Effect of specimen thickness on the creep response of a Ni-based single-crystal superalloy

Creep tests on Ni-based single-crystal superalloy sheet specimens typically show greater creep strain rates and/or reduced strain or time to creep rupture for thinner specimens than predicted by current theories, which predict a size-independent creep strain rate and creep rupture strain. This size-dependent creep response is termed the thickness debit effect. To investigate the mechanism of the thickness debit effect, isothermal, constant nominal stress creep tests were performed on uncoated PWA1484 Ni-based single-crystal superalloy sheet specimens of thicknesses 3.18 and 0.51 mm under two test conditions: 760 °C/758 MPa and 982 °C/248 MPa. The specimens contained initial microvoids formed during the solidification and homogenization processes. The dependence of the creep response on specimen thickness differed under the two test conditions: at 760 °C/758 MPa there was a reduction in the creep strain and the time to rupture with decreasing section thickness, whereas at 982 °C/248 MPa a decreased thickness resulted in an increased creep rate even at low strain levels and a decreased time to rupture but with no systematic dependence of the creep strain to rupture on specimen thickness. For the specimens tested at 760 °C/758 MPa microscopic analyses revealed that the thick specimens exhibited a mixed failure mode of void growth and cleavage-like fracture while the predominant failure mode for the thin specimens was cleavage-like fracture. The creep specimens tested at 982 °C/248 MPa in air showed the development of surface oxides and a near-surface precipitate-free zone. Finite-element analysis revealed that the presence of the alumina layer at the free surface imposes a constraint that locally increases the stress triaxiality and changes the value of the Lode parameter (a measure of the third stress invariant). The surface cracks formed in the oxide scale were arrested by further oxidation; for a thickness of 3.18 mm the failure mode was void nucleation, growth and coalescence, whereas for a thickness of 0.51 mm there was a mixed mode of ductile and cleavage-like fracture.     

High-temperature and low-stress creep anisotropy of single-crystal superalloys

The high-temperature and low-stress creep (1293 K, 160 MPa) of the single-crystal Ni-based superalloy LEK 94 is investigated, comparing the tensile creep behavior of miniature creep specimens in [0 0 1] and [1 1 0] directions. In the early stages of creep, the [0 0 1]-direction loading shows higher minimum creep rates, because a greater number of microscopic crystallographic slip systems are activated, the dislocation networks at γ/γ′ interfaces accommodate lattice misfit better, and γ channels are wider. After the creep rate minimum, creep rates increase more strongly as a function of strain for [1 1 0] tensile loading. This may be related to the nature of rafting during [1 1 0] tensile creep, which results in a more open topology of the γ channels. It may also be related to more frequent γ′ cutting events compared with [1 0 0] tensile creep.     

Tensile creep of directionally solidified NiAl–9Mo in situ composites

Well-aligned Mo fiber-reinforced NiAl in situ composites were produced by specially controlled directional solidification. The creep behavior parallel to the growth direction was studied in static tensile tests at temperatures between 900 °C and 1200 °C. A steady-state creep rate of 10^?6 s^?1 was measured at 1100 °C under an initial applied tensile stress of 150 MPa. Compared to binary NiAl and previously investigated NiAl–Mo eutectics with irregularly oriented Mo fibers, this value demonstrates a remarkably improved creep resistance in NiAl–Mo with well-aligned unidirectional Mo fibers. A high-resolution transmission electron microscope investigation of the NiAl/Mo interface revealed a clean semi-coherent boundary between NiAl and Mo, which enabled an effective load transfer from the NiAl matrix to the Mo fibers, and thus leads to the remarkably increased creep strength. The stress exponent, n, was found to be between 3.5 and 5, dependent on temperature. The activation energy for creep, Q_c, was measured to be 291 ± 19 kJ mol^–1, which is close to the value for self-diffusion in binary NiAl. Transmission electron microscopy observations substantiated that creep occurred by dislocation climb in the NiAl matrix. The Mo fiber was found to behave in a quasi-rigid manner during creep. A creep model for fiber-reinforced metal matrix composites was applied for an in-depth understanding of the mechanical behavior of the individual components and their contribution to the creep strength of the composite.     

The effect of thickness on the steady-state creep properties of freestanding aluminum nano-films

To clarify the effects of film thickness on the creep properties of nano-films we conducted tensile creep experiments on freestanding aluminum films with thickness values in the range ～100–800 nm at room temperature. The nano-films showed typical creep behavior comprising transient, steady-state, and accelerated creep stages. The steady-state creep exponents of the 100–800 nm thick specimens were 0.84–2.7 in the stress range 30–120 MPa, which are close to the value for diffusion creep (1). Creep deformation clearly shows a thickness effect: the steady-state creep rate increases as the thickness decreases from 800 to 400 nm, shows a peak in the range 400–200 nm, and then decreases in the 200–100 nm thickness range. The creep experiments under a small stress of 1 MPa show a negative strain rate, indicating the presence of a driving force to reduce the surface area due to surface tension. The explanation for the thickness effect is as follows. Since the ratio of surface and grain boundary area to volume increases with decreasing thickness, diffusion creep along these paths is enhanced, resulting in an increase in the creep rate. As the thickness decreases to 200–100 nm, however, the surface tension effect to reduce the surface area becomes dominant, decreasing the creep rate. In addition, the creep rate of the nano-films is about two or three orders of magnitude smaller than that of the bulk material dominated by the dislocation creep mechanism.     

Modelling and analysis of the oxidation influence on creep behaviour of thin-walled structures of the single-crystal nickel-base superalloy René N5 at 980 °C

Creep test results of thin-walled specimens of the single-crystal nickel-base superalloy René N5 at 980 °C under vacuum as well as under air show different creep properties depending on material thickness and atmosphere. The differences in creep strength and strain were analysed based on a creep-oxidation model. The model specifies the primary and secondary creep stages of thin-walled specimens by a sequence of layers. The model takes different zones affected by oxidation into account. Four layers were experimentally observed and considered in the model: oxide layer, γ′-free layer, γ′-reduced layer and the two-phase substrate in the sample as centre. Material parameters for growth laws of each layer were identified both by experimental analyses and by thermodynamic simulations. The final creep-oxidation model characterizes the creep behaviour of samples with small thicknesses and low initial stress with high accuracy.     

Lattice strain evolution during creep in single-crystal superalloys

In situ neutron diffraction studies are carried out to characterize the micromechanical deformation occurring during tensile creep of a typical single-crystal nickel-based superalloy, CMSX-4. The loading responses of the matrix γ phase and the precipitate γ′ are distinct. Moreover, the behaviour in the tertiary creep regime (in which the γ′ phase remains intact) is qualitatively different from that in the primary creep regime (when γ′ is sheared). In tertiary creep, initial deformation of the matrix leads to a release of misfit between the phases in the (1 0 0), resulting in elastic compression of the γ in the loading direction. The load state then remains fairly constant during creep. During the initial stages of primary creep, elastic compression of the γ phase is observed until at around 2–4% creep strain this compression stabilizes as the (1 0 0) misfit is released. This is the point at which γ′ shearing is thought to begin. Subsequently, the load in the γ increases by around 200 MPa until a maximum is reached at around 8% creep strain. This load is then suddenly released, which may be due to the release of back-stress.     

Creep and microstructure of Nextel™ 720 fiber at elevated temperature in air and in steam

Creep of Nextel? 720 alumina–mullite fiber tows was investigated at 1100 and 1200 °C for tensile stresses of 100–400 MPa in air and in steam. Fiber microstructures were characterized after creep by transmission electron microscopy. At low stresses steam increased creep rates by up to an order of magnitude and reduced creep lifetimes. At high stresses creep rates in steam and air were similar. Cavitation was prevalent in steam but not in air. The creep-rupture data obtained at 1200 °C were analyzed in terms of a Monkman–Grant (MG) relationship. The MG parameters were independent of the test environment. Results reveal that the MG relationship can be used to predict creep rupture for Nextel? 720 fibers and composites reinforced with these fibers at 1200 °C in air and in steam. In steam the mullite in the Nextel? 720 fibers decomposed to porous alumina. Decomposition kinetics were linear and had an activation energy of ～200 kJ mol^?1. Intergranular films were not observed on alumina grain boundaries or alumina–mullite interphase boundaries after creep in steam. Creep mechanisms are discussed.     

Mechanical properties of dual multi-phase single-crystal intermetallic alloy composed of geometrically close packed Ni3X (X: Al and V) type structures

Dual multi-phase intermetallic alloy, which is composed of Ni₃Al(L1₂) and Ni solid solution (A1) phases at high-temperature annealing and is additionally refined by a eutectoid reaction at low temperature aging, according to which the Al phase is transformed into the Ni₃Al(L1₂) + Ni₃V(DO₂₂) phases, was prepared based on the pseudo-ternary system Ni₃Al–Ni₃Ti–Ni₃V. The high-temperature tensile deformation, fracture behavior and tensile creep were investigated using single crystalline material. The alloy with such a novel microstructure shows extremely high yield and tensile strength with good temperature retention, when compared not only with conventional Ni-based superalloys but also with polycrystalline materials reported previously. Over a broad temperature range fracture occurred along octahedral plane in the major component L1₂ phase, accompanied with high tensile elongation and ductile fracture mode. The tensile creep test conducted at 1173 K and 1223 K showed the presence of threshold stress, and also extremely low creep rate and long creep rupture time when compared with conventional Ni-based superalloys. The obtained results are promising for the development of a new-type of high-temperature structural material.     

Atomistic simulations of diffusional creep in a nanocrystalline body-centered cubic material

Molecular dynamics (MD) simulations are used to study diffusion-accommodated creep deformation in nanocrystalline molybdenum, a body-centered cubic metal. In our simulations, the microstructures are subjected to constant-stress loading at levels below the dislocation nucleation threshold and at high temperatures (i.e., T > 0.75T_melt), thereby ensuring that the overall deformation is indeed attributable to atomic self-diffusion. The initial microstructures were designed to consist of hexagonally shaped columnar grains bounded by high-energy asymmetric tilt grain boundaries (GBs). Remarkably the creep rates, which exhibit a double-exponential dependence on temperature and a double power-law dependence on grain size, indicate that both GB diffusion in the form of Coble creep and lattice diffusion in the form of Nabarro–Herring creep contribute to the overall deformation. For the first time in an MD simulation, we observe the formation and emission of vacancies from high-angle GBs into the grain interiors, thus enabling bulk diffusion.     

Void initiation in fcc metals: Effect of loading orientation and nanocrystalline effects

It is shown, through molecular dynamics simulations, that the emission and outward expansion of special dislocation loops, nucleated at the surface of nanosized voids, are responsible for the outward flux of matter, promoting their growth. Calculations performed for different orientations of the tensile axis, [0 0 1], [1 1 0] and [1 1 1], reveal new features of these loops for a face-centered cubic metal, copper, and show that their extremities remain attached to the surface of voids. There is a significant effect of the loading orientation on the sequence in which the loops form and interact. As a consequence, the initially spherical voids develop facets. Calculations reveal that loop emission occurs for voids with radii as low as 0.15 nm, containing two vacancies. This occurs at a von Mises stress approximately equal to 0.12G (where G is the shear modulus of the material), and is close to the stress at which dislocation loops nucleate homogeneously. The velocities of the leading partial dislocations are measured and found to be subsonic (～1000 m s^?1). It is shown, for nanocrystalline metals that void initiation takes place at grain boundaries and that their growth proceeds by grain boundary debonding and partial dislocation emission into the grains. The principal difference with monocrystals is that the voids do not become spherical and that their growth proceeds along the boundaries. Differences in stress states (hydrostatic and uniaxial strain) are discussed. The critical stress for void nucleation and growth in the nanocrystalline metal is considerably lower than in the monocrystalline case by virtue of the availability of nucleation sites at grain boundaries (von Mises stress ～0.05G). This suggests a hierarchy of nucleation sites in materials, starting with dispersed phases, triple points and grain boundaries, and proceeding with vacancy complexes up to divacancies.     

Microstructures of Rene 142 nickel-based superalloy fabricated by electron beam melting

Rene 142, a commercial, columnar grained, gas turbine airfoil Ni-based superalloy, has been fabricated from a pre-alloyed, atomized powder by additive manufacturing using electron beam melting. Monolithic components having [0 0 1] oriented, columnar grain structures exhibited a creep-optimized 59% volume fraction of cuboidal, coherent, γ′-phase precipitates averaging 275 nm on the side, and with γ/γ′ channel widths ranging from 25 to 75 nm. Transmission electron microscopy, utilizing bright and dark field imaging of optimally oriented γ/γ′ interfaces showed prominent misfit coherency strains as δ-fringe patterns. Corresponding hardness measurements also indicated the possibility of creep strength comparable with the commercial alloy. The notable feature of this study was the monolithic development of desirable superalloy properties without conventional, multi-step heat treatments.     

Hold-time effect on the elevated-temperature crack growth behavior of solid-solution-strengthened superalloys

The fatigue crack growth behavior of two solid-solution-strengthened superalloys, Ni-based HAYNES^® 230 and HASTELLOY^® X, was studied at 816 and 927 °C in laboratory air. The fatigue crack growth tests were conducted following a baseline triangular waveform of 0.33 Hz. Various hold times were introduced at the maximum load to study the hold-time effect. Fracture mechanics parameters, K, C¹, C_t, and (C_t)_avg, were applied to correlate the crack growth rates at different temperatures for both HAYNES 230 and HASTELLOY X alloys. For both alloys, the fatigue cracking path was mainly transgranular at 816 and 927 °C. The cracking path became dominantly intergranular if the hold time increased to 2 min, indicating that the time-dependent creep damage mechanisms were in control. When the time-dependent damage dominated (temperature ⩾816 °C and hold time ⩾2 min), the crack growth rates can be correlated with C_t or (C_t)_avg parameters. The C_t and (C_t)_avg parameters were capable of consolidating data from different temperatures and different alloys.     

Creep behaviour of a new air-hardenable intermetallic Ti–46Al–8Ta alloy

Creep behaviour of a new cast air-hardenable intermetallic Ti–46Al–8Ta (at.%) alloy was investigated. Constant load tensile creep tests were performed at initial applied stresses ranging from 200 to 400 MPa in the temperature range from 973 to 1073 K. The minimum creep rate is found to depend strongly on the applied stress and temperature. The power law stress exponent of the minimum creep rate is n = 5.8 and the apparent activation energy for creep is calculated to be Q_a = (382.9 ± 14.5) kJ/mol. The kinetics of creep deformation of the specimens tested to a minimum creep rate (creep deformation about 2%) is suggested to be controlled by non-conservative motion of dislocations in the γ(TiAl) matrix. Besides dislocation mechanisms, deformation twinning contributes significantly to overall measured strains in the specimens tested to fracture. The initial γ(TiAl) + α₂(Ti₃Al) microstructure of the creep specimens is unstable and transforms to the γ + α₂ + τ type during creep. The particles of the τ phase are preferentially formed along the grain and lamellar colony boundaries.     

Influence of casting surface on creep behaviour of thin-wall Ni-base superalloy Inconel100

The conventionally cast nickel-base superalloy Inconel100 was investigated with the focus on the influence of the surface condition and the thin-section size on the creep properties. Flat samples with 0.9 mm and 1.3 mm thickness were wire eroded from precision cast ingots. Thin cast specimens with a remaining casting surface were compared with machined specimens which flat surfaces were subsequently ground to the final thicknesses. Microstructures were examined by optical and scanning electron microscopy. Creep tests in air at 980 °C under a constant load of 150 MPa revealed decreasing creep strength with decreasing sample thickness. However, thin cast samples and samples with a machined surface having the same thickness showed similar creep properties.     

Influence of minor aluminum concentration changes in zirconium-based bulk metallic glasses on the elastic,anelastic, and plastic properties

The Poisson’s ratio of Zr-based bulk metallic glasses in the system Zr_63?xCu₂₄Al_xNi₁₀Co₃ was found to exhibit a non-monotonous behavior as a function of x when measured with ultrasound by the pulse–echo technique. In addition, from wave propagation velocity measurements at different frequencies, i.e. f = 2.25 MHz and f = 10 MHz, a composition-dependent anelastic behavior as a function of x is found, exhibiting a similar non-monotonous behavior. In this work we further investigated the plastic deformation and the creep properties of this glass system in compression tests and by nanoindentation. The plastic strain and the measured creep deformation show correlations with the Poisson’s ratio. We then discuss the anelastic behavior observed while measuring the sound-wave propagation velocity in the frame of the thermoelastic damping and the bond reorientation as proposed by Egami. Finally we discuss these effects with regard to X-ray diffraction analysis.     

On the effect of long-term creep on the microstructure of a 12% chromium tempered martensite ferritic steel

In the present study we investigate the evolution of the microstructure of a 12% Cr tempered martensite ferritic steel under conditions of long-term aging and creep (823 K, 120 MPa, t_R = 139,971 h). We show how subgrains coarsen, that the close correlation between carbides and subgrain boundaries loosens during long-term creep and that the frequency of small-angle boundaries increases. All these elementary deformation processes have been discussed in short-term creep studies. The present study shows that they also govern long-term creep. However, during long-term creep, precipitation and coarsening reactions occur that are not observed during short-term creep. Three types of particles (M₂₃C₆, VX and Laves-phase) were identified after long-term creep. M₂₃C₆ particles coarsen at constant volume fraction and establish their equilibrium concentration after 51,072 h; VX particles are stable; and the Laves-phase particles never reach thermodynamic equilibrium.     

Ni–(Zr/Hf)–(Nb/Ta)–Al bulk metallic glasses with high thermal stabilities

Thermal stability is a critical consideration in the application of metallic glasses as hydrogen separation material. The development of new Ni-based bulk metallic glasses (BMGs) with enhanced thermal stability is desirable. The present work investigated the alloying effects of refractory metals Hf and Ta on the Ni₆₀Zr₂₀Nb₁₅Al₅ bulk metallic glass. Two serial alloys, namely, Ni₆₀Zr_20 ? xHf_xNb₁₅Al₅ (x = 0～20 at.%) and Ni₆₀Zr₂₀Ta_yNb_15 ? yAl₅ (y = 0～15 at.%), were investigated in the present work. The addition of Hf or Ta was revealed to be effective in improving the thermal stability of the basic alloy, while the glass-forming ability of the alloy is slightly reduced resulting from the addition of Hf or Ta. The possible mechanism of these alloying effects is discussed.     

On the effect of grain boundary segregation on creep and creep rupture

The present work investigates the effect of grain boundary chemistry and crystallography on creep and on creep damage accumulation in Cu–0.008 wt.% Bi and Cu–0.92 wt.% Sb at stresses ranging from 10 to 20 MPa and temperatures between 773 and 873 K. Small additions of Bi and Sb significantly reduce the rupture strain and rupture time during creep of Cu. High stress exponents (Cu–Bi) and high apparent activation energies for creep (Cu–Bi and Cu–Sb) are obtained. Sb promotes creep cavitation on random high-angle grain boundaries. Bi, on the other hand, causes brittle failure when small crack-like cavities cause decohesion. Both elements suppress dynamic recrystallization, which occurs during creep of Cu at high stresses and temperatures.     

Observations of a〈0 1 0〉 dislocations during the high-temperature creep of Ni-based superalloy single crystals deformed along the [0 0 1] orientation

A NASAIR-100 superalloy single crystal was tested in tension creep at 1000 °C at a stress of 148 MPa, for a time period of 20 h and to a strain of 1.1%. Analysis of the resulting dislocation structures after rafting was completed reveals the frequent presence of all three types of a〈0 1 0〉 dislocations in the γ′ particles. Two of these families experience no resolved forces due to the applied stress. It is proposed that these a〈0 1 0〉 dislocations form as a result of the combination of two dissimilar a/2〈0 1 1〉 dislocations entering from γ channels. The possible driving forces for the movement of these a〈0 1 0〉 dislocations are discussed, and a novel recovery mechanism during creep of rafted microstructures is introduced on the basis of these observations.     

Low-temperature processing of Fe–Al intermetallic coatings assisted by ball milling

A novel technique has been developed to produce Fe–Al intermetallic coatings on steel. This technique applies mechanical vibration to a retort, which is loaded with Al powder, alumina filler, ammonium chloride activator and FeCrAl alloy balls. The operation temperature was from 440 to 600 °C. This technique produced coatings with thickness of 17 μm for 15 min and 90 μm for 120 min treatment at 560 °C. The coatings appear to be homogeneous, with a high density and free of porosity, and have excellent adherence to the substrate. The coatings consisted mainly of η-Fe₂Al₅ with small amounts of θ-FeAl₃ and β-FeAl, and exhibited a nano-structure. Microstructure studies suggested that the formation of the intermetallic phases at a low temperature has a complex mechanism, including the formation of a thin Al layer on the substrate by ball milling; Al-rich phases nucleation, growth and formation of an initial alloy layer; severe plastic deformation which increases the local temperature and produces a nano-structure; and fast outward diffusion of Fe and formation of Fe–Al intermetallics. This technique reduced the treatment temperature and duration significantly compared with the conventional Al pack cementation processes, providing a new approach to industrial diffusion coatings with great energy and time savings.