Engineering nanocomposite polymer membranes for olefin/paraffin separation

In this dissertation, I have investigated applying the laser ablation of microparticle aerosol (LAMA) process to the production of nanocomposite polymer membranes for olefin/paraffin separation. Experimental results for three major thrusts are presented: 1) an investigation into the scalability of the LAMA process, 2) a new laser ablation technique for nanoparticle production from aqueous feedstocks, and 3) characterization of olefin-selective polymer nanocomposite membranes produced using LAMA. The propensity for Ag nanoparticles to form agglomerates in LAMA is investigated. Nanoparticle samples were collected on TEM grids at several feedstock aerosol densities. As the density increased, the particle morphology shifted from single nanoparticles 5 nm in diameter to chained agglomerates of 20 nm diameter primary particles. The results are in agreement with a numerical model of Brownian agglomeration and diffusion. Factors influencing nanoparticle morphology, such as temperature, initial nanoparticle charge, and feedstock aerosol density are discussed. It is shown that agglomeration occurs on a much longer timescale than the other processes, and can be treated independently. A new nanoparticle synthesis technique is presented: laser ablation of aqueous aerosols. A Collison nebulizer is used to generate a mist of ~10 [mu]m diameter water droplets containing dissolved transition metal salts. Water from the droplets quickly evaporates, leaving solid particles which are ablated by an excimer laser. Ablation results in plasma breakdown and photothermal decomposition of the feedstock material. For AgNO₃ ablated in He gas, metallic Ag nanoparticles were produced. For Cu(NO₃)₂ ablated in He gas, crystalline Cu₂O nanoparticles were produced. For Ni(NO₃)₂ ablated in He gas, crystalline NiO nanoparticles were produced. A combination of AgNO₃ and Cu(NO₃)₂ ablated in a reducing atmosphere of 10%H₂/He yielded nonequilibrium Ag-Cu alloy nanoparticles. Membranes composed of poly(ethylene glycol diacrylate) (PEGDA) and Ag nanoparticles were produced by the LAMA process. Permeation and sorption measurements for the light olefins and paraffins were conducted for these membranes. The membranes showed very little improvement in olefin/paraffin selectivity compared with neat PEGDA membranes. Using the LAMA implementation described here, it was impossible to produce membranes with high Ag loading. Whether membranes containing more Ag would exhibit improved selectivity remains an open question.