Metal-Organic Frameworks (MOFs) are nanoporous framework materials with tunable pore size and functionality, and hence attractive for gas separation membrane applications. Zeolitic Imidazolate Frameworks (ZIFs), a subclass of MOFs, are known for their high thermal and chemical stability. ZIF-8 has demonstrated potential to kinetically separate propane/propene in powder and membrane form. ZIF-8 membranes propane-propene separation performance is superior in comparison to polymer, mixed matrix and carbon membranes. The overarching theme of my research is to address challenges that hinder fabrication of MOF membranes on a commercial scale and in a reproducible and scalable manner. 1. Current approaches, are specific to a given ZIF, a general synthesis route is not available. Use of multiple steps for surface modification or seeding causes reproducibility and scalability issues. 2. Conventional fabrication techniques are batch processes, thereby limiting their commercialization. Here we demonstrate two new approaches that can potentially address these challenges. First, we report one step in situ synthesis of ZIF-8 membranes on more commonly used porous α-alumina supports. By incorporating sodium formate in the in situ growth solution, well intergrown ZIF-8 membranes were synthesized on unmodified supports. The mechanism by which sodium formate promotes heterogeneous nucleation was investigated. Sodium formate reacts with zinc source to form zinc oxide layer, which in turn promotes heterogeneous nucleation. Sodium formate promotes heterogeneous nucleation in other ZIF systems as well, leading to ZIF-7, Zn(Im)2 (ZIF-61 analogue), ZIF-90, and SIM-1 films. Thus one step in situ growth using sodium formate provides a simplified, reproducible and potentially general route for ZIF film fabrication. One step in situ route, although advantageous; is still conventional in nature and batch process with long synthesis time. This limits commercialization, due to scalability and manufacturing cost issues. Taking advantage of coordination chemistry of MOFs and using temperature as driving force, continuous well-intergrown membranes of HKUST-1 and ZIF-8 in relatively short time (15 min) using Rapid Thermal Deposition (RTD). With minimum precursor consumption and simplified synthesis protocol, RTD provides potential for a continuous, scalable, reproducible and commercializable route for MOF membrane fabrication. RTD-prepared MOF membranes show improved separation performances, indicating improved microstructure.

The goal of this project was to develop a mixed matrix membrane with enhanced properties for propylene/propane separations. To start with the project, one of the high performance 6FDA based polyimides was identified as the polymer matrix for the rest of the project. The chosen polymer (6FDA-6FpDA) was successfully synthesized in the laboratory. During the synthesis process the key objectives for high molecular weight and low polydispersity index polymer were identified. High molecular weight 6FDA-6FpDA was achieved via laboratory synthesis and was tested successfully. After successful synthesis of the high performance polymer, pure polymer dense films were tested for transport properties. One problem identified with 6FDA-6FpDA polymer films for propylene/propane separations was plasticization. A major objective of this research was to develop a method for plasticization suppression. A carefully controlled annealing procedure with high temperature permeation experiments was used in this research to suppress plasticization in a mixed gas environment. To the best of our knowledge, this is for the first time plasticization suppression was achieved with pure polymeric membrane material for propylene/propane separations with pure and mixed gases. The observed mixed gas experimental selectivity was lower than the pure gas selectivity which was explained by the combination effect of dual mode and bulk flow effect. The last objective of this project was to successfully incorporate molecular sieve materials to form a mixed matrix membrane hybrid material with enhanced transport properties First, an ideal molecular sieve for propylene/propane separation was identified and characterized. AlPO-14 was chosen for this research following its success with propylene/propane pressure swing adsorption. Mixed matrix membranes were successfully produced and tested for enhanced transport properties. Both pure and mixed gas results showed promising results with enhanced propylene permeability and propylene/propane selectivity. The experimental results were modeled with the Cussler and Maxwell models. A modified Cussler model was presented in this work. This is the first time an enhancement in the transport properties with mixed matrix membrane for propylene/propane separations has been observed. This fundamental dense film work holds a bright future for the scale up of propylene/propane separations.

abstract: Membrane-based gas separation is promising for efficient propylene/propane (C3H6/C3H8) separation with low energy consumption and minimum environment impact. Two microporous inorganic membrane candidates, MFI-type zeolite membrane and carbon molecular sieve membrane (CMS) have demonstrated excellent thermal and chemical stability. Application of these membranes into C3H6/C3H8 separation has not been well investigated. This dissertation presents fundamental studies on membrane synthesis, characterization and C3H6/C3H8 separation properties of MFI zeolite membrane and CMS membrane. MFI zeolite membranes were synthesized on α-alumina supports by secondary growth method. Novel positron annihilation spectroscopy (PAS) techniques were used to non-destructively characterize the pore structure of these membranes. PAS reveals a bimodal pore structure consisting of intracrystalline zeolitic micropores of ~0.6 nm in diameter and irregular intercrystalline micropores of 1.4 to 1.8 nm in size for the membranes. The template-free synthesized membrane exhibited a high permeance but a low selectivity in C3H6/C3H8 mixture separation. CMS membranes were synthesized by coating/pyrolysis method on mesoporous γ-alumina support. Such supports allow coating of thin, high-quality polymer films and subsequent CMS membranes with no infiltration into support pores. The CMS membranes show strong molecular sieving effect, offering a high C3H6/C3H8 mixture selectivity of ~30. Reduction in membrane thickness from 500 nm to 300 nm causes an increase in C3H8 permeance and He/N2 selectivity, but a decrease in the permeance of He, N2 and C3H6 and C3H6/C3H8 selectivity. This can be explained by the thickness dependent chain mobility of the polymer film resulting in final carbon membrane of reduced pore size with different effects on transport of gas of different sizes, including possible closure of C3H6-accessible micropores. CMS membranes demonstrate excellent C3H6/C3H8 separation performance over a wide range of feed pressure, composition and operation temperature. No plasticization was observed at a feed pressure up to 100 psi. The permeation and separation is mainly controlled by diffusion instead of adsorption. CMS membrane experienced a decline in permeance, and an increase in selectivity over time under on-stream C3H6/C3H8 separation. This aging behavior is due to the reduction in effective pore size and porosity caused by oxygen chemisorption and physical aging of the membrane structure. / Dissertation/Thesis / Doctoral Dissertation Materials Science and Engineering 2015

To date there exists a great deal of energetic and economic inefficiency in the separation of olefins from paraffins because the principal means of achieving industrial purity requirements is accomplished with very energy intensive cryogenic distillation. Mitigation of the severe energy intensity of the propylene/propane separation has been identified as one of seven chemical separations which can change the landscape of global energy use, and membranes have been targeted as an emerging technology because they offer scalability and lower capital and operating costs. The focus of this work was to evaluate a new direction of material development for the very industrially relevant propylene/propane separation using membranes. The objective was to develop a rational design approach for generating highly selective membranes using a relatively new platform of materials known as polyimides of intrinsic microporosity (PIM-PIs), the prospects of which have never been examined for the propylene/propane separation. Structurally, PIMs comprise relatively inflexible macromolecular architectures integrating contortion sites that help disrupt packing and trap microporous free volume elements (< 20 Å). To date most of the work reported in the literature on this separation is based on conventional low free volume 6FDA-based polyimides which in the best case show moderate C3H6/C3H8 selectivities (<20) with C3H6 permeabilities too low to garner industrial interest. Due to propylene and propane’s relatively large molecular size, we hypothesized that the use of more open structures can provide greater accessibility to the pores necessary to enhance membrane sieving and flux. It has been shown for numerous key gas separations that introduction of microporosity into a polymer structure can defy the notorious permeability/selectivity tradeoff curve and induce simultaneous boosts in both permeability and selectivity. The cornerstone approach to designing state of the art high performance PIM-PI membranes for the light gas separations involving maximizing the intra-segmental rigidity of the polymer chain was applied to the C3H6/C3H8 separation. A study regarding a stepwise maximization of intra-molecular rigidity and its effects on C3H6/C3H8 permeation was evaluated by conducting systematic structural modifications to high performance PIM-PIs. State of the art increases in performance were observed in pure-gas measurements as there were significant increases in C3H6/C3H8 selectivity and C3H6 permeability upon doing so. However, mixed-gas measurements showed that there were 65% losses in selectivity due to competitive sorption and mainly plasticization. Based on the conclusions drawn, a fundamental departure from conventional PIM design principles was used, instead emphasizing enhancing inter-chain interactions by introduction of a flexible diamine and functionalization with hydroxyl groups to attempt to immobilize the polymer chains. In doing so, the polymer chains may be able to pack more efficiently and upon sub-Tg annealing cause a microstructural reorganization to form a coplanarized configuration due to the combination of inter-chain charge transfer complexes (CTC) and hydrogen bonding networks. This approach successfully mitigated plasticization, but more importantly resulted in a tightening of the microstructure, especially in the ultra-microporous range (<7 Å) thereby yielding significant boosts in C3H6/C3H8 selectivity. Based on the PIM platform and novel polymer design approach thereof, the C3H6/C3H8 upper bound was thrust to new limits and led to the generation of the most selective solution processable polymers reported for the C3H6/C3H8 separation. Although the PIM platform has redefined the polymer upper bound, the permeability/selectivity tradeoff still endures, as the C3H6 permeabilities were on the order of 1 to 3.5 Barrer for the most selective polymers. To bridge that gap in permeability, several different approaches were taken. For the first time attempted for C3H6/C3H8 separation, high temperature heating of a PIM-PI to form thermally-rearranged and carbon molecular sieve membranes was employed. The TR membrane showed increased C3H6 permeability and about 50% losses in C3H6/C3H8 selectivity, while the CMS membrane formed at 600 oC showed modest gains in C3H6/C3H8 selectivity with significant improvements in C3H6 permeability. Finally, hybrid nanocomposite membranes incorporating a metal-organic framework structure into a PIM-PI matrix was used. ZIF-8, which has demonstrated high diffusive selectivities for C3H6/C3H8, was dispersed within the polymer, since previous work by the Koros group indicated that its incorporation into polyimide matrices can facilitate major improvements in both C3H6/C3H8 selectivity and C3H6 permeability compared to the respective neat polymer. Focus was directed towards attempting to improve polymer/nanoparticle adhesion by enhancing the interactions between the polymer and filler particles to mitigate the interfacial defects notorious in mixed-matrix membranes (MMM). To do so, ZIF-8 was dispersed into one of the best performing hydroxyl functionalized PIM-PI for the C3H6/C3H8 separation. The highest loaded mixed-matrix membrane in a glassy polymer to date of 65% (w/w) was achieved. The membranes showed pure-gas selectivities ranging from 34 with 10 Barrer at 30% loading to 43 with 38 Barrer at 65% loading. Strong performance and plasticization resistance were sustained in mixed-gas experiments even to feed pressures approaching the vapor pressure of the C3H6/C3H8 mixture, as selectivities well over 20 were achieved with high permeabilities, thereby demonstrating the potential commercial viability. Based on the work reported in this dissertation, we hope to help lay a framework to be able to tailor membrane performance and future membrane design to meet the demands of the different applications of the propylene/propane separation and hence show that there can be a marketplace for membranes in the separation. These include the debottlenecking of cryogenic distillation towers for production of polymer-grade propylene (99.5%) to reduce the associated extensive energy load, production of chemical-grade propylene (92-95% propylene), or for the recovery and recycling of olefins from reactor purges of petrochemical processes.

Olefin/paraffin separation is a large potential market for membrane applications. Carbon molecular sieve membranes (CMS) are promising for this application due to the intrinsically high separation performance and the viability for practical scale-up. Intrinsically high separation performance of CMS membranes for olefin/paraffin separations was demonstrated. The translation of intrinsic CMS transport properties into the hollow fiber configuration is considered in detail. Substructure collapse of asymmetric hollow fibers was found during Matrimidﾮ CMS hollow fiber formation. To overcome the permeance loss due to the increased separation layer thickness, 6FDA-DAM and 6FDA/BPDA-DAM polyimides with higher rigidity were employed as alternative precursors, and significant improvement has been achieved. Besides the macroscopic morphology control of asymmetric hollow fibers, the micro-structure was tuned by optimizing pyrolysis temperature protocol and pyrolysis atmosphere. In addition, unexpected physical aging was observed in CMS membranes, which is analogous to the aging phenomenon in glassy polymers. For performance evaluation, multiple "proof-of-concept" tests validated the viability of CMS membranes under realistic conditions. The scope of this work was expanded from binary ethylene/ethane and propylene/propane separations for the debottlenecking purpose to mixed carbon number hydrocarbon processing. CMS membranes were found to be olefins-selective over corresponding paraffins; moreover, CMS membranes are able to effectively fractionate the complex cracked gas stream in a preferable way. Reconfiguration of the hydrocarbon processing in ethylene plants is possible based on the unique CMS membranes.