Dilatometry Study of a High-Chromium Cast Iron

Sammanfattning: High-chromium cast irons are used in certain applications where the demand on abrasion resistance is high. Such applications can be found in the milling industry and in pumps for transport of abrasive particles in liquid suspension. Soft annealed high-chromium cast iron containing 2.6 % C and 24.7 % Cr was supplied by Xylem Water Solutions, Sundbyberg, and investigated by dilatometry. The heat treatments were inspired by induction hardening procedures. The purpose of the investigation was to evaluate the effect of maximum temperature reached during heat treatment on the final length of the test specimen. The aim with this was to find the treatment yielding the maximum possible length which should be profitable to create desirable compressive stresses in the surface hardened area. The experimental results were used to create a finite element model in COMSOL Multiphysics accommodating for the maximum temperature, simulating the phase changes occurring in a geometry based on the experimental test specimen. The experimental results did not reveal any clear correlation between the maximum temperature and the final length change. The hardness, however, increased with the increasing temperature in the treatment interval 900-1150 °C. The, by light optical microscopy, observed amount of secondary precipitated carbides decreased with increasing temperature. Martensite transformation was also affected; the transformation temperature decreased for increased treatment temperatures. From dilatometry it was also seen that the thermal strains were greatly affected by the direction of which the material was cut from the original cast material. Samples taken perpendicular to the mainly investigated direction showed lower coefficients of thermal expansion and the final strain was clearly positive compared to the slightly negative values found for the main direction. This phenomenon could possibly be explained by different macrostructures created during solidification of the melt causing anisotropy in the eutectic. The implementation in COMSOL by describing the phase transformation as ordinary differential equations did show partially good results in the simulation of thermal expansion. The difference in original material is noticeable in the dilatometry and the simulated martensite transformation deviates from the experimental results. The model needs to be validated against new intermediate test temperatures and the martensite transformation kinetics must be investigated further to yield better results to be able to combine the phase transformations with mechanical calculations.