TMS Logo

1997 TMS Annual Meeting: Thursday Abstracts

RECENT ADVANCES IN FRACTURE--A Symposium Dedicated to Emeritus Professor Frank A. McClintock: Session VII: Fracture at High Temperature

Sponsored by: MSD Flow and Fracture; SMD Mechanical Metallurgy Committees
Program Organizers: Dr. R.K. Mahidhara, Tessera Inc., 3099 Orchard Drive, San Jose, CA 95134; Dr. A.B. Geltmacher, Naval Research Laboratory, Code 6380, 4555 Overlook Drive SW, Washington DC 20375; Dr. K. Sadananda, Naval Research Laboratory, Code 6323, 4555 Overlook Drive SW, Washington DC 20375; Dr. P. Matic, Naval Research Laboratory, Code 6380, 4555 Overlook Drive SW, Washington DC 20375

Proceedings Info
Poster Session
Previous Session
Return To Program Contents Page

Room: 314A

Session Chairpersons: Professor Amiya K. Mukherjee, Department of Chemical Engineering and Materials Science, University of California, Davis, CA 95616; Professor Kenji Higashi, Department of Mechanical Systems Engineering, Osaka Prefecture University, Sakai, Osaka, Japan

8:25 am INVITED

MODELLING OF CREEP CRACK GROWTH: George R. Webster, Department of Mechanical Engineering, Imperial College, London, SW7 2BX, UK

Failure in high temperature components which suffer creep can occur by net section rupture, crack growth or some combination of both processes depending on the loading conditions. Failure by crack growth is most likely to occur from sites of stress concentration or in components which contain an initial defect. In many practical situations, cracking is preceded by an incubation period, or at least a transient region of very slow growth, prior to the onset of steady state behaviour. This transient region can occupy the majority of life and it is important that it is taken into account to obtain reliable lifetime predictions. In this presentation non linear fracture mechanics concepts will be used to predict the behaviour. A model involving the build up of damage in a process zone at a crack tip will be employed to describe an incubation period, the transient region and steady state crack propagation rates. The role of superimposed fatigue loading will be examined. The analysis will be applied to characterize high temperature crack growth in polycrystalline, directionally solidified and single crystal materials.

8:50 am INVITED

INTERFACIAL DECOHESION FROM SURFACE - ABSORBED EMBRITTLING ELEMENTS: R.C. Muthiah1, Y. Xu2, C.J. McMahon1 and J.L. Bassani2, 1Department of Materials Science and Engineering; 2Department of Mechanical Engineering and Applied Mechanics, University of Pennsylvania, Philadelphia, PA

Motivated by problems in high-temperature cracking in superalloys exposed to an oxidizing environment, McClintock and Bassani (1981) developed a model that considered the diffusion along a grain boundary due to a concentration gradient alone and the resulting decohesion in the presence of the time-dependent crack-tip stresses. More recently, Bika and McMahon (1995) extended their ideas to include the influence of the crack-tip stress gradient on the diffusion process. In this paper we develop a model where diffusion is driven both by concentration and stress gradients while those stresses are directly influenced by cracking process. The formulation utilizes a cohesive zone model that couples creep deformation, diffusion, and damage ahead of a crack to predict the cracking due to dynamic embrittlement. This cracking process is being studied experimentally in systems in which the surface-active embrittling element comes either from the material itself, i.e., from surface segregation, or from the surrounding atmosphere. Both polycrystalline and bicrystal specimens are being employed. We have found that a precipitation-hardened copper-beryllium alloy makes an ideal model material for the study of stress-driven oxygen-induced embrittlement in high-strength alloys. The effect of varying the diffusion coefficient as function of direction in bicrystals is being investigated in a copper-tin alloy, in which the embrittlement comes from surface-segregated tin.

9:15 am INVITED

STRESS INDUCED ELECTRICAL FIELDS AND THEIR INFLUENCE ON HIGH TEMPERATURE FRACTURE IN CERAMICS: Rishi Raj, Department of Materials Science and Engineering, Cornell University, Ithaca, NY 14853

High temperature fracture in ceramics is nearly always controlled by the nucleation and growth of intergranular cavities. The mechanism of growth is stress induced diffusion of point defects. In ceramics the point defects are charged and therefore, their movement can be influenced by local electrical fields. Internal fields can arise near interfaces as a result of net segregation of charged defects to the interfaces. These electrical fields can, in turn, change the defect concentration, thereby, the self diffusivity in the nanoscale region adjacent to the interface. I will discuss, and present results from fundamental experiments that show how these fields can be influenced by applied stress and how these fields participate in the fracture kinetics. Further consideration leads us to speculate on the use of externally applied electrical fields to control nucleation and growth of cavities at grain boundaries.

9:40 am INVITED

CRITICAL ASSESSMENTS OF CAVITATION FAILURE PROCESS IN HIGH STRAIN-RATE SUPERPLASTIC MATERIALS: Kenji Higashi1, M. Mabuchi2 H. Iwasaki3, 1Department of Mechanical Systems Engineering, Osaka Prefecture University, Gakuen-cho, Sakai, Osaka 593, Japan; 2National Industrial Research Institute of Nagoya, Hirate-cho, Kita-ku, Nagoya 462, Japan; 3Department of Materials Science, Himeji Institute of Technology, Shosha, Himeji, Hyogo 671-22, Japan

A new accommodation process for high-strain rate superplastic flow is analyzed from a viewpoint of the relaxation of stress concentrations at triple junctions of grain boundaries for the alloys and around reinforcement particles for the composites resulting from sliding at boundaries and interfaces. A special process by an accommodation helper such as a liquid phase is required to continue superplastic flow when the stress concentration is insufficiently relaxed by diffusional flow and/or diffusion-controlled dislocation movement under the given deformation conditions. A liquid plays an important role as an accommodation helper in the accommodation mechanisms of high strain-rate superplasticity, that is, in an assistance to relax stress concentrations caused by sliding. However, the presence of a liquid phase does not always lead to the high strain-rate superplasticity. The critical conditions such as optimum distribution, thickness and volume in liquid phase are discussed based on the observation results by transmission electron microscopy and cavitation behavior. Cavitation behavior at various conditions for liquid phases are investigated by a quantitative analysis for high strain rate superplastic materials. It is suggested from theoretical analysis that diffusion-controlled cavity growth is limited and the plastically-controlled cavity growth is dominant when stress concentrations at triple junctions of boundaries and around reinforcements are relaxed by the presence of a liquid phase, so that the cavity growth is significantly slow in a small cavity size range. This view was in agreement with the experimental data of the cavity growth rates.

10:05 am INVITED

ROLE OF DIFFUSIONAL RELAXATION IN FRACTURE OF ALUMINUM MATRIX COMPOSITES DURING CREEP AND SUPERPLASTICITY: Rajiv S. Mishra and Amiya K. Mukherjee, Department of Chemical Engineering and Materials Science, University of California, Davis, CA 95616

Reinforcement of metallic matrix with second phase ceramic particles can lead to significant increase in creep strength. Some of these composites also exhibit high strain rate superplasticity. The diffusional relaxation of stresses during creep and superplastic deformation plays an important role. For example, the creep fracture behavior depends on the diffusional relaxation rate. Diffusional relaxation models can be used to calculate the critical creep rate. Below the critical creep rate intergranular fracture is observed, whereas above the critical creep rate transgranular fracture is observed. These observations can be explained on the basis of metal/ceramic interface decohesion. The importance of early onset of interfacial cavitation in the analysis of creep data to obtain mechanistic interpretation is discussed. The change in high strain rate superplasticity mechanism with the size of reinforcement is also explained using the diffusional relaxation models.

10:30 am BREAK

10:40 am

CREEP FRACTURE DURING SOLUTE-DRAG CREEP AND SUPERPLASTIC DEFORMATION: Eric M. Taleff*, Donald R. Lesuer** and Chol K. Syn**, *Department of Aerospace Engineering and Engineering Mechanics, The University of Texas, Austin, TX 78712; **Manufacturing and Materials Engineering Division, Lawrence Livermore National Laboratory, Livermore, CA 94550

Creep fracture has been studied in Al-Mg and Al-Mg-Mn alloys undergoing solute-drag creep and in micro-duplex stainless steel undergoing superplastic deformation. Failure in these materials can be controlled by two mechanisms, neck formation and cavitation. The mechanism of creep fracture during solute-drag creep in Al-Mg is found to change from necking-controlled fracture to cavitation-controlled fracture as Mn content is increased. Binary Al-Mg material fails by neck formation during solute-drag creep, and cavities are formed primarily in the neck region due to high hydrostatic stresses. Ternary alloys of Al-Mg-Mn containing 0.25 and 0.50 wt pct Mn exhibit more uniform cavitation, with the 0.5Mn alloy clearly failing by cavity interlinkage. Failure in the micro-duplex stainless steel is controlled by cavity growth and interlinkage during superplastic deformation. Cavitation was measured at several strains, and cavitation is found to increase as an exponential function of strain. An important aspect of cavity growth in the superplastic stainless steel is the long latency time before cavitation occurs. For a short latency period, cavitation acts to significantly reduce ductility below that by neck growth alone. This effect is most pronounced in materials with high strain-rate sensitivity, for which neck growth occurs very slowly.

11:05 am

FAILURE BEHAVIOR OF DUCTILE LAYERS IN LAMINATED COMPOSITES: S. Bulent Biner, Ames Laboratory, Iowa State University, 208 Metals Development Building, Ames, IA 50011

In this study, the failure of the ductile layers from collinear, multiple and delaminated cracks that occur in laminated composite systems was studied using a constitutive relationship that accounts for strength degradation resulting from the nucleation and growth of voids. The results indicate that in laminated composites, void nucleation and growth ahead of cracks occur at a much faster growth rate due to evolution of much higher stress values at the interface region. Except for short crack extensions, collinear and multiple cracks develop crack resistance curves similar to that seen for a crack in the ductile layer material as in homogenous isotropic cases. For delaminated crack cases, the fracture behavior is strongly influenced by the delaminating length. The resistance of the ductile layers to crack extension can be significantly reduced by short delamination lengths; however, for large delamination lengths the resistance to crack extension becomes greater than that seen for the ductile material.

11:25 am

HIGH TEMPERATURE CRACK GROWTH UNDER MIXED-MODE CONDITIONS: William E. Churley and James C. Earthman, Department of Chemical and Biomedical Engineering, University of California, Irvine, CA 92717

Results from finite element analyses have long been used to model high temperature crack growth processes beginning with the pioneering work of Bassani and McClintock. Recently, finite element results have been used to to study the mechanisms of high temperature crack growth under mixed-mode loading. Experimental studies of this failure process in high temperature alloys and intermetallics have been performed by measuring crack growth direction and crack growth rate in specially designed specimens under Mode I, Mode II and a range of mixed-mode loading conditions. The focus has been on how certain multiaxial stress parameters determined from finite element results can be used to predict both the crack paths and crack growth rates observed experimentally. This approach has led to a better understanding of the dominant crack growth processes in the high temperature materials investigated.

11:45 am

ON VOID GROWTH IN VISCOPLASTIC SOLIDS UNDER CREEP-FATIGUE CONDITIONS: Raj Mohan and F.W. Brust, Engineering Mechanics Department, Battelle Memorial Institute, Columbus, OH 43201 USA

The growth of intergranular voids in elastic-viscoplastic solids is studied using an axisymmetric micromechanical model. Numerical unit cell calculations are performed under remote slow as well as fast cyclic loading. Among the many issues examined include the effect of initial void shape, the effect of on boundary diffusion, the effect of primary creep mechanism, the effect of stress triaxiality as well as the role of elastic accommodation. The results of the study demonstrate that the void growth history and void shape evolution are significantly affected by stress triaxiality, material nonlinearity and initial void shape. The importance of accounting for primary creep mechanism, in addition to secondary creep (power-law creep) is demonstrated for cyclic loading conditions. The analyses shed some light on experimentally observed peculiar behavior under balanced cyclic loading.

12:05 pm

TWO PARAMETER CHARACTERIZATION OF CRACK TIP FIELDS UNDER THERMOMECHANICAL LOADING: Noel P. O'Dowd, Department of Mechanical Engineering, Imperial College, London, SW7 2BX, United Kingdom

Two parameter approaches (K-T, J-Q) have been used to account for constraint and geometry in fracture under mechanical testing. This paper examines the effect of thermal loading on the near tip constraint. Finite element analyses of representative thermal loading which give rise to high and low constraint fields (high and low T and Q) have been conducted. As in previous analyses linear elastic, power law hardening materials have been examined. Following the thermal loading the structure is subjected to mechanical loading, both tension and bending dominated to assess the effect on the near tip constraint. On subsequent mechanical loading it is observed that the two parameter structure of the fields is maintained and the values of Q stress have been obtained from the analyses. The thermal loading initially has a strong effect on the Q value but at higher loads when the mechanical loading dominates, this effect is much weaker. The ability of a combined thermo-mechanical T stress approach to characterize the variation in constraint is assessed.

12:25 pm

HIGH TEMPERATURE FRACTURE OF 6061/Al2O3 MMC'S DEFORMED AT HIGH STRAIN RATES: P. Ganguly and Warren J. Poole, Department of Metals and Materials Engineering, The University of British Columbia,, 309-6350 Stores Road, Vancouver, B.C., V6T 1Z4 Canada

The deformation of the particulate reinforced metal-matric composites (PRMMC) at high temperatures and high strain rates is of critical importance for a number of industrial forming applications, such as hot rolling and hot extrusion. The ductile fracture mechanism in this regime is the classic void nucleation, growth and coalescence, with the void nucleation arising from particle fracture or decohesion of the matrix-particle interface. To predict the onset of damage, it is important to be able to estimate the stress and strain states in and around the reinforcing particles. However, the situation is complicated by relaxation processes in the matrix, which tend to lower the stresses in the particles and at the matrix-particle interface. Furthermore, at large volume fractions (>0.25), the interaction between adjacent particles becomes significant resulting in substantially larger hydrostatic stresses which can aid or deter the formation of the voids depending on whether the hydrostatic stresses are compressive or tensile. The present work aims at studying the failure of PRMMC at high temperature (200°C to 550°C) and moderately high strain rates (0.1 to 10 s-1). Collar compression testing will be used to evaluate the ductility of these materials. The deformation of the macroscopic sample (i.e. w/o the explicit presence of the particles, but with the mechanical properties of the composite) has been modelled using the FEM code ABAQUS. The deformation at the particle length scale has been then determined by imposing the boundary conditions from the macroscopic model on a unit-cell FEM model with single or multiple particles. Metallographic examination of the fracture sites has also been conducted to aid the interpretation of the fracture mechanism.

12:45 pm

ON THE MECHANISMS AND TOUGHNESS OF DUCTILE FRACTURE: Alexander D. Vasilev and S. A. Firstov, Francevich Institute for Problems of Materials Science, 3 Krjijanivskoho str., Kyiv-142, UA-252680, Ukraina

The mechanism of ductile fracture of metallic and some ceramic materials subjected to uniaxial and bending loading in a wide temperature range as well as fracture toughness, and its temperature and structural dependencies are discussed. The main instrument that was used to formulate the ductile fracture mechanism was scanning and transmission electron microscopy of single and polycrystalline, and strengthened by particles, materials. On the basis of study of fracture surfaces of pre-deformed pure particleless materials, the mechanisms of void nucleation, growth and coalescence of voids was found. The pore in particleless materials nucleate along the boundaries of dislocation cellular structure that formed in a course of deformation preceding fracture. The nucleation of pores and subsequent delaminations along the cell and grain boundaries is the main reason of the fracture toughness increase of pre-deformed materials. With the longitudinal cleavage technique it is shown that the intergranular pores nucleate first of all. The decrease of grain size promotes to the transition from cleavage to ductile, by void coalescence, fracture. In materials strengthened with particles there is the temperature region where particles, even with weak interface, do not play the preferable sites of pores nucleation. In those materials, two kinds of ductile, by void coalescence, fracture mechanism may be determined. It is shown that the same mechanism of pores nucleation is valid in ionic (NaCl) single crystals but at temperature above 0.6 of melting temperature. In partially stabilized zirconia single crystals some dislocation plasticity arises at temperature above 1300°C and results in delaminations along interdomain boundaries that points out the same mechanism of pores nucleation. In silicon nitride ceramics the dimple-like, the so-called foam-like, fracture may be found as a result of decomposition of silicon nitride into silicon oxide and gases at temperature above 1000°C. To estimate the fracture toughness of materials failed by ductile manner, the formula: where is the yield stress, is the elastic modulus and is the diameter of the dimples, is proposed. It is shown also that toughness of ductile fracture decreases with temperature and is proportional to the root square of the yield stress. All the conclusions are proven using fractographical data.

Poster Session
Previous Session
Search Technical Program Contents 1997 Annual Meeting Page TMS Meetings Page TMS OnLine