BP has a programme to develop technologies that could reduce greenhouse gas emissions, by the capture and storage Of CO2 from existing industrial boilers and process heaters. One generic technology under development is oxyfuel combustion, with flue gas recycle. Previous studies, by three of the authors, have concluded that refinery steam boilers could be successfully converted to oxyfuel firing. Fired heaters, however, differ from boilers in several respects and so it was decided to study the feasibility of converting process heaters. Three heaters, located on BP’s Grangemouth refinery, were chosen as examples, as they are typical of large numbers of heaters worldwide. In establishing the parameters of the study, it was decided that the heat fluxes to the process tubes should not be increased, compared to conventional air firing. For two of the heaters this was achieved by proposing a slightly higher recycle rate than for the boiler conversion studied earlier – the heater duty would be retained with no changes to the tubes. For the third heater, where the process duty uses only the radiant section, the CO2 capture cost and the firing rate could be reduced by lowering the recycle rate. Some air inleakage to these heaters was considered inevitable, despite measures to control it, and therefore plant to remove residual inerts from the CO2 product was designed. Cryogenic oxygen production was selected for two heaters, but for the smallest heater vacuum swing adsorption was more economic.