Objective of this program was to identify the poisoning mechanism(s) responsible for performance losses in molten carbonate fuel cells operating on fuels contaminated with ppM levels of H/sub 2/S and to develop models and codes to predict cell performance under poisoned conditions. This objective was addressed by focusing out-of-cell and in-cell experiments on single mechanistic issues, followed by incorporation of the results into a model that correlated the performance decline to the contaminant concentration. The importance of this work is that a justifiable specification for gas cleanup for MCFC power plants can now be made and a reasonably accurate model now exists which is capable of predicting the impact that upsets in the sulfur cleanup system will have upon cell performance. As the result of this work, it is now known that a complex relationship exists between the level of sulfur contamination and the amount of performance loss that is experienced by a fuel cell operating on contaminated fuel. This loss is dependent upon the total amount of sulfur present in the fuel, the heating value of the fuel, specifically the hydrogen content, the load placed on the fuel cell, and the pressure at which the cell is operating. Therefore, no single set of sulfur cleanup specifications is possible; however, in general, a MCFC system operating on a medium-Btu fuel at atmospheric pressure can tolerate about 1 ppM of total sulfur without suffering any loss in performance and with no adverse long-term effects on materials survivability. Such a system could also handle as much as 5-ppM total sulfur during a temporary upset in the sulfur cleanup system but with a concomitant loss in performance of about 20% provided the upset lasted less than 24 hours.