(460a) Controlling Combustion Location in Reverse-Flow Steam Reforming

In contemplating a broader use of hydrogen in our energy infrastructure, it is instructive to consider that our most economical route to hydrogen, Steam-Methane Reforming (SMR) is only 70-75% efficient to hydrogen. SMR inefficiency is rooted in a high-temperature endothermic nature of the reaction. Substantial energy is used to heat streams to reaction temperature, which energy is ineffectively captured as product streams are returned to ambient conditions. Heat exchange can improve efficiency, but the corrosive nature of the syngas makes this heat exchange costly and problematic.

Our approach to raising the efficiency of steam reforming is to perform this heat exchange using the catalyst bed itself. We operate the steam reforming bed as a reverse flow reactor, in which heat addition (by combustion) alternates with heat consumption (by reforming), with the flow being reversed between these steps. By properly designing the bed and other reactor components, the entire bed becomes a high-efficiency heat exchanger. With appropriate design, SMR efficiency can be boosted into the 85-90% range... within a few percent of the theoretical maximum.

We call this approach "Pressure Swing Reforming" (PSR) because we swing the pressure of the reactor such that low pressure air can be used during combustion, while high pressure syngas is produced as reform-product. Efficiency is derived from the heat exchange function of the bed. Hot combustion product adds heat to a reforming zone as it is cooled in one step, and that heat is use to heat and reform hydrocarbon in the following step. The resulting hot syngas is cooled in a recuperating zone of the bed, and this heat is used to preheat the incoming air and fuel during the next step of the reverse-flow cycle. Of central importance is the challenge of deferring combustion to occur after the fuel and air have been heated in this recuperating zone and before entering the reforming zone.

We have explored several approaches to this challenge. In one approach, one of the combustion reactants does not travel through the recuperator, but instead is added at the interface between the recuperator and reforming zones. In another approach, fuel and air are completely mixed before entering the recuperating zone, and the combination of residence time and temperature profile is used to defer combustion. Finally, a third approach has been explored in which the air and fuel travel through the recuperator in separate channels, and are combined and combusted after the passing through the recuperator zone but before entering the reforming zone.

This presentation will provide a brief introduction to PSR and its uses. We will discuss the various approaches we have used for deferring combustion, and will present experimental experiences with these approaches.