Structure/property/continuum synthesis of ductile fracture in inconel alloy 718

Two microscopic ductile fracture processes have been established in a fracture tough superalloy, Inconel 718, aged to five strength levels. At yield strengths less than 800 MPa, the mechanism is a slow tearing process within large pockets of inhomogeneous carbides and nitrides, giving rise to plane strain fracture toughness (K _IC)values greater than 120 MPa-m^1/2. At yield strengths greater than 900 MPa, the mechanism involves fracture initiation at carbides and nitrides followed by off crack plane void sheet growth nucleated at the Laves (σ) phases. Here, the fracture toughness drops to about 80 MPa-m^1/2. A Mode I normal strain growth model for low yield strength conditions and a shear strain void sheet model for high yield strength ones are shown to model K_IC data obtained from a J-integral evaluation of compact tension results.     

Modeling of local strains in ductile fracture

The concept of dividing microvoid coalescence (MVC) ductile fracture into three constituent processes, nucleation, growth, and coalescence, is discussed, with emphasis on needs for additional analytical and experimental work. Statistical and stochastic aspects of the problem are presented. Recent work on modeling of local strains during ductile fracture, and particularly as components of fracture toughness, is summarized and discussed in light of current knowledge of ductile fracture. Such local strain modeling is especially attractive because it permits micromechanisms of fracture to be explicitly included in the fracture model. This paper is based on a presentation made at the symposium “Stochastic Aspects of Fracture” held at the 1986 annual AIME meeting in New Orleans, LA, on March 2-6, 1986, under the auspices of the ASM/MSD Flow and Fracture Committee.     

Layer thickness effect on ductile tensile fracture

Laminated metal composites containing equal volume percentage of ultrahigh carbon steel (UHCS) and brass were prepared in three different layer thicknesses (750, 200, and 50 μm) by press- bonding and rolling at elevated temperature and were tensile tested at ambient temperature. A dramatic increase in tensile ductility (from 13 to 21 to 60 pct) and a decrease in delamination tendency at the UHCS-brass interfaces were observed as the layer thickness was decreased. The layer thickness effect on ductility is attributed to residual stress whose influence on delamination is decreased as the layer thickness is decreased. Suppression of delamination inhibits neck for- mation in the UHCS layers, allowing for extended uniform plasticity. For a given layer thick- ness, the tensile ductility decreases as the ratio of hardness of component layers is increased.     

Cavity formation from inclusions in ductile fracture

The previously proposed conditions for cavity formation from equiaxed inclusions in ductile fracture have been examined. Critical local elastic energy conditions are found to be necessary but not sufficient for cavity formation. The interfacial strength must also be reached on part of the boundary. For inclusions larger than about 100? the energy condition is always satisfied when the interfacial strength is reached and cavities form by a critical interfacial stress condition. For smaller cavities the stored elastic energy is insufficient to open up interfacial cavities spontaneously. Approximate continuum analyses for extreme idealizations of matrix behavior furnish relatively close limits for the interfacial stress concentration for strain hardening matrices flowing around rigid non-yielding equiaxed inclusions. Such analyses give that in pure shear loading the maximum interfacial stress is very nearly equal to the equivalent flow stress in tension for the given state of plastic strain. Previously proposed models based on a local dissipation of deformation incompatibilities by the punching of dislocation loops lead to rather similar results for interfacial stress concentration when local plastic relaxation is allowed inside the loops. At very small volume fractions of second phase the inclusions do not interact for very substantial amounts of plastic strain. In this regime the interfacial stress is independent of inclusion size. At larger volume fractions of second phase, inclusions begin to interact after moderate amounts of plastic strain, and the interfacial stress concentration becomes dependent on second phase volume fraction. Some of the many reported instances of inclusion size effect in cavity formation can thus be satisfactorily explained by variations of volume fraction of second phase from point to point. This work has been presented in part orally at the Third International Conference on Fracture in Munich, Germany April 1973.     

Cavity formation from inclusions in ductile fracture

The previously proposed conditions for cavity formation from equiaxed inclusions in ductile fracture have been examined. Critical local elastic energy conditions are found to be necessary but not sufficient for cavity formation. The interfacial strength must also be reached on part of the boundary. For inclusions larger than about 100Å the energy condition is always satisfied when the interfacial strength is reached and cavities form by a critical interfacial stress condition. For smaller cavities the stored elastic energy is insufficient to open up interfacial cavities spontaneously. Approximate continuum analyses for extreme idealizations of matrix behavior furnish relatively close limits for the interfacial stress concentration for strain hardening matrices flowing around rigid non-yielding equiaxed inclusions. Such analyses give that in pure shear loading the maximum interfacial stress is very nearly equal to the equivalent flow stress in tension for the given state of plastic strain. Previously proposed models based on a local dissipation of deformation incompatibilities by the punching of dislocation loops lead to rather similar results for interfacial stress concentration when local plastic relaxation is allowed inside the loops. At very small volume fractions of second phase the inclusions do not interact for very substantial amounts of plastic strain. In this regime the interfacial stress is independent of inclusion size. At larger volume fractions of second phase, inclusions begin to interact after moderate amounts of plastic strain, and the interfacial stress concentration becomes dependent on second phase volume fraction. Some of the many reported instances of inclusion size effect in cavity formation can thus be satisfactorily explained by variations of volume fraction of second phase from point to point.     

An acoustic emission investigation of microscopic ductile fracture

A high-strength 4340 steel fracture-toughness specimen was heat treated to give a ductile-rupture type of slow crack growth under rising load. For evaluation of the step-wise growth process, the specimen was instrumented with acoustic stress wave emission (SWE) detection equipment. The resulting crack area swept out by the advancing crack was correlated to the magnitude and number of the acoustic emission pulses. A crack growth model was developed which accounts for the direct relationship between crack area swept out and the sum of the individual SWE amplitudes, and for the experimentally observed bimodal distribution of the SWE amplitudes. The model postulates that slow crack growth takes place in a step-wise mechanism. This involves a repeated two-step process where the first step is the formation of a multitude of individual thumbnail cracks and the second step is the simultaneous interconnection of these thumbnail cracks to form a new continuous crack front. Formerly , Postdoctorate Associate, Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, Minn.     

Hydrogen assisted ductile fracture of spheroidized carbon steels

The effects of hydrogen on ductile fracture were studied in two spheroidized plain carbon steels, containing 0.16 and 0.79 pct C. A combination of fractography and quantitative metallography on sectioned, deformed specimens permitted separation of the effects of hydrogen on the initiation, growth, and link-up of voids. In both steels, hydrogen was found to have no significant effect on either the initiation of voids at carbides, or early growth of voids, prior to link-up. In the higher carbon steel the fracture surface dimple size increased after hydrogen exposure with no other evident change in the fracture surface appearance; it is therefore inferred that hydrogen primarily assists void growth during link-up in this steel. In the lower carbon steel the fracture appearance changed and a decrease in void size due to hydrogen was found fractographically; thus, both initiation and growth of voids are apparently enhanced during the link-up phase of fracture in this steel. It is hypothesized that these effects may be due largely to a void pressure mechanism if hydrogen is transported by mobile dislocations. Formaly Graduate Student, Department of Metallurgy and Materials Science, Carnegie-Mellon University     

An acoustic emission investigation of microscopic ductile fracture

A high-strength 4340 steel fracture-toughness specimen was heat treated to give a ductile-rupture type of slow crack growth under rising load. For evaluation of the step-wise growth process, the specimen was instrumented with acoustic stress wave emission (SWE) detection equipment. The resulting crack area swept out by the advancing crack was correlated to the magnitude and number of the acoustic emission pulses. A crack growth model was developed which accounts for the direct relationship between crack area swept out and the sum of the individual SWE amplitudes, and for the experimentally observed bimodal distribution of the SWE amplitudes. The model postulates that slow crack growth takes place in a step-wise mechanism. This involves a repeated two-step process where the first step is the formation of a multitude of individual thumbnail cracks and the second step is the simultaneous interconnection of these thumbnail cracks to form a new continuous crack front.     

Ferritic-bainitic structure and ductile fracture resistance of high-strength pipe steels

The formation of the products of bainitic transformation in pipe steels of strength classes K60 (X70) and K65 (X80) during thermomechanical treatment is considered. It is shown that the structural factors that affect the ductile-fracture resistance of steel are the bainite morphology and the carbon distribution over structural constituents.     

Damage leading to ductile fracture under high strain-rate conditions

Quantitative metallographic studies of damage evolution leading to ductile fracture under high strain-rate loading conditions are presented. A model material is considered, namely, leaded brass, which contains a dispersed globular lead phase that acts as void nucleation sites. Interrupted tensile split Hopkinson bar tests have been performed to capture the evolution of porosity and void aspect ratio with deformation at strain rates up to 3000 s⁻¹. Both uniaxial and notched specimen geometries were considered to allow the effects of remote stress triaxiality to be investigated. Plate impact testing has also been performed to investigate the evolution of damage under the intense tensile triaxiality and extremely high rates of deformation (10⁵ s⁻¹) occurring within a spall layer. Quantitative metallographic measurements of damage within deformed specimens are used to assess predictions from a Gurson-based constitutive model implemented within an explicit dynamic finite element code. A stress-controlled void nucleation treatment is shown to capture the effect of triaxiality on damage initiation for the range of experiments considered. This article is based on a presentation given in the symposium entitled “Dynamic Behavior of Materials—Part II,” held during the 1998 Fall TMS/ASM Meeting and Materials Week, October 11–15, 1998, in Rosemont, Illinois, under the auspices of the TMS Mechanical Metallurgy and the ASM Flow and Fracture Committees.     

Camage leading to ductile fracture under high strain-rate conditions

Quantitative metallographic studies of damage evolution leading to ductile fracture under high strainrate loading conditions are presented. A model material is considered, namely, leaded brass, which contains a dispersed globular lead phase that acts as void nucleation sites. Interrupted tensile split Hopkinson bar tests have been performed to capture the evolution of porosity and void aspect ratio with deformation at strain rates up to 3000 s⁻¹. Both uniaxial and notched specimen geometries were considered to allow the effects of remote stress triaxiality to be investigated. Plate impact testing has also been performed to investigate the evolution of damage under the intense tensile triaxiality and extremely high rates of deformation (10⁵ s⁻¹) occurring within a spall layer. Quantitative metallographic measurements of damage within deformed specimens are used to assess predictions from a Gursonbased constitutive model implemented within an explicit dynamic finite element code. A stresscontrolled void nucleation treatment is shown to capture the effect of triaxiality on damage initiation for the range of experiments considered. J.P. FOWLER, formerly Research Associate with the Mechanical and Aerospace Engineering Department, Carleton University This article is based on a presentation given in the symposium entiled “Dynamic Behavior of Materials—Part II,” held during the 1998 Fall TMS/ASM Meeting and Materials Week, October 11–15, 1998, in Rosemont, Illinois, under the auspices of the TMS Mechanical Metallurgy and the ASM Flow and Fracture Committees.     

Evaluation of static and dynamic fracture toughness in ductile cast iron

Crack extension behavior and fracture toughness of ductile cast iron were examined by three-point bend tests, where various detection methods of crack initiation under static and dynamic loading conditions were adopted. Loading on the specimens was interrupted at various displacement points, and the final fracture surfaces of the specimen were observed via scanning electron microscopy (SEM). Crack-tip opening displacement (CTOD) obtained under the dynamic loading condition was smaller than that under the static loading condition in ferritic ductile cast iron, and CTOD additionally decreased with increasing pearlite content in the matrix. The relationship between J (ΔC) obtained by the compliance changing rate method and J(R) established by the intersection of the crack extension resistance curve and the theoretical blunting line varied with pearlite content. The average value of .J(ΔC) and J(R), that is J (mid), was proposed to define the fracture toughness of ductile cast iron; J (mid) was considered to be a reasonable measure for the fracture toughness of ductile cast iron, irrespective of loading condition and the pearlite content in the matrix.     

The role of plasticity in bimaterial fracture with ductile interlayers

Evaluation of the plasticity effects in fracture along ductile/brittle interfaces requires appropriate models for plastic dissipation in a ductile component. For thin ductile films, constitutive properties appropriate to the small volumes involved are essential for adequate modeling. Here, yield stress is of primary importance. With nanoindentation, one can obtain both a large strain flow stress as well as the far field yield stress representing the small strain elastic-plastic boundary. Using these to estimate an appropriate plastic strain energy density, the crack tip plastic energy dissipation rates associated with the interfacial crack extension can be estimated for a ductile film. With the preceding analysis, plasticity effects on the interfacial toughness have been evaluated for external measures of strain energy release rates as obtained from indentation tests using the axisymmetric bilayer theory. Comparison involved RF sputtered 200-to 2000-nm-thick Cu interlayers between oxidized silicon and sputtered tungsten. Experimental values for the Cu/SiO₂ interface increased with Cu film thickness from 1 to 15 J/m². This was in qualitative agreement with the theoretical predictions for plastic energy dissipation rates. In contrast, first-order estimates suggest that the observed interfacial toughness increases cannot be attributed to either mode mixity effects or increased intrinsic interfacial fracture energies. As such, crack tip plasticity is identified as the dominant mechanism for increasing interfacial toughness. This article is based on a presentation made in the Symposium “Mechanisms and Mechanics of Composites Fracture” held October 11–15, 1998, at the TMS Fall Meeting in Rosemont, Illinois, under the auspices of the TMS-SMD/ASM-MSCTS Composite Materials Committee.     

Influence of microstructure on fracture toughness of austempered ductile iron

An investigation was carried out to examine the influence of microstructure on the plane strain fracture toughness of austempered ductile iron. Austempered ductile iron (ADI) alloyed with nickel, copper, and molybdenum was austenitized and subsequently austempered over a range of temperatures to produce different microstructures. The microstructures were characterized through optical microscopy and X-ray diffraction. Plane strain fracture toughness of all these materials was determined and was correlated with the microstructure. The results of the present investigation indicate that the lower bainitic microstructure results in higher fracture toughness than upper bainitic microstructure. Both volume fraction of retained austenite and its carbon content influence the fracture toughness. The retained austenite content of 25 vol pct was found to provide the optimum fracture toughness. It was further concluded that the carbon content of the retained austenite should be as high as possible to improve fracture toughness.     

Quantitative description of damage evolution in ductile fracture of tantalum

Dynamic ductile fracture has been studied through incipient spallation experiments on two grades of tantalum. A commercially pure Ta material incipiently spalled at 252 m/s, a highly pure Ta material incipiently spalled at 246 m/s, and a highly pure Ta material preshocked at 250 m/s and incipiently spalled at 246 m/s were used. Microstructural parameters of the fracture process such as porosity, void-size distributions, and void aspect ratios have been quantified using image analysis and optical profilometry techniques. The commercially pure Ta, the highly pure Ta preshocked prior to spall, and the annealed high-purity Ta exhibited 27, 16.6, and 5.5 pct porosity, respectively. The void-size distribution observed in all three tests was adequately represented by either a log-normal or a linear combination of a log-normal and a Weibull distribution function. At least 80 pct of the aspect ratios observed in all three tests were adequately represented by a gamma distribution function. This article is based on a presentation given in the symposium entitled “Dynamic Behavior of Materials—Part II,” held during the 1998 Fall TMS/ASM Meeting and Materials Week, October 11–15, 1998, in Rosemont, Illinois, under the auspices of the TMS Mechanical Metallurgy and the ASM Flow and Fracture Committees.     

The role of plasticity in bimaterial fracture with ductile interlayers

Evaluation of the plasticity effects in fracture along ductile/brittle interfaces requires appropriate models for plastic dissipation in a ductile component. For thin ductile films, constitutive properties appropriate to the small volumes involved are essential for adequate modeling. Here, yield stress is of primary importance. With nanoindentation, one can obtain both a large strain flow stress as well as the far field yield stress representing the small strain elastic-plastic boundary. Using these to estimate an appropriate plastic strain energy density, the crack tip plastic energy dissipation rates associated with the interfacial crack extension can be estimated for a ductile film. With the preceding analysis, plasticity effects on the interfacial toughness have been evaluated for external measures of strain energy release rates as obtained from indentation tests using the axisymmetric bilayer theory. Comparison involved RF sputtered 200- to 2000-nm-thick Cu interlayers between oxidized silicon and sputtered tungsten. Experimental values for the Cu/SiO₂ interface increased with Cu film thickness from 1 to 15 J/m². This was in qualitative agreement with the theoretical predictions for plastic energy dissipation rates. In contrast, first-order estimates suggest that the observed interfacial toughness increases cannot be attributed to either mode mixity effects or increased intrinsic interfacial fracture energies. As such, crack tip plasticity is identified as the dominant mechanism for increasing interfacial toughness. This article is based on a presentation made in the Symposium “Mechanisms and Mechanics of Composites Fracture” held October 11–15, 1998, at the TMS Fall Meeting in Rosemont, Illinois, under the auspices of the TMS-SMD/ASM-MSCTS Composite Materials Committee.     

Effect of porosity on the fracture stress of powder materials in the ductile fracture mechanism

Translated from Poroshkovaya Metallurgiya, No. 5(353), pp. 95–99, May, 1992.     

A Criterion for ductile fracture in sheets under biaxial loading

A criterion for ductile fracture is developed based on the statistical process of shear joining of voids and on the assumption that the voids responsible for fracture have experienced considerable growth prior to this stage of shearing. From the knowledge of uniaxial flow properties and fracture strain measurement, this model is capable of predicting the strain at fracture for other strain states. The predicted data are in good agreement with experiments. Although this model assumes spherical inclusions, some quantitative estimates for elongated inclusions can also be made.     

Quantitative description of damage evolution in ductile fracture of tantalum

Dynamic ductile fracture has been studied through incipient spallation experiments on two grades of tantalum. A commercially pure Ta material incipiently spalled at 252 m/s, a highly pure Ta material incipiently spalled at 246 m/s, and a highly pure Ta material preshocked at 250 m/s and incipiently spalled at 246 m/s were used. Microstructural parameters of the fracture process such as porosity, void-size distributions, and void aspect ratios have been quantified using image analysis and optical profilometry techniques. The commercially pure Ta, the highly pure Ta preshocked prior to spall, and the annealed high-purity Ta exhibited 27, 16.6, and 5.5 pct porosity, respectively. The void-size distribution observed in all three tests was adequately represented by either a log-normal or a linear combination of a log-normal and a Weibull distribution function. At least 80 pct of the aspect ratios observed in all three tests were adequately represented by a gamma distribution function. This article is based on a presentation given in the symposium entitled “Dynamic Behavior of Materials - Part II,” held during the 1998 Fall TMS/ASM Meeting and Materials Week, October 11–15, 1998, in Rosemont, Illinois, under the auspices of the TMS Mechanical Metallurgy and the ASM Flow and Fracture Committees.