Tyler L. Crawford and Dr. John D. Lamb, Chemistry and Biochemistry
Ion chromatography (IC) has a need for a suppressor which would allow the simultaneous conductitnetric detection of both cations and anions. Such a suppressor’s membrane would need to exhibit a large flux and a high level of selectivity. Polymer inclusion membranes (PIMS) containing macrocycles have been shown to exhibit high selectivity over a long period of time. The structure of the macrocycle 1 determines the affinity these Ugands exhibit toward cations. By completing metal salts, the macrocycles are able to extract ionic species into solvents that these metal salts would not be soluble in alone. The macrocycle can then transport a cation for which it has specific affinity through a nonpolar media to a receiving phase where the cation may be released. This process can be driven by employing a proton-ionizable macrocycle. However, such polymer inclusion membranes have not exhibited the high levels of proton flux which would be required in a suppressor membrane. Current suppressors employ sulfonated teflon membranes which allow enormous proton flux. The aim of my research has been to construct and investigate a composite membrane that could be utilized in a suppressor for IC. Such a composite membrane would need the high found in sulfonated teflon membranes and the selectivity of PIMS.
Composite membranes were formed from PIMs and sulfonated teflon. The sulfonated teflon membranes (Dionex, Sunnyvale, CA) were cut into 9 cm diameter circles or into squares 9 cm on a side. A containing vessel was constructed by applying silicone glue to the wide mouth end of a glass funnel and placing it on a sheet of plastic, so as to form a smooth surface which would seal. The teflon membrane was placed in a flat bottom petri dish with the containing vessel resting over it. PIMs were then cast over the sulfonated teflon membranes. In casting each PIM, 3.5 mL of CTA solution and 1.5 mL of a 10% vol. solution of 2-nitrophenyl octyl ether (Aldrich) in methylene chloride. The CTA solution was made by dissolving 2.5 g of CTA (Aldrich) into 200 mL of methylene chloride. PIMs were cast over both dry and water-swelled teflon membranes and allowed to cure for twenty-four hours.
A membrane’s proton flux was determined by placing it between two glass vessels which held the source and receiving phases. The source phase was 50 mM LIOH (Spectrum) and the receiving phase was 100 mM sulfuric acid (Mallinckrodt). Three bladed glass propellers in both the receiving and source phases stirred at 600 rpm over the course of the experiment. Three ml of receiving phase was removed after an hour, three hours, six hours and twenty-four hours. After each aliquot of receiving phase was removed, 3 ml of 100 mm sulfuric acid was replaced. The samples were analyzed for lithium against a standard curve by IC (Dionex 4000).
My first objective on this project was to cast a PIM on top of a piece of circular sulfonated teflon membrane which contained no macrocycle. Previous research had showed that a PIM without macrocycles exhibited no flux. I hoped to show that a composite membrane would likewise not transport ionic species from a source phase to a receiving phase because of its PIM component. I chose to perform a flux experiment on this composite membrane to see if it was permeable to lithium. I also performed a simultaneous flux experiment on plain sulfonated teflon membrane. I found that both membranes transported lithium at the same rate, indicating that the PIM portion of the composite membrane had been compromised. I concluded that the problem was probably due to the fact that the sulfonated teflon swelled and buckled as it came in contact with water. I made many attempts to cast the PIM on top of the sulfonated teflon membrane so as to avoid compromising the composite membrane. I investigated the viability of pulling the membrane taut, but I found that the composite membrane invariably buckled in the center when it got wet. I next wetted the membrane and attempted to stretch it flat, but I observed the same buckling phenomenon. I decided that the buckling may have resulted from the inability of the center portion of the membrane to expand to the same extent as its outer edges. Square-cut membranes buckled as well, though less than circularly cut membranes. I also stretched slightly dampened teflon membranes over paper media that delivered water by capillary action to keep it from shrinking during the casting procedure. Despite numerous attempts, I was not able to develop a viable casting procedure for making composite membranes.
During this time I was also able to research more of the literature and talk to Dionex scientists regarding the composite suppressor membrane. Besides the structural obstacles described previously, the sulfonated teflon membrane itself may prevent the use of a composite membrane based on these materials in future suppressors. The sulfonated teflon must face the basic element which carries the analyses because the PIM is unstable in base. The cations in the element could then diffuse into and be retained by the sulfonated teflon, thus causing enormous band broadening of these cations’ peaks. One possibility of getting around this problem would be to integrate the macrocycle into the teflon membrane itself. A plasticizer, such as 2-nitrophenyl octyl ether in methylene chloride, could absorb into the teflon membrane. If macrocycle was dissolved into the methylene chloride as well, it would be trapped within the membrane as the membrane cures. Unsulfonated teflon membrane would be required in order to lower the possibility of a nonpolar macrocycle partitioning out of the membrane. Such a membrane could be made extremely thin. I had requested unsulfonated membrane from Dionex, but we had not received any by the time this research was completed.
References
- A.J. Schow et. al., “Polymer inclusion membranes containing macrocycle carriers for use in cation separations,” Journal ofmembraneScience,in publication.
- Special thanks to the Dionex Corp. of Sunnyvale, CA for materials and technical advice, and the assistance of Dr. John Lamb and Travis
Niederhauser is thankfully acknowledged.
