Oxygen-mediated mitigation of Ru/Al2O3 deactivation during propane steam reforming

Cotterrell, Stephen

Oxygen-mediated mitigation of Ru/Al2O3 deactivation during propane steam reforming

Thesis / Dissertation

2023

Publisher

University of Cape Town

Department

Department of Chemical Engineering

Faculty

Faculty of Engineering and the Built Environment

Abstract

PEMFCs promise efficient and low emission energy generation, compared to internal combustion engines (ICEs). If operated using pure hydrogen, the only by-products of a PEMFC system are heat and water. Indeed, even when the hydrogen is generated from hydrocarbon fuels, emissions are limited to CO2 (and small amount of hydrocarbons during start up), in contrast to ICEs which typically may produce significant environmental pollutants, such as NOx. A pivotal issue in the deployment of PEMFC systems is the availability of hydrogen feedstock. While much research has been done into hydrogen storage technologies, there are significant shortcomings, primarily low volumetric and gravimetric storage density. An alternative is the small-scale on-site production of hydrogen by catalytic processing of a readily available hydrocarbon feed. Reforming is an important step in this process and industrial applications typically employ nickel- based catalysts which suffer from complicated pre-operational reduction procedures and may be pyrophoric on exposure to air. In contrast, the requirements for small-scale distributed hydrogen production are significantly different from those in industrial settings. This study focuses on the performance of an air-stable ruthenium catalyst, Ru/Al2O3, which avoids the needs to extended pre-operational activation/reduction procedure, specifically catalyst stability and the introduction of a limited oxygen (air) in the reformer feed for purposes of mitigating deactivation. The study included the construction of an experimental apparatus to perform catalytic performance evaluation and an investigation of the effect of changing certain key variables (air co-feed and steam- to-carbon ratio) on catalyst performance and stability. Further work was performed to characterise the catalyst before and after use to determine the cause of deactivation. Catalytic performance testing results indicated that the addition of an air co-feed at 1-3% of the total dry feed composition had a very limited effect on the deactivation of the catalyst, but appeared to slightly shift the composition of the reaction products favourably, with a greater effect incurred by a larger co-feed. The favourable shift in reaction products was reversible on removal of the air co-feed. Characterisation of the spent catalyst from these catalytic performance tests indicated a significant extent of coking, supporting the hypothesis that coking is the primary driver of deactivation.