Study of Adsorbate Interactions of Alumina Surfaces Using Nuclear Magnetic Resonance Methods
Gamma type alumina (gamma-Al2O3) is a fundamental material in heterogeneous catalysis with a growing demand for environmentally benign solid acid catalysts to replace toxic and corrosive liquid solvents for a wide range of applications including renewable energy production, fine chemicals and petrochemicals. Given the complexity and amorphous nature of gamma-Al2O3, the role of water continues to be ambiguous and of significant interest for industry. This thesis has applied a combination of analytical techniques including high throughput experimentation (HTE), advanced NMR techniques, multivariate linear regression (MLR), X-ray diffraction (XRD), nitrogen porosimetry and thermogravimetric analysis (TGA) to address the fundamental research questions; what is the role of water in thermal, hydrothermal pretreatments and operating conditions for dimethyl methyl ether (DME) synthesis and is the hydroxyl coverage related to the catalytic activity?
Advanced NMR techniques were performed, including NMR relaxometry and diffusiometry at 43 MHz to determine the absorbate-adsorbent interactions of the reactant (methanol, CH3OH) of DME synthesis and the by-product (water, H2O) following a thermal or a variation of a steam pretreatment step for gamma-Al2O3. Two types of steaming protocols were investigated, the critical difference being that one protocol steamed the alumina catalyst at the set-point only, whereas the alternative protocol also steamed the samples during the ramp and cool down. The two types of steam pretreatments and thermal pretreatments as a function of temperature (200-450 °C) were compared and showed a negligible interaction strength of small, rigid alkanes relative to the strongly interacting polar water and methanol guest molecules within the mesoporous oxide. Functional-specific analysis of methanol was performed via deuterated variants (CD3OH, CH3OD), with the molecular motion of the hydroxyl and aliphatic functional group differentiated. Tortuosity analysis of the steamed and calcined catalysts was used to determine the mass transport properties of the pretreated catalysts. For both types of thermal pretreatments investigated (in vacuo and under inert gas) the structural properties of gamma-Al2O3 remained constant within the temperature range: 200-450 °C. However, a variation of the steaming pretreatment (steaming sequence 1) was shown to significantly alter the structural network by 14 %. The results of XRD analysis for both thermal and water vapour pretreatments confirmed that the effect of both the thermal and steam pretreatments was localised to the surface with no bulk phase transitions observed.
High throughput experimentation (HTE) supported the NMR relaxation and tortuosity analysis of steamed alumina to identify the catalyst pretreatment as a critical factor to optimise the catalytic conversion of DME synthesis where gamma-Al2O3 is the solid acid catalyst. Furthermore, novel results implied that steam pretreatments can play a dual role in catalysis both as an inhibitor and activator being significantly dependent on the steaming protocol and temperature. Therefore, these insights question the traditional dogma in literature of water’s role as an inhibitor for gamma type alumina, independent of conditions. Counter to the general assumption, steaming sequence 2 (S2) above 350 °C was shown to produce the highest DME conversion rate for both pure gamma and mixed phase polymorphs. Relative to the conventional thermal pretreatment under the equivalent conditions of 390 °C, the conversion rate of the S2 pretreated catalysts was shown to be 15-57 % higher. Furthermore, a 10-day lifetime study proved the positive affect of the S2 pretreatment process to be stable and long term, supporting the initial 6 h activity studies.
To understand the HTE, relaxometry and diffusiometry studies, solid state NMR was performed to reveal the coordination of the surface aluminium species. Surface sensitive NMR techniques implied that formation of penta- and four-coordinated species may strongly influence the DME conversion rate due to an increase of the surface Lewis acidity. The surface hydroxyl population for both thermal and steam pretreatments of interest were quantified using NMR with an optimum hydroxyl density range of approximately 5-6 OH/nm2 linked to the catalytic behaviour. These findings have the potential to influence the rational design of activation procedures for other metal oxides and zeolites.
Engineering and Physical Sciences Research Council (1945288)