Date of Award

Summer 2011

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Mechanical Engineering

First Advisor

Goldsborough, Samuel S.

Second Advisor

Borg, John

Third Advisor

Koch, Jon

Abstract

Because of its high energy density, hydrogen is a desirable energy source for the achievement of a renewable energy landscape. Though production methods like thermolysis, electrolysis and biomass conversion, among others, are thought to be long term renewable solutions, catalytic steam methane reforming (SMR) is currently the predominant mechanism to produce hydrogen on an industrial scale. The highly endothermic, transport-limited reforming process has also been scaled down through process intensification to create efficient small-scale hydrogen-generating systems. One proposed geometry utilizes a catalytic finned cylinder that provides a manufacturable solution to enable high-efficiency heat exchange and SMR reaction. An accurate representation of the reactor performance characteristics is imperative to the design of small-scale systems.

The Nusselt and Sherwood numbers, the respective dimensionless temperature and concentration gradients, are commonly used to model the transport characteristics. Previous works have outlined the significance of modeling techniques that include radial diffusion to capture the bulk-phase diffusive resistance. However, prior studies have either over-simplified the transport to neglect diffusion in the bulk fluid or employed CFD to include the relevant effects. A considerable limitation of CFD-derived solutions is a high degree of computational intensity.

In the current study, local transport coefficients are determined for the SMR reaction in a catalytic microchannel. The 2-D cylindrical transport equations are simplified based on approximations from prior work to represent the channel geometry. The applied assumptions dramatically decrease the model's computation time. A finite central-differencing scheme is implemented to solve the coupled transport equations with the reaction kinetics, and is solved through simultaneous matrix inversion. A kinetic model for SMR reactions is included as a model subroutine to describe the highly non-linear transport/kinetic interactions, while accounting for species adsorption/desorption to and from the catalyst. The transport model is compared to known solutions for the desired boundary conditions to validate the diffusive effects. The full model is validated against experimental data, and is able to reasonably predict the expected transport behavior and chemical kinetic interactions in the catalytic microchannel.

COinS