On the Roles of Oxidation and Vaporization in Surface Micro-structural Instability during Solution Heat Treatment of Ni-base Superalloys

On the Roles of Oxidation and Vaporization in Surface Micro-structural Instability during Solution Heat Treatment of Ni-base Superalloys NEIL D'SOUZA 0 DEAN WELTON 0 GEOFF D. WEST 0 IAN M. EDMONDS 0 HANG WANG 0 0 NEIL D'SOUZA , Materials Technologist, DEAN WELTON, Turbines Metallurgist, and IAN M. EDMONDS , Engineering Manager Materials (Nuclear) , are with the Rolls-Royce plc, PO Box 31, Derby, DE24 8BJ , U.K. GEOFF D. WEST, Research Fellow, is with the Loughborough University , Loughborough, Leicestershire LE11 3TU , U.K. HANG WANG, is with the School of Metallurgy and Materials, University of Birmingham , Birmingham B15 2TT , U.K. Contact Micro-structural instability at the surface that develops during solution heat treatment of a typical third generation Ni-base superalloy, CMSX10N has been reported. It is shown that elemental Ni vaporizes from the surface during solutioning leading to de-stabilization of c phase. With increasing extent of vaporization, a phase mixture of b, c, and the refractory (W and Re-rich) precipitates occur within the surface layers resulting in the complete breakdown of the cuboidal c/c phase morphology that is usually observed. It is demonstrated that the conditions at the surface have a marked effect on the vaporization kinetics and subsequent evolution of surface phases-the presence of a continuous dense oxide such as Al2O3 or the presence of sacrificial Ni-foils interspersed in the furnace significantly suppresses elemental vaporization from the sample surface. - recently shown that a very different response is exhibited by the surface, where an anomalous microstructure develops following solutioning.[11] Specifically, de-stabilization of c phase was observed, and the surface layer showed a mixture of c and refractory-rich precipitates, which is markedly absent within the bulk of the sample. The stabilization of c arises from enrichment in Al, while the formation of refractory-rich TCP phases incorporates the excess W and Re. To account for the evolution of this microstructure, it was proposed therefore that volatilisation of Ni and Cr occurs from the surface during solutioning, which is consistent with the high vapor pressures for both these elements in vacuum.[12] It is common practice therefore for solution heat treatment to be carried out under a partial pressure of an inert gas, such as Ar rather than in vacuum to prevent the volatilisation of high vapor pressure elements, such as Cr from the surface of a casting. A particular application where this principle is critical is in brazing, where the elemental vapor pressure is quite high at the melting point of these alloys.[13, 14] Other applications, where elemental vaporization is also critical is in electron beam melting; e.g., in refining of Tibase alloys, where evaporation of Ti from the liquid residing in the hearth occurs, in a vacuum atmosphere.[15] This clearly demonstrates that surface effects need to be considered, and the notion of a closed system for mass balance considerations might not necessarily be correct. However, notwithstanding this preliminary study,[11] many aspects remain un-answered. The principal ones being; Solutioning is not carried out strictly in vacuum, but there is the localised circulation of Ar (carrier gas) and therefore, vaporization kinetics cannot be determined in a straightforward manner using the Langmuir equation.[16] The nature and type of the as-cast oxide will have a profound effect on the rate of vaporization; the casting surface is covered in part by an Al2O3 reaction layer, while in some regions, where mold/metal separation had occurred during casting, a scaled NiO layer is observed which arises from subsequent oxidation of the bare surface.[17] The as-cast microstructure at the surface of the casting varies; in some cases a layer of eutectic is observed proud of the secondary dendrite arms, unlike in other instances, the microstructure comprises dendrites and intermittent inter-dendritic channels (eutectic phases) that abut the surface.[18] The purpose of this article is to elucidate these aspects more clearly. Specifically; (i) The occurrence of vaporization is demonstrated unambiguously through a novel experiment where the Ni vapor pressure in the furnace chamber can be varied to study the effect on vaporization. (ii) The role of the surface oxide in vaporization is also demonstrated through carefully controlled pre-oxidation experiments and possible mechanisms assessed. (iii) Finally, the resultant solute re-distribution within the surface layers that follow vaporization is quantitatively dealt with using a simplified approach to determine the evolution of phases, and limitations of this method are outlined. In this study, the third generation alloy, CMSX10N whose nominal composition is given in Table I is considered owing to its prevalence as the alloy of preference for intermediate pressure turbine blades, where this effect is often observed. It is also worth highlighting that there are consequences arising from such an anomalous morphology on the surface of turbine blades developing during heat treatment. The decreased aerofoil section thickness with the optimum cuboidal c/c morphology results in a reduction in thickness of the load-bearing cross-section of the aerofoil during service and subsequently having a detrimental effect on creep properties. II. EXPERIMENTAL A. Casting and Heat Treatment Experiments Directional solidification was carried out at the Precision Casting Facility, Rolls-Royce, plc, Derby, UK. Cylindrical test bars (70 mm in length with a 10 mm diameter) were cast in an industrial directional solidification furnace. The basic parameters included the following; furnace heater temperature nominally 1773 K (1500 C), vacuum level of the furnace chamber was maintained at 0.1 Pa (10 6 atm), and a mold withdrawal rate of around 5 10 5 m s 1 was used. Additional details are available elsewhere.[14] Solution heat treatment of the samples was carried out in a TAV vacuum furnace at the University of Birmingham, UK. All solutioning experiments were conducted in an Ar atmosphere at a pressure of 2 10 4 atm (0.2 mBar). The test bars were located in alumina boats and positioned in the center of the furnace, which was fitted with eight thermocouples at the corners and one at the center of the load. The heat treatment cycle consisted of a ramp with intermediate temperature holds followed by an isothermal hold at the final solutioning temperature, 1633 K (1360 C). The time corresponding to the ramp cycle was 10 hour and this was subsequently followed by a 1-hour hold at 1633 K (1360 C). Quenching at the end of the soak cycle was done using a forced argon flow resulting in a cooling rate of approximately 90 K min 1. In a second experiment, the test bars were positioned in alumina boats as before; however, dispersed sheets of Ni-foils were interspersed between the test bars. The thickness of the foil was 100 lm. In yet (...truncated)