Corrosion of Stainless Steel in Diesel Exhaust Gas After-treatment Systems with Urea Injection

Licentiate thesis, 2010

The demands on lowering the emissions from fossil fuel keep increasing. Together with higher demands of longer component lifetime and lighter vehicles for lower fuel consumption this has led to an increasing use of stainless steels in exhaust after-treatment systems. In diesel exhaust systems one method used for removing the NOx from the exhaust gases is selective catalytic reduction. On heavy-duty trucks the reducing agent ammonia is commonly added in the form of urea water solution. In large scale use this method is fairly young and there is much yet to learn about the corrosivity of such systems. This thesis is a step in the direction of increasing this knowledge. To that effect laboratory experiments simulating a part of the muffler just before the catalyst have been performed as well as analysis of both laboratory and bench tested materials. For the laboratory experiments four stainless steels were used, three ferritic alloys and the austenitic 304L. From bench tests 304L and the high alloyed austenitic steel 904L were investigated. The laboratory experiments exposed the samples to artificial diesel exhaust gases and evaporated urea solution under isothermal and cyclic conditions, including a cold zone with condensate formation. These revealed that isothermal exposure at 30°C did not give any corrosion and that cyclic exposure caused more severe corrosion than isothermal exposure at 450°C. AES and XPS analyses of the corrosion products showed that while the austenitic alloy performed best, the ferritic alloys all displayed the same oxide composition and amount of oxide formation. Experiments replacing the urea solution with water showed that urea additions rather decrease the corrosion under these conditions. Sulphidation, on the other hand, appears to be influencing the corrosion as sulphur enrichment was found in the inner Cr-rich oxides on all cyclically exposed samples. Analyses of the bench tested materials from near the urea injection did not indicate any influence of sulphur, but rather an influence of nitrogen uptake. Silicon oxide was observed in high amounts on both alloys, but is not believed to be active in the corrosion. Deposit formation and the possible presence of different complex formers may influence the process and influence of flow is clearly visible. The 304L steel had suffered severe corrosion and mainly exhibited Fe-oxide and what might be nitride precipitates under the oxide. The 904L steel showed an inner Cr-rich oxide and was only mildly corroded, but displayed a thick nitrided layer. This, however, appears to be nitrogen in solution. That may contribute to the difference in formation of Cr-rich oxides in the two alloys.