Technically, the capture Of CO2 from the flue gas of natural gas/coal fired power plants, using a monoethanolamine (MEA) absorption process is a viable short to medium term strategy for mitigation of the atmospheric CO2 emissions from large point sources. The costs, however, of such systems are currently perceived to be too high to be economically feasible. A practical research objective is the analysis of the CO2 capture process with a view towards minimizing the cost of implementation. Because large scale CO2 capture operations are expensive, process simulation and modelling have an important role to play in evaluation of the process alternatives. In the case of amine scrubbing, the convergence of the process flow sheet is a nontrivial. exercise due to the highly nonlinear nature of the process and the large recycle stream involved. This paper presents a method of decomposing the flow sheet that assists in converging the process flow sheet and in optimizing key process operating variables, in particular the amine loadings and temperature of MEA entering the stripper. This method was applied to three different CO2 concentrations (molar fraction, wet basis): 3% (to simulate flue gas from a gas turbine), 14% (flue gas from a coal plant) and 25% (flue gas from a cement plant). The decomposition method proved useful for investigation of the process parameters in both the absorber and the stripper. The results from the decoupled flow sheet provided good initial estimates for the convergence of the integrated flow sheet. The results from the decoupled and integrated flow sheets were similar. A minimum reboiler duty was found at a lean MEA loading of 0.25 for all CO2 concentrations studied. (C) 2004 Elsevier Ltd. All rights reserved.