A system to produce hydrogen with high purity and without CO2 emissions briefly consists of two operating units: production and separation. The coupling of the steam-iron process to the cracking of methane can manage that goal. However the steam-iron process needs an active and stable redox solid at moderate temperatures. The most suitable (pure iron oxide) suffers quick and strong deactivation mainly due to the structural changes upon reduction-oxidation cycles. Among those tested in our laboratory, one promising solid was proposed: 98 wt%Fe2O3-1.75 wt%Al2O3-0.25 wt%CeO2. In this work, the expensive and rare cerium has been substituted by molybdenum. After optimizing the Mo amount in the solid, the long lasting experiments show that the new triple oxide, 98 wt%Fe2O3-1.75 wt% Al2O3-0.25 wt%MoO3, in spite of some initial deactivation, maintains slightly better hydrogen production rates than the cerium sample. At temperature and conditions studied the Mo-solid was able to run, without coke formation, under real exhaust gas from natural gas thermocatalytic decomposition, producing about 8.1 g of high purity (>99.995%) hydrogen h(-1) kg of solid(-1). This means a natural gas processing of about 68 Nm(3) h(-1) 1000 kg of solid(-1) (at 67% conversion of methane to hydrogen). Copyright (C) 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.