Revealing the prior austenite grain boundaries from a martensitic structure is well known to be very difficult and dependent on the chemical composition and the thermomechanical processing of the steel. In the present study, four different chemical etching reagents and additional thermal etching have been conducted for thermomechanical simulated tool steels Orvar Supreme and Stavax ESR. The etching results have been characterized using light optical microscopy and electron backscattered diffraction. The obtained results show that saturated aqueous picric acid, oxalic and sodium bisulfite based acid reveals prior austenite grain boundaries well for Orvar Supreme. For Stavax ESR, only aqueous CrO3-NaOH-picric acid provides good results in revealing the prior austenite grain boundaries. Thermal etching shows good potential and if conducted properly, thermal etching is a good alternative to the chemical reagents from a health- and environmental perspective.

Effects of mould flux contaminant on scale-steel adhesion and hydraulic descaling of scale formed on slabs were investigated. In this investigation, stainless steel type 304 (austenitic with 18% Cr and 8% Ni) and specific mould fluxes were used when growing the scale on contaminated samples under simulated industrial reheating conditions, with subsequent high pressure water hydraulic descaling. The basic hypothesis was that the steel-scale adhesion depends on the microstructure of different phases present in the scale, the segregation of specific elements at the interface and the interfacial morphology of the scale after reheating. It was found that mould flux contaminant decreases scale-steel adhesion and therefore improved the descaling effectiveness significantly compared to non contaminated stainless steel. The descaling effectiveness of contaminated and uncontaminated slab was dependent to the presence of metal free paths (chromite layers along the austenite grains boundaries) and the presence of unoxidized metal in the scale due to nickel enrichment at the interface. Compared to the uncontaminated samples, the descaling of contaminated samples was efficient which could be due to the fact that some mechanisms which increase scale– steel adhesion (notably nickel enrichment at the interface) were considerably reduced. For all contaminated samples, the descaling effectiveness after visual observation were close to 100% and it was found that mould flux type 832 ( low basicity) gave a high descaling efficiency with better steel surface quality after descaling compared to mould fluxes type 810 and RF1. / Dissertation (MSc)--University of Pretoria, 2009. / Materials Science and Metallurgical Engineering / unrestricted