Erosion of oxide scales formed on Cr₃C₂–NiCr thermal spray coatings

Abstract

Cr₃C₂–NiCr thermal spray coatings are extensively used to mitigate high temperature erosive wear in fluidised bed combustors and power generation/transport turbines. The aim of this work was to characterise the variation in oxide erosion response as a function of the Cr₃C₂–NiCr coating microstructure. Erosion was carried out at 700 °C and 800 °C with erodent impact velocities of 225–235 m/s. The erosion behaviour of the oxide scales formed on these coatings, was influenced by the coating microstructure and erosion temperature. Development of the carbide microstructure with extended heat treatment lead to variations in the erosion–corrosion response of the Cr₃C₂–NiCr coatings.

Introduction

Cr₃C₂–NiCr thermal spray coatings are extensively used to mitigate high temperature erosive wear in fluidised bed combustors and power generation/transport turbines. The useful operating temperature range of these coatings starts at the upper limit of the WC based cermet system (450–500 °C [1], [2], [3], [4], [5]) to a maximum of 850–900 °C [6], [7], [8], [9], [10], [11]). At such temperatures oxide scale development and response to erosion becomes increasingly important, especially when the duty lifetime is considered. The combination of erosion and high temperature oxidation generate a wear phenomenon termed “erosion–corrosion”, whereby the extent of degradation is greater then the sum of the two mechanisms acting independently [12], [13], [14], [15], [16], [17], [18], [19], [20]. A broad range of mechanisms have been defined to characterise the response under these conditions, ranging from erosion dominated behaviour through to oxidation dominated behaviour [15], [16], [17]. The most industrially relevant mechanisms involve a more balanced contribution from each mechanism, commonly being defined as “oxidation-affected-erosion” and “erosion-affected-oxidation” [12], [21]. In both instances the oxide response to erosion impact is critical to determining the extent of degradation [17]. This in turn is dependent upon the erosion conditions, the mechanical properties of the oxide at the exposure temperature and the response of the substrate to erosion [16], [18], [19], [22], [23]. From a practical viewpoint mitigation of erosion–corrosion requires the formation of a slow growing, adhesive oxide scale that is sufficiently ductile at the operating temperature to undergo deformation by impact without extensive cracking or spalling [18], [19].

The variation in mechanical properties of the oxide, and underlying substrate, with temperature plays a critical role in the mechanism of degradation. The ductile to brittle transition temperature of the substrate determines the degree of deformation upon impact. High ductility results in plastically deformed mounds and platelets to which the oxide may find it difficult to adhere. Low ductility substrates focus the impact load within the oxide and the oxide–substrate interface by not dissipating the impact load through plastic deformation [18]. The same principle applies to the response of the oxide phase itself. Oxide plasticity during erosion impact has been reported for several high temperature studies for chromium oxide [16], [18], [24], nickel oxide [25], [26] and iron based alloy scales [27], [28]. In an industrial system then, the degree of erosion deformation is determined largely by the temperature and the effective modulus of the oxide and substrate combination. The modulus of that combination is dictated primarily by the scale thickness [16], [17], [29], [30].

The complexity of erosion–corrosion and the variation in response with different service conditions means that most reported work is based on the assessment of candidate materials in comparative trials. Few studies have investigated the wear phenomenon from a mechanistic, microstructural viewpoint [16], [19], [24], [30]. This is especially true for thermal spray coatings and for carbide based coatings in particular.

The aim of this work was to characterise the variation in oxide erosion response as a function of the Cr₃C₂–NiCr coating microstructure. Two powders were used which span the range of those used industrially – an agglomerated/sintered powder, characterised by a homogeneous distribution of fine Cr₃C₂ grains in a NiCr matrix, and a blended powder mixed from pure Cr₃C₂ and Ni–20Cr powders. The first represents the current state of technology while the latter represents the coating structure upon which the initial industrial application of this cermet composition was based. Previous works [31], [32], [33] have highlighted how the as-sprayed microstructure of these coatings continues to develop as a function of in-service exposure at elevated temperature. To determine how such changes influence the erosion–corrosion response several coatings were heat treated to achieve a steady state microstructure representative of that formed with extended in-service exposure. Specimens were oxidised for periods of up to 48 h in order to develop oxides of varying thickness prior to testing under high velocity erosion conditions.