Advanced coal-power generation combined with fuel cell and hydrogen production from hot reformate gas for various applications, including SOFCs as battery chargers (military applications) or auxiliary power units (APUs) for commercial uses have been receiving much attention lately. Removal of H2S from the fuel gas is necessary in all the above energy generation systems to meet the rigorous environmental regulations for sulfur emissions, prevent corrosion of the blades of the turbines in IGCC and fouling of the anode materials in the fuel cell systems. Hot gas cleanup from H2S not only saves up to three percentage points in the overall energy efficiency of the process but also results in considerably lower hardware costs since the volume of the fuel gas stream is a lot smaller than that of the flue gas stream (diluted by the air used in combustion). Rare earth oxides have recently received attention because of their excellent sulfidation equilibria, which allows sub-ppmv H2S concentration levels to be achieved at temperatures as high as 800 oC. In on-going work in our laboratory, cerium and lanthanum oxides have shown excellent equilibria and good kinetics as high-temperature sorbents for the removal of H2S upstream of the SOFC anode. However, similar to other sorbents, such as ZnO and ZnO-based materials, regeneration of the bulk of the sorbent is slow and sulfates may form under certain oxygen partial pressures. Single or mixed lanthanide (La, Pr) oxides can be used for non-regenerative high temperature desulfurization because of their excellent sulfidation thermodynamics in realistic reformate gas compositions. In this work, potential improvements of the sulfur capacity of the La oxide-based materials at 1 ppmv H2S breakthrough will be shown. Regenerative desulfurization is possible, if one uses just the sorbent surface capacity. This is demonstrated with pre-sulfided sorbents. The kinetics of adsorption and desorption are very fast and small reactor volumes can be used. Complete sorbent regenerability and stability in long cyclic operation makes this approach highly suitable for small-footprint fuel cell systems.