(113g) Integrated Hydrogen Production, Purification and Compression Demonstration Project

As part of an ongoing US Department of Energy sponsored project, a novel hydrogen production demonstration unit using a natural gas feed was built and tested by the project partners (Linde North America Inc., Membrane Reactor Technologies Ltd. and Ergenics Corp.). The system comprises of an autothermal fluidized bed membrane reformer coupled with a non-mechanical, metal hydride compressor (MHC). The system is designed to produce 15 Nm3/h of compressed hydrogen at 100 bar. The reformer includes 25 palladium-alloy membranes (25 um thick foil) to provide in-situ separation of hydrogen from the reforming zone. Removal of hydrogen from the reactor shifts the thermodynamic equilibrium forward, so that high conversion rates of methane are achieved at relatively mild operating conditions (550°C, 25 bar). Air is directly added to the fluidized reforming bed to provide the endothermic reaction heat, thereby eliminating the need for costly indirect heat transfer. The fluidized bed environment facilitates an isothermal reactor, which enhances membrane performance. Heat is transferred from the heating (air addition) zones to the reforming (membrane) zones through solids movement. The MHC is a 3-stage system. Hydrogen from the reformer is absorbed in cool metal hydride beds, and then desorbed at higher pressure after heating. The system compresses hydrogen permeate from the reformer membranes from sub-atmospheric to 100 bar in a single system by engineering the composition of the metal hydrides in each compression stage. Compression work is provided by thermal heat, which in the current system is a hot water solution heated by a small natural gas burner. Coupling the membrane reformer and MHC provides thermal efficiency and capital cost benefits: - The MHC can provide sub-atmospheric pressure (suction) on the hydrogen membranes. This increases the hydrogen permeation driving force, thereby reducing the membrane surface area in the reformer. - Excess energy from the reformer can provide a significant portion of the compression work. - Common heat transfer systems (e.g. burner, gas cooling). The results from demonstration testing will be presented, including: - Reformer performance - Hydrogen purity - Membrane longevity - Autothermal operation - Compressor performance - System operability - Effect of cyclic compressor on the fluidized bed reformer Efficiency and cost data for an advanced prototype will also be presented, including a discussion of the key technical and cost issues with this novel technology.