Influence of grain size and niobium content on the creep resistance of ferritic stainless steels

Includes abstract.Includes bibliographical references (leaves 98-102).Type 441 ferritic stainless steel is used for the production of catalytic converter housings. As the housing is subjected to high temperatures it is necessary that the material offers creep resistance. Type 441 is dual stabilised with Ti and Nb to provide improved weldability; however, Nb addition also enhances the hot strength and creep resistance by means of precipitation and solid solution strengthening. Notwithstanding the Nb strengthening effect, the strong dependence of creep resistance on grain size also means that the relationship between creep resistance and Nb content may be complicated by grain growth inhibition that arises from aspects of solute drag and grain boundary pinning. Thus it may not be simple to predict the relative creep resistance of standard production heats on the basis of Nb level alone and other factors affecting solid solution and grain size also need to be taken into account. Consequently, it is pertinent to evaluate more closely the sensitivity of these parameters in influencing creep resistance by choosing two alloy heats with different Nb contents and subjecting them to a range in heat treatments that will modify solute level and grain size. This thesis reports on the examination of the dependence of creep resistance on Nb level by eliminating the influence of grain size. The latter was achieved by manipulating the post-cold roll recrystallisation temperature in such a way that equivalent grain sizes were produced in two alloy heats with Nb levels of 0.46 and 0.74 wt.% respectively. Although the grain size was essentially stabilised by recrystallisation between 1050-1100 C for 30 minutes, the solution treatment prior to creep testing was varied for each heat to evaluate not only the influence of bulk Nb level on creep resistance, but also to consider the influence of the distribution of Nb in the microstructure. Consequently, the total heat treatment cycle prior to constant load creep testing at 850 C involved recrystallisation, ageing at 700 C, and final solution treatment at 950, 1000 or 1050 C for 200 seconds. The microstructure after the different heat treatments was investigated using light microscopy, scanning electron microscopy (SEM) and electron backscattered diffraction (EBSD). The latter was particularly useful in accurately measuring grain size. The microstructural evolution of both alloys during creep testing was also monitored. This was done in order to examine the microstructural changes that occur during the prolonged creep testing period. Not surprisingly, the creep tests at initial stresses of 5, 10 and 15 MPa all revealed greater creep resistance for the higher Nb-containing alloy heat. However, the correlation with solution treatment practice was much less obvious, particularly for the alloy with the lower Nb content. Detailed analysis of the precipitate distribution after the various heat treatments is presented to illustrate the difference in microstructure that can arise and consequently consideration is given to the influence of precipitation on creep behaviour