UC Berkeley

The work herein describes the investigation of metal-organic frameworks for industrial applications, specifically gas phase separations of mixtures. Metal-organic frameworks are crystalline molecular scaffolds built from cationic metal vertices and organic bridging ligands. They are porous on a molecular scale and can separate gas mixtures when one component interacts more strongly with the pore walls than others. The near-infinite combination of metals and ligands allows for optimization of metal-organic framework structures for specific separations.

Chapter 1 begins with an explanation of why gas separations are relevant to protecting the environment and preventing atmospheric CO2 emissions, followed by an overview of the basic chemistry of metal-organic frameworks. The chapter then presents an extensive survey of previously reported literature on CO2/H2 separations and hydrocarbon separations using metal-organic frameworks, which are the two specific separations reported in this work.

Chapters 2, 3, and 4 focus on CO2/H2 mixtures. The first chapter describes the first experimental survey of metal-organic frameworks for this separation. Five materials were chosen as representative of different types of metal-organic frameworks: two with high internal surface areas, one with structural flexibility, and two with highly polarizing cationic metal sites lining the pores. The materials with these coordinatively-unsaturated metal centers demonstrated vastly improved behavior over the materials that are currently used industrially for this separation. Of these two, the framework with exposed Mg2+ sites, Mg2(dobdc), was deemed to be superior due to its high selectivity for CO2 over H2. The framework with exposed Cu2+ sites, Cu-BTTri, had a lower selectivity but higher capacity for CO2. Both selectivity and CO2 capacity are important metrics in evaluating adsorbents, and in this case selectivity has been reported to be paramount.

Chapter 3 examines Mg2(dobdc) in further detail due to its promising behavior in Chapter 2. Industrially, CO2/H2 mixtures are contaminated with a variety of impurities, primarily methane and carbon monoxide. Carbon monoxide is very reactive and therefore more manageable to remove. Methane, however, behaves very similarly to H2 and therefore materials that isolate H2 from both CO2 and methane are highly desirable. Chapter 3 evaluates the performance of Mg2(dobdc) when separating CO2/H2/CH4 mixtures and describes its excellent performance.

Chapter 4 describes an attempt to improve upon the CO2/H2 selectivity of Cu-BTTri. Mg2(dobdc) performs better than Cu-BTTri in part because Mg2+ cations are more polarizing than Cu2+ cations. By generating a material isostructural to Cu-BTTri but featuring Mn2+ cations which are also more polarizing than Cu2+, greater CO2/H2 selectivities can be achieved.

Chapter 5 is a departure from the previous chapters and investigates the separation of alkane isomers. Isomers are a particularly challenging mixture to separate because their reactivity and physical characteristics are very similar. This chapter presents the synthesis of a metal-organic framework with triangular channels that can separate hexane isomers based on their shape. The acute angles of the channels allow for isomers with a higher aspect ratio to wedge themselves into the channel corners, while the bulkier isomers have a weaker interaction.