Carbon capture and storage is a concept to reduce greenhouse gas emissions resulting from the use of fossil fuels for power generation. In oxy-fuel combustion, the fuel is burned in a mixture of oxygen and recirculated flue gas. Due to the absence of nitrogen the recirculated flue gas is enriched with CO 2, which makes the post-combustion extraction easier. Recirculation increases also the levels of SO x and water vapour in the flue gas. Limestone (CaCO 3) is commonly used as an absorbent for SO 2 capture. In oxy-fuel combustion calcination may be locally inhibited due to lower combustion temperature (than in air-combustion) and higher CO 2 concentration, thus the capture mechanism may change from normal sulphation (CaCO 3→>CaO→CaSO 4) to direct sulphation (CaCO 3→CaSO 4). The simultaneous occurrence of calcination, sulphation and recarbonation may cause formation of hard deposits which are difficult to remove from heat transfer surfaces. Such a deposit may cause operational problems like plugging of gas channels and/or corrosion of superheaters. In this paper, the corrosion performance of three different superheater austenitic steels (TP347HFG, HR3C and Sanicro 25) has been determined in laboratory tests under simulated oxy-fuel conditions (60 CO 2 - 4 O 2 - bal N 2/Ar vol%) with three levels of H 2O (0, 10 and 30 vol%), with and without a synthetic carbonate based deposit (85 CaCO 3 - 15 CaSO 4 wt%) at 650°C for 168, 500 and 1000 hours. The results obtained from oxy-fuel conditions are compared with corresponding tests simulating conventional air-combustion and co-combustion in air (with and without the 85CaO-14CaSO 4-1KCl synthetic deposit). After the tests, the specimens were analysed using optical and SEM/EDX microscopy to determine the oxide layer thickness and the elemental distribution in the oxide scale as well as the type of the corrosion attack. The carburisation tendency was evaluated from etched cross sections using optical microscopy. The results from the 168 h tests in a moist and a high CO 2 atmosphere showed clear corrosion of TP347HFG. In addition, carburisation was observed already in the 168 h tests with the CaCO 3-CaSO 4 deposits. The HR3C and Sanicro 25 performed better than TP347HFG. However, growing oxide nodules could be observed already after 168 h in some cases with HR3C and Sanicro 25. A clear oxide layer was observed on all materials after 168 h exposures in air, while oxy-fuel conditions with water vapor caused significant corrosion only to the TP347HFG.