A study on the effect of lateral interactions on methanation over Fe(100)

Doctoral Thesis

2018

Permanent link to this Item
Authors
Journal Title
Link to Journal
Journal ISSN
Volume Title
Publisher
Publisher

University of Cape Town

License
Series
Abstract
In this thesis, the lateral interactions involved in conversion of synthesis gas, a mixture of H2 and CO, to methane over Fe(100) and the effect they have on the kinetics of the process is explored. Understanding the methanation of syngas allows for a better understanding of the initial stages of Fischer-Tropsch synthesis. Density functional theory was used to calculate the energies and properties of simple methanation adsorbates on an Fe(100) surface. All of the parameters were tested and optimized in order to find a balance between efficiency and accuracy. A number of configurations were calculated to investigate nearest neighbour and next nearest neighbour interactions. An energetic break down of the lateral interactions is postulated using the components of the Hamiltonian. The charges associated with the different atoms in each configuration were identified using the Mulliken population analysis and the Bader population analysis. These gave insights into configurations which displayed large electrostatic lateral interactions. Lateral interactions were investigated using larger unit cells than typically utilized in molecular modelling up to now (viz. p(4x4) and p(3x2) unit cells) to enable the estimation of nearest neighbour and next nearest neighbour interactions. When using larger p(4x4) unit cells for CO adsorption on Fe(100), the results showed that the heat of adsorption can differ by as much as 0.24 eV at 0.25 ML. It was concluded that lateral interactions are a function of local coverage (i.e. number of nearest and next nearest neighbours) and not necessarily global coverage. Nearest neighbour interactions are typically repulsive and much larger than next nearest neighbour interactions, which varied between repulsive and attractive interactions. While this is not a unique conclusion it did allow for the creation lateral interaction matrices that vary with temperature. The study has shown that lateral interactions can be broken down into kinetic and potential energy and an inverse relationship exists between these component energies. If this relationship is truly understood, then the total energy can be calculated knowing either kinetic or potential energy instead of both. This would then give additional value to well explored electrostatic interaction models. The lateral interactions were empirically related to nearest neighbour and next nearest neighbour interactions. Two kinetic studies were investigated in this thesis and in both cases, mean field approximations and quasi chemical approximation (QCA) were used and compared to incorporate lateral interactions into the kinetics. The mean field approximation over estimates the lateral interactions and considers global coverage while the QCA approximation considers probability of local combinations. The first kinetic study was a simulated CO TPD experiment on Fe(100). The mean field approximation was an improvement on systems which considered no lateral interactions but did not describe all the aspects observed in the experimental TPD. The prediction by the quasi-chemical approximation shows good agreement for the desorption of associatively bound CO. The deviation observed for the dissociatively adsorbed CO is attributed to the presence of alternative pathways for the adsorbed species (specifically the diffusion of oxygen into the lattice of the solid). A microkinetic model for the methanation of syngas over Fe(100) was also created. The results showed that different methods of lateral interaction incorporation resulted in significantly different coverage profiles and reaction energy profiles. Both methods showed a build-up of oxygen on the surface towards the end of the simulation. The build-up of oxygen on the surface of Fe(100) may indicate that iron-based catalysts need to undergo phase changes to complete the catalytic cycle.
Description

Reference:

Collections