Graphene oxide membranes have shown promising gas separation characteristics specially for hydrogen that make them of interest for industrial applications. However, the gas transport mechanism for these membranes is unclear due to inconsistent permeation and separation results reported in literature. Graphene oxide membranes made by filtration, the most common synthesis method, contain wrinkles affecting their gas separation characteristics and the method itself is difficult to scale up. Moreover, the production of graphene oxide membranes with fine-tuned interlayer spacing for improved molecular separation is still a challenge. These unsolved issues will affect their potential impact on industrial gas separation applications.

In this study, high quality graphene oxide membranes are synthesized on polyester track etch substrates by different deposition methods and characterized by XRD, SEM, AFM as well as single gas permeation and binary (H2/CO2) separation experiments. Membranes are made from large graphene oxide sheets of different sizes (33 and 17 micron) using vacuum filtration to shed more light on their transport mechanism. Membranes are made from dilute graphene oxide suspension by easily scalable spray coating technique to minimize extrinsic wrinkle formation. Finally, Brodie’s derived graphene oxide sheets were used to prepare membranes with narrow interlayer spacing to improve their (H2/CO2) separation performance.

An inter-sheet and inner-sheet two-pathway model is proposed to explain the permeation and separation results of graphene oxide membranes obtained in this study. At room temperature, large gas molecules (CH4, N2, and CO2) permeate through inter-sheet pathway of the membranes, exhibiting Knudsen like diffusion characteristics, with the permeance for the small sheet membrane about twice that for the large sheet membrane. The small gases (H2 and He) exhibit much higher permeance, showing significant flow through an inner-sheet pathway, in addition to the flow through the inter-sheet pathway. Membranes prepared by spray coating offer gas characteristics similar to those made by filtration, however using dilute graphene oxide suspension in spray coating will help reduce the formation of extrinsic wrinkles which result in reduction in the porosity of the inter-sheet pathway where the transport of large gas molecules dominates. Brodie’s derived graphene oxide membranes showed overall low permeability and significant improvement in in H2/CO2 selectivity compared to membranes made using Hummers’ derived sheets due to smaller interlayer space height of Brodie’s sheets (~3 Å).

An urgent need for developing new chemical separations that address the capture of dilute impurities from fluid streams are needed. These separations include the capture of carbon dioxide from the atmosphere, impurities from drinking water, and toxins from blood streams. A challenge is presented when capturing these impurities because the energy cost for processing the bulk fluid stream to capture trace contaminants is too great using traditional thermal separations. The development of sorbents that may capture these contaminants passively has been emphasized in academic research for some time, producing many designer materials including metal-organic frameworks (MOFs) and polymeric resins. Scaffolds must be developed to effectively anchor these materials in a passing fluid stream. In this work, two design techniques are presented for anchoring these sorbents in electrospun fiber scaffolds.

The first technique involves imbedding sorbent particles inside the fibers: forming particle-embedded fibers. It is demonstrated that particles will spontaneously coat themselves in the fibers at dilute loadings, but at higher loadings some get trapped on the fiber surface. A mathematical model is used to show that when these particles are embedded, the polymeric coating provided by the fibers may be designed to increase the kinetic selectivity and/or stability of the embedded sorbents. Two proof-of-concept studies are performed to validate this model including the increased selectivity of carbon dioxide over nitrogen when the MOF ZIF-8 is embedded in a poly(ethylene oxide) and Matrimid polymer blend; and that increased hydrothermal stability is realized when the water-sensitive MOF HKUST-1 is embedded in polystyrene fibers relative to pure HKUST-1 powder.

The second technique involves the creation of a pore network throughout the fiber to increase accessibility of embedded sorbent particles. It is demonstrated that the removal of a blended highly soluble polymer additive from the spun particle-containing fibers leaves a pore network behind without removing the embedded sorbent. The increased accessibility of embedded sorbents is validated by embedding a known direct air capture sorbent in porous electrospun fibers, and demonstrating that they have the fastest kinetic uptake of any direct air capture sorbent reported in literature to date, along with over 90% sorbent accessibility.

Separation of carbon dioxide and methane for the upgrade of natural gas through use of pressure swing adsorption could potentially save large amounts of energy from the current, costly process of cryogenic distillation and provides greater cost effectiveness for carbon dioxide capture, and provide larger product flowrates than membrane permeation separation. The purpose of this study is to analyze the effects of varying initial conditions of a MatLab simulation, courtesy of Mai Xu, a graduate student at ASU, designed to use Langmuir isotherms, mass transfer equations, and adsorbent and gas properties to simulate a pressure swing adsorption process with a mixture of methane and carbon dioxide gas feed. The effects that will be varied are the adsorption/desorption time, pressurization/depressurization time, adsorption feed composition, desorption purge composition, adsorption pressure, desorption pressure, adsorption flow rate, and desorption flow rate. The study found that the trends in methane purity and production generally follow the trends predicted by literature and relevant equations, with pressure boundaries being the largest impacting factor. In addition there was a markedly inverse correlation between purity of methane product and the productivity of the system. This trend was only violated in one instance, at very low vacuum pressure during desorption, which could indicate an area that requires further study. Overall, the main areas of improvement in pressure swing adsorption for this system would be improving the selectivity of adsorption of carbon dioxide over methane, which requires improvement and change of the adsorbent, and more extreme vacuum pressures during desorption, both of which will increase methane yield and reduce operating costs.

Membranes are a key part of pervaporation processes, which is generally a more

efficient process for selective removal of alcohol from water than distillation. It is

necessary that the membranes have high alcohol permeabilities and selectivities.

Polydimethylsiloxane (PDMS) based mixed matrix membranes (MMMs) have

demonstrated very promising results. Zeolitic imidazolate framework-71 (ZIF-71)

demonstrated promising alcohol separation abilities. In this dissertation, we present

fundamental studies on the synthesis of ZIF-71/PDMS MMMs.

Free-standing ZIF-71/ PDMS membranes with 0, 5, 25 and 40 wt % ZIF-71

loadings were prepared and the pervaporation separation for ethanol and 1-butanol from

water was measured. ZIF-71/PDMS MMMs were formed through addition cure and

condensation cure methods. Addition cure method was not compatible with ZIF-71

resulting in membranes with poor mechanical properties, while the condensation cure

method resulted in membranes with good mechanical properties. The 40 wt % ZIF-71

loading PDMS nanocomposite membranes achieved a maximum ethanol/water selectivity

of 0.81 ± 0.04 selectivity and maximum 1-butnaol/water selectivity of 5.64 ± 0.15.

The effects of synthesis time, temperature, and reactant ratio on ZIF-71 particle

size and the effect of particle size on membrane performance were studied. Temperature

had the greatest effect on ZIF-71 particle size as the synthesis temperature varied from -

20 to 35 ºC. The ZIF-71 synthesized had particle diameters ranging from 150 nm to 1

μm. ZIF-71 particle size is critical in ZIF-71/PDMS composite membrane performance

for alcohol removal from water through pervaporation. The membranes made with

micron sized ZIF-71 particles showed higher alcohol/water selectivity than those with

smaller particles. Both alcohol and water permeability increased when larger sized ZIF-

71 particles were incorporated.

ZIF-71 particles were modified with four ligands through solvent assisted linker

exchange (SALE) method: benzimidazole (BIM), 5-methylbenzimidazole (MBIM), 5,6-

dimethylbenzimidazole (DMBIM) and 4-Phenylimidazole (PI). The morphology of ZIF-

71 were maintained after the modification. ZIF-71/PDMS composite membranes with 25

wt% loading modified ZIF-71 particles were made for alcohol/water separation. Better

particle dispersion in PDMS polymer matrix was observed with the ligand modified ZIFs.

For both ethanol/water and 1-butanol/water separations, the alcohol permeability and

alcohol/water selectivity were lowered after the ZIF-71 ligand exchange reaction.

Membrane technology is a viable option to debottleneck distillation processes and minimize the energy burden associated with light hydrocarbon mixture separations. Zeolitic imidazolate frameworks (ZIFs) are a new class of microporous metal-organic frameworks with highly tailorable zeolitic pores and unprecedented separation characteristics. ZIF-8 membranes demonstrate superior separation performance for propylene/propane (C3) and hydrogen/hydrocarbon mixtures at room temperature. However, to date, little is known about the static thermal stability and ethylene/ethane (C2) separation characteristics of ZIF-8. This dissertation presents a set of fundamental studies to investigate the thermal stability, transport and modification of ZIF-8 membranes for light hydrocarbon separations.

Static TGA decomposition kinetics studies show that ZIF-8 nanocrystals maintain their crystallinity up to 200○C in inert, oxidizing and reducing atmospheres. At temperatures of 250○C and higher, the findings herein support the postulation that ZIF-8 nanocrystals undergo temperature induced decomposition via thermolytic bond cleaving reactions to form an imidazole-Zn-azirine structure. The crystallinity/bond integrity of ZIF-8 membrane thin films is maintained at temperatures below 150○C.

Ethane and ethylene transport was studied in single and binary gas mixtures. Thermodynamic parameters derived from membrane permeation and crystal adsorption experiments show that the C2 transport mechanism is controlled by adsorption rather than diffusion. Low activation energy of diffusion values for both C2 molecules and limited energetic/entropic diffusive selectivity are observed for C2 molecules despite being larger than the nominal ZIF-8 pore aperture and is due to pore flexibility.

Finally, ZIF-8 membranes were modified with 5,6 dimethylbenzimidazole through solvent assisted membrane surface ligand exchange to narrow the pore aperture for enhanced molecular sieving. Results show that relatively fast exchange kinetics occur at the mainly at the outer ZIF-8 membrane surface between 0-30 minutes of exchange. Short-time exchange enables C3 selectivity increases with minimal olefin permeance losses. As the reaction proceeds, the ligand exchange rate slows as the 5,6 DMBIm linker proceeds into the ZIF-8 inner surface, exchanges with the original linker and first disrupts the original framework’s crystallinity, then increases order as the reaction proceeds. The ligand exchange rate increases with temperature and the H2/C2 separation factor increases with increases in ligand exchange time and temperature.

Polymer-nanoparticle composites (PNCs) show improved chemical and physical properties compared to pure polymers. However, nanoparticles dispersed in a polymer matrix tend to aggregate due to strong interparticle interactions. Electrospun nanofibers impregnated with nanoparticles have shown improved dispersion of nanoparticles. Currently, there are few models for quantifying dispersion in a PNC, and none for electrospun PNC fibers. A simulation model was developed to quantify the effects of nanoparticle volume loading and fiber to particle diameter ratios on the dispersion in a nanofiber. The dispersion was characterized using the interparticle distance along the fiber. Distributions of the interparticle distance were fit to Weibull distributions and a two-parameter empirical equation for the mean and standard deviation was found. A dispersion factor was defined to quantify the dispersion along the polymer fiber. This model serves as a standard for comparison for future experimental studies through its comparability with microscopy techniques, and as way to quantify and predict dispersion in polymer-nanoparticle electrospinning systems with a single performance metric.

Metal-organic frameworks (MOFs) are a new set of porous materials comprised of metals or metal clusters bonded together in a coordination system by organic linkers. They are becoming popular for gas separations due to their abilities to be tailored toward specific applications. Zirconium MOFs in particular are known for their high stability under standard temperature and pressure due to the strength of the Zirconium-Oxygen coordination bond. However, the acid modulator needed to ensure long range order of the product also prevents complete linker deprotonation. This leads to a powder product that cannot easily be incorporated into continuous MOF membranes. This study therefore implemented a new bi-phase synthesis technique with a deprotonating agent to achieve intergrowth in UiO-66 membranes. Crystal intergrowth will allow for effective gas separations and future permeation testing. During experimentation, successful intergrown UiO-66 membranes were synthesized and characterized. The degree of intergrowth and crystal orientations varied with changing deprotonating agent concentration, modulator concentration, and ligand:modulator ratios. Further studies will focus on achieving the same results on porous substrates.

Zinc oxide nanowires ( NWs) have broad applications in various fields such as nanoelectronics, optoelectronics, piezoelectric nanogenerators, chemical/biological sensors, and heterogeneous catalysis. To meet the requirements for broader applications, the growth of high-quality ZnO NWs and functionalization of ZnO NWs are critical. In this work, specific types of functionalized ZnO NWs have been synthesized and correlations between specific structures and properties have been investigated. Deposition of δ-Bi2O3 (narrow band gap) epilayers onto ZnO (wide band gap) NWs improves the absorption efficiency of the visible light spectrum by 70%. Furthermore, the deposited δ-Bi2O3 grows selectively and epitaxially on the {11-20} but not on the {10-10} facets of the ZnO NWs. The selective epitaxial deposition and the interfacial structure were thoroughly investigated. The morphology and structure of the Bi2O3/ZnO nanocomposites can be tuned by controlling the deposition conditions.

Various deposition methods, both physical and chemical, were used to functionalize the ZnO NWs with metal or alloy nanoparticles (NPs) for catalytic transformations of important molecules which are relevant to energy and environment. Cu and PdZn NPs were epitaxially grown on ZnO NWs to make them resistant to sintering at elevated temperatures and thus improved the stability of such catalytic systems for methanol steam reforming (MSR) to produce hydrogen. A series of Pd/ZnO catalysts with different Pd loadings were synthesized and tested for MSR reaction. The CO selectivity was found to be strongly dependent on the size of the Pd: Both PdZn alloy and single Pd atoms yield low CO selectivity while Pd clusters give the highest CO selectivity.

By dispersing single Pd atoms onto ZnO NWs, Pd1/ZnO single-atom catalysts (SACs) was synthesized and their catalytic performance was evaluated for selected catalytic reactions. The experimental results show that the Pd1/ZnO SAC is active for CO oxidation and MSR but is not desirable other reactions. We further synthesized ZnO NWs supported noble metal (M1/ZnO; M=Rh, Pd, Pt, Ir) SACs and studied their catalytic performances for CO oxidation. The catalytic test data shows that all the fabricated noble metal SACs are active for CO oxidation but their activity are significantly different. Structure-performance relationships were investigated.

The accurate and fast determination of organic air pollutants for many applications and studies is critical. Exposure to volatile organic compounds (VOCs) has become an important public health concern, which may induce a lot of health effects such as respiratory irritation, headaches and dizziness. In order to monitor the personal VOCs exposure level at point-of-care, a wearable real time monitor for VOCs detection is necessary. For it to be useful in real world application, it requires low cost, small size and weight, low power consumption, high sensitivity and selectivity.

To meet these requirements, a novel mobile device for personal VOCs exposure monitor has been developed. The key sensing element is a disposable molecularly imprinted polymer based quartz tuning fork resonator. The sensor and fabrication protocol are low cost, reproducible and stable. Characterization on the sensing material and device has been done. Comparisons with gold standards in the field such as GC-MS have been conducted. And the device’s functionality and capability have been validated in field tests, proving that it’s a great tool for VOCs monitoring under different scenarios.

An accurate knowledge of the complex microstructure of a heterogeneous material is crucial for quantitative structure-property relations establishment and its performance prediction and optimization. X-ray tomography has provided a non-destructive means for microstructure characterization in both 3D and 4D (i.e., structural evolution over time). Traditional reconstruction algorithms like filtered-back-projection (FBP) method or algebraic reconstruction techniques (ART) require huge number of tomographic projections and segmentation process before conducting microstructural quantification. This can be quite time consuming and computationally intensive.

In this thesis, a novel procedure is first presented that allows one to directly extract key structural information in forms of spatial correlation functions from limited x-ray tomography data. The key component of the procedure is the computation of a “probability map”, which provides the probability of an arbitrary point in the material system belonging to specific phase. The correlation functions of interest are then readily computed from the probability map. Using effective medium theory, accurate predictions of physical properties (e.g., elastic moduli) can be obtained.

Secondly, a stochastic optimization procedure that enables one to accurately reconstruct material microstructure from a small number of x-ray tomographic projections (e.g., 20 - 40) is presented. Moreover, a stochastic procedure for multi-modal data fusion is proposed, where both X-ray projections and correlation functions computed from limited 2D optical images are fused to accurately reconstruct complex heterogeneous materials in 3D. This multi-modal reconstruction algorithm is proved to be able to integrate the complementary data to perform an excellent optimization procedure, which indicates its high efficiency in using limited structural information.

Finally, the accuracy of the stochastic reconstruction procedure using limited X-ray projection data is ascertained by analyzing the microstructural degeneracy and the roughness of energy landscape associated with different number of projections. Ground-state degeneracy of a microstructure is found to decrease with increasing number of projections, which indicates a higher probability that the reconstructed configurations match the actual microstructure. The roughness of energy landscape can also provide information about the complexity and convergence behavior of the reconstruction for given microstructures and projection number.