Laboratory simulation of metal dusting corrosion
Permanent link to this Item
Link to Journal
University of Cape Town
A laboratory carburising furnace in which metal dusting conditions are simulated has been designed and constructed. This furnace has been used to simulate and study the metal dusting corrosion of four iron-based alloys viz. 9Cr Mo 45 steel, Incoloy 800H, AISI 310 stainless steel and Chromanite - an experimental high-nitrogen Cr-Mn stainless steel (HNSS). Tests conducted on the carburising furnace show that the rig is capable of heating a flowing gas environment to temperatures of 800°C in the horizontal ceramic tube. The design allows the testing of up to thirty-six test specimens in a constanttemperature test zone. Systems for the safe heating and disposal of gases such as hydrogen and carbon monoxide have been incorporated into the design. Four twenty-four hour exposures were performed on specimens of the CrMo steel and as received samples of AISI 310. This was followed by a series of seven week-long exposures of Incoloy 800H, AISI 310 and the high nitrogen stainless steel (HNSS). These specimens were tested in an annealed and polished condition in order to increase their susceptibility to metal dusting. In a third test series, specimens of these three alloys were tested in an annealed and abraded condition in order to determine the effect of grain size and surface roughness on metal dusting resistance. Exposures of the CrMo specimens resulted in general metal loss and massive carbon deposition after the first exposure of 24 hours. Filamentous carbon deposits containing metal particles showed that metal dusting corrosion of the specimens had taken place. The as received AISI 310 specimens showed no signs of metal dusting attack over the same exposure time. This was attributed to a protective surface chromia layer that prevented carburisation of the specimens. During the second test series, specimens of Incoloy 800H in the sensitised condition showed a high susceptibility to metal dusting. Carburisation of the matrix carburisation was accompanied by large carbon protrusions growing from the specimens' surfaces. Large pits were observed on the specimens after five weeks. Sensitised AISI 310 specimens also showed signs of metal dusting but at a slower rate. The difference in performance between these two alloys was attributed to the difference in alloying contents, notably chromium, nickel and silicon. The HNSS specimens showed a high resistance to carburisation, carbon deposition and metal loss during the first six weeks. Small amounts of carbon deposition and pitting were observed after the seventh exposure. The good resistance to metal dusting of this alloy was attributed to its alloying contents, which included chromium, manganese, sulphur and nitrogen. The results of the third test senes showed that resistance to metal dusting was significantly improved by increasing the surface roughness and decreasing the grain size of the specimens. A new alloy, Fe-25Cr-12Ni-9Mn-4.5Al-2Si-0.5N is proposed for fabrication and exposure to metal dusting environments to evaluate its suitability for use in industrial applications. It is also recommended that further work be carried out in evaluating the effect of increasing the nitrogen, chromium and manganese contents of the Fe-18Cr- 9Mn-0.5N alloy that performed well in this project. Investigations into the effect of aluminising and nitriding components should also be carried out.
Vaughan, A., Vaughan, A. 1997. Laboratory simulation of metal dusting corrosion. University of Cape Town.