Single Crystal Superalloys: The Transition from Primary to Secondary Creep

Abstract

In this paper we compare the effect of stress, orientation and alloy composition on primary and secondary creep rate in a number of single crystal alloys. The activation energies and stress dependence of primary and secondary creep, measured in the alloy CMSX-4, show a distinctive low activation energy associated with stacking fault shear during primary creep which disappears at low stress. Twostage creep tests show that a fully developed dislocation network, introduced into the microstructure by limited deformation at 950 C, results in a substantial drop in primary creep strain during subsequent testing at 750 C. Comparison with identically-oriented samples without prestrain confirms the role of dislocation networks in the channels as the major hardening mechanism and shows that secondary creep is unaffected by the pre-strain. These observations are inconsistent with existing understanding of the transition between primary and secondary creep in some important respects. We argue that the mechanisms producing primary and secondary creep strain are similar, but that the secondary creep rate is determined by the rate of formation of defects at the / interface. This enables many aspects of the dependence of creep rate on orientation, alloy composition and micro-structure to be rationalised.

Non-Isothermal Creep Behavior of a Second Generation Ni-Based Single Crystal Superalloy: Experimental Characterization and Modeling

“…Actually, Ni-based single crystal superalloys exhibit superior creep resistance at temperature as high as 1100°C [2] due to their particular microstructure, consisting of 70% in volume of coherently precipitated γ' cubes (L1 2 lattice) separated by thin channels of face centered cubic (fcc) γ matrix [3]. Extensive literature is devoted to the study of their properties, especially on isothermal creep over a wide range of stress and temperature [4][5][6][7][8][9][10][11]. Two temperature/stress domains can be distinguished: (1) at low temperature (T < 950°C) and high stress, no significant evolution of the initial optimal microstructure occurs during the creep experiment, while (2) at higher temperature/lower stress, great modifications of the microstructure are observed during primary creep stage: rafting of the γ' phase perpendicular to the applied stress for negative γ/γ' misfit alloys and development of stable dislocation networks at the γ/γ' interfaces.…”

Secondary Creep of Thin-walled Specimens Affected by Oxidation

“…However as the γ′-volume fraction increases above 50%, the dislocation movement becomes restricted to the narrow γ channels and the density is governed by the local conditions in the channels. Norton creep exponents in high volume fraction superalloys such as CMSX-4 are typically much higher than conventional value of 4-6; values as high as 13 have been measured at the somewhat lower temperature of 850ºC [32]. > 50%) by-pass mechanism by Orowan bowing can be assumed as dominating mechanism.…”

Dislocations in a Ni-based superalloy during low temperature creep

“…The primary slip plane was identified using Laue X-ray imaging and the test-piece sectioned vertically, parallel to the tensile axis. For high resolution imaging the primary slip plane (-111) was perpendicular to the foil normal [110] and the Burgers vector of the highest Schmid Factor [1][2][3][4][5][6][7][8][9][10][11][12](-111) slip system lies in the plane of the foil. Hence an edge dislocation in this system a/3 [1][2][3][4][5][6][7][8][9][10][11][12] has its line vector parallel to the foil normal and would be in the ideal configuration to image the atom alignment at high resolution.…”

“…For high resolution imaging the primary slip plane (-111) was perpendicular to the foil normal [110] and the Burgers vector of the highest Schmid Factor [1][2][3][4][5][6][7][8][9][10][11][12](-111) slip system lies in the plane of the foil. Hence an edge dislocation in this system a/3 [1][2][3][4][5][6][7][8][9][10][11][12] has its line vector parallel to the foil normal and would be in the ideal configuration to image the atom alignment at high resolution. The sample was also cut on the other vertical section normal to [1][2][3][4][5][6][7][8][9][10] and horizontally normal to [001] for imaging the dislocations and stacking faults at lower magnification.…”

“…But in the important low temperature regime, experienced in the hottest part of the engine where the high pressure turbine blades are cooled, dislocations are able to enter the γ . Furthermore a large part of the strain derives from the cutting of the precipitates by suitable partial dislocations [4,5]. The role of stacking fault shear in superalloys has received increasing attention with the focus on high cobalt superalloys that exhibit this mode of behaviour in both tensile and creep deformation [6].…”