Characterisation and Performance Evaluation of Calcium Treatment in Automotive Steels

Advanced High Strength Steels (AHSS) are seeing a dramatic usage increase in order to achieve the reduction in weight required for automotive vehicles to adhere to new legislation on emissions. This has resulted in the increase of use of Calcium (Ca), as a means of both oxide and sulphide-based inclusion control. However, this process is not well understood, with no definitive answer on the level of Ca required to obtain adequate modification of the inclusion population within the steel. As a result, this thesis studied the level of Ca addition applied to the dual-phase steel grade, FB590, by analysing its Calcium Sulphur Ratio (Ca:S). This allowed for the level of modification needed to obtain the required mechanical properties to be studied. It was discovered that even at the low Ca:S of 0.23, sufficient modification of the inclusions was achieved to obtain these properties. Samples across the entire Ca:S range obtained a substantial increase in the number of ‘globular’ inclusions when compared to non-Calcium treated samples. The second aspect of this project was to automate the process of analysing the inclusion populations within the steel. Typically, this is done manually and is very time consuming. The resulting automated process analysed a stitched image of the entire sample surface (200mm2) taken at a typical magnification (100x) and was conducted as closely as possible to the BS EN 10247 and ASTM E45 standards. A custom macro was written that allowed this image to be broken up into fields, each of which were analysed individually, for both inclusion population and size. This method proved significantly less time consuming while generating a vast amount of data that would be unavailable utilising standard analysis techniques. The key problem with the technique, however, is that it did not include the ‘grouped’ inclusions stated in the standards. Nevertheless, EDS analysis proved that a significant proportion of these inclusions were in fact distinct globular inclusions and therefore not the inclusions separated by the polishing process that the standard assumes. A suggestion has been made to conduct all future inclusion analysis on FB590 at 500x magnification so as to reduce the negative effects of pixel size. A series of different regions would then be analysed to better ensure the accuracy of the results