Scouring erosion resistance of metallic materials with composite microstructures

Abstract

Slurry erosion can cause severe damage and often constitutes a significant portion of the total production cost in many industries such as oil sands and mining. The most commonly used materials for such applications are metallic materials having a composite microstructure consisting of hard reinforcement phases in a more ductile metal matrix. Typical examples are high chromium white irons and tungsten carbide welding overlays. The main purpose of this work is to evaluate the scouring erosion resistance of such materials for applications in handling fine sand slurries and to introduce a wear model to help understand the correlation between erosion resistance and the characteristics of microstructures.Five high chromium white cast irons and four spray and fusion coatings composed of cast tungsten carbide particles and Ni–Cr–B–Si–Fe matrix have been studied using a Coriolis scouring erosion tester in a fine sand slurry. The relationship between the erosion resistance and different microstructure parameters was explored. For the chromium white cast irons, the erosion resistance increased with increase in carbide volume fraction but the spray and fusion coatings showed the opposite trend, i.e., their erosion resistance decreased with increase in the carbide volume fraction. No clear correlation was found between erosion resistance and free path for the spray and fusion coatings while for the chromium white cast irons, the erosion resistance was found to increase rather than decrease with increase in free path.To help explain the experimental results, a wear model has been proposed for the wear resistance of composite microstructures composed of a reinforcement phase (RP) and a metal matrix. In the model, the wear surface of the material is distinguished by four different features/regions, each having unique wear resistance. The rule of mixtures (ROM) for wear resistance is applied to formulate the wear resistance of the material with the above four wear surface features. It is shown that the results can be reasonably well explained based on the proposed model by using the total edge length, L, of the RP particles exposed on a unit area of the wear surface.