Our research is focused on developing compact dense ceramic membranes that will enable the efficient and cost-effective production of hydrogen by reforming natural gas (NG) using pure oxygen that is formed by water splitting and transported by the membrane. In our approach, hydrogen is produced on the steam side of the membrane while synthesis gas (a mixture of hydrogen and carbon monoxide) is produced on the NG side. Water dissociates into oxygen and hydrogen at high temperatures, however, only very small concentrations of hydrogen and oxygen are generated even at relatively high temperatures. For example, hydrogen and oxygen concentrations of only 0.1 and 0.042%, respectively, are produced at 1600°C, because the equilibrium constant for this reaction is small. Significant amounts of hydrogen or oxygen can be generated at moderate temperatures, however, if the equilibrium is shifted toward dissociation by using a mixed-conducting (i.e., conducts both electrons and ions) membrane to remove either oxygen or hydrogen. We have studied hydrogen production via water splitting at moderate temperatures (500-900°C) with novel mixed-conducting membranes. Hydrogen production rates were investigated as a function of temperature, water partial pressure, membrane thickness, and chemical potential gradient across the membranes. The hydrogen production rate increased with increases in temperature, water partial pressure, and oxygen chemical potential gradient and with a decrease in membrane thickness. The maximum hydrogen production rate, ≈10 cm3(STP)/min-cm2, was measured with a 0.30-mm-thick membrane at 900°C using 50 vol.% water vapor on one side of the membrane and 80% hydrogen (balance helium) on the other side. It is expected that the hydrogen production rate can be increased to >20 cm3(STP)/min-cm2 by reducing the thickness of the membrane. In these experiments, hydrogen was used only as a model gas to establish a high oxygen potential gradient. While producing hydrogen from hydrogen is impractical, our results illustrate that significant quantities of hydrogen and oxygen can be generated by water dissociation. Pure oxygen that is generated can be used to reform NG. To demonstrate this concept, we used methane (instead of hydrogen) in several experiments to maintain the high chemical potential gradient across the membrane. With methane as the feed gas, hydrogen was produced on both sides of the membrane. The oxygen produced by water dissociation was transported across the membrane and reacted with methane to form synthesis gas. The synthesis gas can be subjected to a water-gas shift reaction to produce more hydrogen and a stream enriched in CO2 for sequestration. In this talk, we will present test results of hydrogen production by water dissociation using mixed-conducting membranes.

Work supported by U.S. Department of Energy (DOE), Energy Efficiency and Renewable Energy, Office of Hydrogen, Fuel Cells, and Infrastructure Technologies Program and Office of Fossil Energy, National Energy Technology Laboratory's Hydrogen and Gasification Technologies Program, under Contract W-31-109-Eng-38.