Matching Items (2)
Filtering by
Description
A scheme has been developed for finding the gas and temperature profiles in an environmental transmission electron microscope (ETEM), using COMSOL Multiphysics and the finite element method (FEM). This model should permit better correlation between catalyst structure and activity, by providing a more accurate understanding of gas composition than the assumption of homogeneity typically used. While more data is needed to complete the model, current progress has identified several details about the system and its ideal modeling approach.
It is found that at the low pressures and flowrates of catalysis in ETEM, natural and forced convection are negligible forms of heat transfer. Up to 250 °C, radiation is also negligible. Gas conduction, being enhanced at low pressures, dominates.
Similarly, mass transport is dominated by diffusion, which is most accurately described by the Maxwell-Stefan model. Bulk fluid flow is highly laminar, and in fact borders the line between continuum and molecular flow. The no-slip boundary condition does not apply here, and both viscous slip and thermal creep must be considered. In the porous catalyst pellet considered in this work, Knudsen diffusion dominates, with bulk flow being best described by the Darcy-Brinkman equation.
With these physics modelled, it appears as though the gas homogeneity assumption is not completely accurate, breaking down in the porous pellet where reactions occur. While these results are not yet quantitative, this trend is likely to remain in future model iterations. It is not yet clear how significant this deviation is, though methods are proposed to minimize it if necessary.
Some model-experiment mismatch has been found which must be further explored. Experimental data shows a pressure dependence on the furnace temperature at constant power, a trend as-yet unresolvable by the model. It is proposed that this relates to the breakdown of the assumption of fluid continuity at low pressures and small dimensions, though no compelling mathematical formulation has been found. This issue may have significant ramifications on ETEM and ETEM experiment design.
It is found that at the low pressures and flowrates of catalysis in ETEM, natural and forced convection are negligible forms of heat transfer. Up to 250 °C, radiation is also negligible. Gas conduction, being enhanced at low pressures, dominates.
Similarly, mass transport is dominated by diffusion, which is most accurately described by the Maxwell-Stefan model. Bulk fluid flow is highly laminar, and in fact borders the line between continuum and molecular flow. The no-slip boundary condition does not apply here, and both viscous slip and thermal creep must be considered. In the porous catalyst pellet considered in this work, Knudsen diffusion dominates, with bulk flow being best described by the Darcy-Brinkman equation.
With these physics modelled, it appears as though the gas homogeneity assumption is not completely accurate, breaking down in the porous pellet where reactions occur. While these results are not yet quantitative, this trend is likely to remain in future model iterations. It is not yet clear how significant this deviation is, though methods are proposed to minimize it if necessary.
Some model-experiment mismatch has been found which must be further explored. Experimental data shows a pressure dependence on the furnace temperature at constant power, a trend as-yet unresolvable by the model. It is proposed that this relates to the breakdown of the assumption of fluid continuity at low pressures and small dimensions, though no compelling mathematical formulation has been found. This issue may have significant ramifications on ETEM and ETEM experiment design.
ContributorsLangdon, Jayse Tanner (Author) / Crozier, Peter (Thesis director) / Hildreth, Owen (Committee member) / Chemical Engineering Program (Contributor) / Materials Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2017-05
Description
Titanium dioxide is an essential material under research for energy and environmental applications, chiefly through its photocatalytic properties. These properties allow it to be used for water-splitting, detoxification, and photovoltaics, in addition to its conventional uses in pigmentation and sunscreen. Titanium dioxide exists in several polymorphic structures, of which the most common are rutile and anatase. We focused on anatase for the purposes of this research, due to its promising results for hydrolysis.
Anatase exists often in its reduced form (TiO2-x), enabling it to perform redox reactions through the absorption and release of oxygen into/from the crystal lattice. These processes result in structural changes, induced by defects in the material, which can theoretically be observed using advanced characterization methods. In situ electron microscopy is one of such methods, and can provide a window into these structural changes. However, in order to interpret the structural evolution caused by defects in materials, it is often necessary and pertinent to use atomistic simulations to compare the experimental images with models.
In this thesis project, we modeled the defect structures in anatase, around oxygen vacancies and at surfaces, using molecular dynamics, benchmarked with density functional theory. Using a “reactive” forcefield designed for the simulation of interactions between anatase and water that can model and treat bonding through the use of bond orders, different vacancy structures were analyzed and simulated. To compare these theoretical, generated models with experimental data, the “multislice approach” to TEM image simulation was used. We investigated a series of different vacancy configurations and surfaces and generated fingerprints for comparison with TEM experiments. This comparison demonstrated a proof of concept for a technique suggesting the possibility for the identification of oxygen vacancy structures directly from TEM images. This research aims to improve our atomic-level understanding of oxide materials, by providing a methodology for the analysis of vacancy formation from very subtle phenomena in TEM images.
Anatase exists often in its reduced form (TiO2-x), enabling it to perform redox reactions through the absorption and release of oxygen into/from the crystal lattice. These processes result in structural changes, induced by defects in the material, which can theoretically be observed using advanced characterization methods. In situ electron microscopy is one of such methods, and can provide a window into these structural changes. However, in order to interpret the structural evolution caused by defects in materials, it is often necessary and pertinent to use atomistic simulations to compare the experimental images with models.
In this thesis project, we modeled the defect structures in anatase, around oxygen vacancies and at surfaces, using molecular dynamics, benchmarked with density functional theory. Using a “reactive” forcefield designed for the simulation of interactions between anatase and water that can model and treat bonding through the use of bond orders, different vacancy structures were analyzed and simulated. To compare these theoretical, generated models with experimental data, the “multislice approach” to TEM image simulation was used. We investigated a series of different vacancy configurations and surfaces and generated fingerprints for comparison with TEM experiments. This comparison demonstrated a proof of concept for a technique suggesting the possibility for the identification of oxygen vacancy structures directly from TEM images. This research aims to improve our atomic-level understanding of oxide materials, by providing a methodology for the analysis of vacancy formation from very subtle phenomena in TEM images.
ContributorsShindel, Benjamin Noam (Author) / Crozier, Peter (Thesis director) / Anwar, Shahriar (Committee member) / Singh, Arunima (Committee member) / Materials Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05