pinnau, 1994). 

Selectivity, often represented as the permeance ratio of two gases, is the capability of the membrane to separate a given gas mixture into its components [7], Selectivity is a measure of the membrane... 

Mol/m s (note that flux is represented not permeance). Ratio of single-gas permeances. 

The single gas permeance ratios N2/SF6 or FF/SFf, are often considered as representative of the zeolite membrane quality. However these values can be quite low even for high quality membranes [131]. Indeed, these ratios are affected by the defect distribution but can also depend on other factors such as the applied feed pressure, the film thickness and the support type. For example the N2/SF6 permeance ratio increases with pressure due to non-linear adsorption of SF6 and also increases with the film thickness due to the mass transfer resistance of the support. In the same way, the ideal or mixture selectivity /-butane/n-butane is often considered as a good indication of the membrane quality, but varies largely (from 10 up to 200) even for good quality membranes. It is then difficult to use only one of these criteria to compare zeolite membranes. 

Fig. 8.6. Pure O2/N2 permeance ratio of asymmetric poly(phenylene oxide) membranes as a function of surface tension of chloroform/nonsolvent additives mixtures. Nonsolvent additives include 2-ethyl-l-hexanol (1m), 1-octanol (2m), 2-propanol (3d), 2-decanol (4m), 3,5,5-trimethyl-1-hexanol (5m), 2,4-dimethyl-3-pentanol (6d), 2,4,4-trimethyl-1-pentanol (7d), 2-methyl-3-hexanol (lOd), 3-ethyl-3-pentanol (12m), and 2-methyl-2-hexanol (13d). Merged is indicated by m discrete is indicated by d. Reprinted from [22], with kind permission from J.Tan...

Asymmetric blend poly(ether sulfone)-polyimide (PES-PI) hollow fiber membranes were prepared at different air gaps and used for gas separation [23,24]. It was observed that the permeance of CO2 and N2 increased with an increase in air gap while the ideal selectivity decreased when the hollow fibers were uncoated. After silicone coating, no significant change was observed either in permeance or in the permeance ratio. Although the feed gas was supplied to the shell side of the hollow fiber, it seems that the skin layer is on the lumen side when the SEM pictures are examined. From AFM [25], it was observed that the mean roughness parameter of the inner surface (uncoated lumen side) increased with an increase in the air gap. 

Fig. 8.9. Pure O2/N2 permeance ratio versus etching time. Reprinted from [22], with kind permission from J. Tan...

Table 8.7 shows the 02/N2and CO2/CH4 permeance ratio for the as-cast P3AcET membrane and the base- or acid-treated membrane. As the table shows, the base or acid treatment resulted in a dramatic increase in the O2/N2 permeance ratio, while... 

Table 8.7. Permeance ratio for selected gas pairs and roughness parameters...

It should be however kept in mind that the membrane preparation procedure could influence its structure and gas transport properties. Thus, casting of the integrally skinned asymmetric membranes from p-DMePO solution using different nonsolvent additives produced the nodule structures in the surface skin layer of the membranes, which affected the permeance ratios for O2/N2 and CO2/CH4 [72]. In the homogeneous films of polyphenylene oxides considered in this chapter such structures apparently do not exist. 

Development of integrally skinned asymmetric membranes from polyphenylene oxide for gas separation to the author s knowledge dates back to 1973 when Kimura of General Electric Company in his patent work disclosed the method of such membrane preparation and use for air separation. He used a solution of PPO in chloroform (volatile and good solvent) and dichlorobenzene (non-volatile and poor solvent). The cast membrane was allowed to desolvate for 30 seconds after which it was precipitated in methanol. The dried membrane was tested for oxygen and nitrogen permeances. Kimura observed O2 permeance of 5.6 GPU with O2/N2 permeance ratio of 4.3. The patent did not mention about the molecular weight of PPO. 

Pure gas permeation data of membranes prepared using these additives were obtained from a constant pressure permeation system for CO2, CH4, O2 and N2. In order to know the effect of the structure of the nonsolvents, nonlinear regression analysis was attempted. Each additive was split into structural components groups. Several linear polynomial first and second order equations as well as nonlinear polynomial equations were attempted to derive an empirical correlation between the number of structural components and gas permeation data. A second order polynomial equation was derived to predict the pure CO2/CH4 permeance ratio from the structural components of the nonsolvents. From the structural studies the authors concluded that nonsolvent additives that possess a long straight hydrocarbon chain such as 2-ethyl-l-hexanol, 1-octanol and 2-decanol showed the highest pure gas permeance ratio. 

Morphology studies done by tapping mode atomic force microscope revealed that membranes prepared from a PPO solution with 2-ethyl-1-hexanol as the additive had the lowest mean diameter of nodules. Interestingly these were the membranes that exhibited the highest permeance ratio of 5.5 for O2/N2 24.5 for CO2/CH4 (second highest after 1-octanol). Pure gas... 

Memb. Type Solvent Solution concentration wt/wt % Permeance, GPU CO2 O2 Permeance ratio CO2/CH4 O2/N2 Ref. 

SPPO was first prepared by sulfonating PPO of intrinsic viscosity of 1.58 dL/g in chloroform at 2S C. Solutions of SPPOH in different solvents like ethoxyether and butoxyether were coated onto the surface of commercial polyethersulfone ultrafiltration membranes. All coated membranes were dried at 60° C for overnight. TFC membranes prepared from a solution of SPPO in ethoxyethanol demonstrated higher permeances and permeance ratios for CO2/CH4 gas pair. These membranes were immersed in solutions of alkali metal hydroxide or alkaline earth metal hydroxide of 0.1 to IN concentrations, depending on the solubility of the respective hydroxide in water. When the solubility was low, the solution saturated with hydroxide was used. The TFC membranes were kept immersed for 48 hours at room temperature to complete the exchange of the proton with metal cations. Solutions of magnesium nitrate and aluminium chloride were used to replace the proton with Mg and AF respectively. 

The effect of the metal cations on the performance of the cation-exchanged SPPO-PES TFC membranes is shown in Table 19. The permeance ratio of CO2/CH4 gas pair for TFC membranes in the divalent cation form is higher than that for the monovalent cation form when ions of similar sizes are... 

Among a series of either monovalent or divalent cations, the permeance ratio decreases as the ionic radius increases. Therefore, it may be inferred that there is a general correlation between the ionic charge density of the cation and the pure gas permeance ratio, i.e., the permeance ratio increases with an increase in the charge density of the cations. This is probably due to the enhancement of the electrostatic crosslinking between charged polymers and metal cations with an increase in the cationic charge density. 

Module Cone. Of SPPO, wt.% Solvent mixture composition, wt./wt.% Methanol Ethanol Nos. of coating layers Permeance, GPU CO2 CH4 Permeance ratio, CO2/CH4... 

The data shown in Table 22 describes the effect of concentration of SPPOBr solutions and the numbers of layers of coating on CO2/CH4 permeances and permeance ratios. Use of more dilute solutions and more... 

Module Surface area, cm Permeance, GPU Permeance ratio ... 

The membranes tested here showed very high permeance ratio for O2/N2 gas pair. The permeance for O2 was on an average of 10.5 GPU. The permeance ratio for CO2/CH4 gas pair in these cases was also high and close to the intrinsic permeability ratio of 59 in most of the membranes tested. 

Membrane modules were then left in an atmosphere of CH4 at 30 psig for one week after which the second set of tests was conducted. It was observed that the membrane permeance in the second test decreased approximately 20% but the permeance ratio remained unchanged. Previous study in the absence of water vapor in the feed stream had indicated a decrease in permeance ratio up to 80%. Therefore, improvement in the stability of the membrane selectivity with a wet feed stream is undeniable. 

It is evident from Table 29 that the ratio of carbon, oxygen, sulfur and bromine has not changed significantly in the used membranes. Therefore, the low values of permeance ratio and separation factor could be the result of leaks through the membrane modules. 

---