Materials Week '97: Wednesday AM Session Abstracts

Focusing on physical metallurgy and materials, Materials Week '97, which incorporates the TMS Fall Meeting, features a wide array of technical symposia sponsored by The Minerals, Metals & Materials Society (TMS) and ASM International. The meeting will be held September 14-18 in Indianapolis, Indiana. The following sessions will be held Wednesday morning, September 17, during Materials Week 1997. To view other programming planned for the meeting, go to the technical program contents page.

ALUMINUM ALLOYS: General Abstract Session

Session Chairperson: Viola Acoff, University of Alabama, Tuscaloosa, AL

EFFECT OF SOLUTIONIZING ON THE DEFORMATIONAL CHARACTERISTICS OF AGE HARDENABLE ALUMINUM ALLOY COMPOSITES: S.K. Varma, Daniel Salas, Erica Corral, Erika Esquivel and Miriam Regaldo, Department of Metallurgical and Materials Engineering, The University of Texas at El Paso, El Paso, TX 79968-0520

Metal-matrix composites (MMCs) with a combination of age hardenable aluminum alloys as matrix and alumina particles as reinforcements have been solutionized at 540 and 550°C for various times of up to 20 hours. The two solution treated alloys of aluminum, 6061 and 2014, have been subjected to room temperature rolling deformation up to fracture. The microstructural characterization has been carried out using SEM, TEM and optical microscopy. The dislocation arrangements in the matrix and the influence of the particle matrix interface on its distribution will be explored. The evolution of microstructures during the deformation in the composites will be compared with those developed in their respective monoliths. The changes in work hardening features of the composites as a function of solutionizing treatment will be examined in details. This research has been supported by the National Science Foundation through the grant number HRD-9353547.

8:50 am

OPEN AIR DIFFUSION BONDING OF Al ALLOY 6061: Li Hang, Suruzi Bin Abu Samah, G. Dong, H. Li, Division of Materials Engineering, School of Applied Science, Nanyang Technological University, Singapore 539798

Conventional diffusion bonding of Al alloys requires the use of vacuum environment. A feasibility study of open air diffusion bonding of Al alloy 6061 with or without using a special in situ bonding interface treatment was carried out within an applied pressure range of 8.08 MPa to 20.21 MPa at 450°C. Bonding times were set ranging from 30 minutes to 90 minutes. The bonding joints were evaluated by tensile testing, optical microscope and SEM examinations. Tensile test results indicated that the in situ bonding interface treated samples scores much higher UTS values and failure frequently occurred in the parent material. Without the treatment, however, the UTS values were low and failure always occurred at the bonding interface. Microscopic investigation clearly indicated that the in situ interface treatment could effectively break the bonding interface oxide layer and the interface grain growth eventually surpassed the bonding interface. The in situ interface treatment was found highly viable in producing high strength Al alloy diffusion bonding joints in air ambiance.

9:10 am

A NEW FACILITY FOR DIRECTIONAL SOLIDIFICATION-EXPERIMENTS WITH EUTECTIC Al-Al₃Ni-ALLOYS WITH AN ACCELERATED SOLIDIFICATION FRONT: J. Alkemper¹, L. Ratke², ¹Department for Materials Science and Engineering, Northwestern University, Evanston, IL 60201; ²Institute for Space Simulation, DLR, 51140 Cologne, Germany

Based on the power method a new facility for directional solidification was developed. It allows to vary independently of each other the solidification velocity at one fixed position, the acceleration of the solidification front and the temperature gradient ahead of it. The temperature along the sample is measured with optical techniques with high resolution in time and location. Results from experiments with eutectic Al-Al₃Ni alloys are shown to demonstrate that the facility is working and the accuracy of the optical temperature measurements can be evaluated. The movement of the solid-liquid interface is extracted from these measurements.

9:30 am

AGING BEHAVIOR AND THERMAL GROWTH OF CAST 319 ALUMINUM: P.M. Reeber, J.W. Jones, Dept. of Materials Science & Engineering, University of Michigan, Ann Arbor, MI; J.E. Allison, Ford Research Laboratory, Dearborn, MI

The automotive use of cast aluminum has greatly increased during the past decade, especially for elevated temperature applications. One physical property that is important in elevated-temperature applications is thermal stability of the microstructure. Certain copper-containing cast aluminum alloys (like 319) can undergo dimensional changes when exposed to elevated temperatures; when these changes occur, the shape of the component is distorted and the performance may be diminished. Thus, the purpose of this study was two fold: first, to examine the aging response of solution-treated and quenched 319 as a function of aging temperature and time at that temperature; and second, the effect of long-term thermal exposure on 319 Al in the T6 and T7 heat-treated conditions. For both portions of the study, dimensional growth measurements and tensile tests were used to quantify the microstructural changes as functions of time and temperature.

9:50 am BREAK

10:00 am

SEMI-SOLID THERMAL TRANSFORMATIONS (SSTT) OF Al-Si ALLOYS AND THE RESULTING MECHANICAL PROPERTIES: S.C Bergsma¹, M.C. Tolle², M.E. Kassner³, X. Li³, E. Evangelista², ¹Northwest Aluminum Company, The Dalles, OR; ²Dept. of Mechanics, University of Ancona, Ancona, Italy, Presently at Kaiser Aluminum and Chemical Company, Pleasanton, CA 94566; ³Dept. of Mechanical Engineering, Oregon State University, Corvallis, OR 97331

This study demonstrated that 356/357 type Al-Si alloys cast with high solidification rates resulting in a fine grain structure could be rapidly thermally transformed into a structure suitable for semi-solid forming. The mechanical properties of these alloys were comparable to those of electromagnetically stirred semi-solid as well as the as-cast properties.

10:20 am

SINGLE AND MULTISTAGE HOT WORKING OF Al and Al-5Mg ALLOYS: H.J. McQueen, I. Poschman, Mechanical Engineering, Concordia University, Montreal, Canada H3G1M8

At 1.16 s^-1 over the T range 300-500°C, Al flow curves monotonically hardened to steady state strains of 0.9, 2.8 at 500, 400°C. Elongated helicoidal grains were observed on planes normal to the radius by both SEM-EBSI and TEM. As T rises, the equiaxed mean subgrain size in Al (inversely Proportiona1 to stress ) and log normal distribution width increase. For Al-5Mg, rises to a peak and declines at a diminishing rate, although above the power law n=3 domain. At 400°C, peak is about 4 times that of Al, declining to 3 times at =2.8. Dislocations are in long planar arrays at =0.1, O.2 and in elongated subgrains at 0.45 (partially at 0.9). At 0.9, 1.8 and 2.8, the subgrains are equiaxed at 1.4 µm, (Al being ~3.9 µm). The industrial rolling schedule idealized as 17 passes of 20% reduction with preheat at 500°C and completion at 300°C, has been torsion simulated with determination of pass flow curves. From fractional softening during intervals and microstructure, static recovery or recrystallization is greater at higher T. For Al, the reloading stress is higher and the strain hardening rate is lower compared to recrystallized material with convergence at high . In Al-5Mg, a pass flow curve is reduced slightly compared to recrystallized material since appears to be principally controlled by the influence of Mg atoms on the mobile dislocations.

10:40 am

EFFECT OF SILICON PARTICLE SIZE ON FRACTURE BEHAVIOR OF Al-Si ALLOY: Manish Dighe, Arun Gokhale, School of Material Science and Engineering, Georgia Institute of Technology, Atlanta, GA 30332

Commercial Al-Si alloy castings exhibit significant variations in the properties from batch to batch, and from piece to piece. The objective of this investigation is to study microstuctural features that cause these variations, and to identify and quantify the features that control the fracture. For this purpose, microstructural damage evolution is studied in a series of tensile specimens strained to different strain levels. The experiments have been performed on sand cast and chill cast Al-Si alloy. Digital image analysis algorithms, stereology, and fractographic techniques are used to obtain statistically reliable microstructural data. It is observed that debonding of silicon particles is the dominant fracture mechanism. The analysis of the particle size data demonstrate that the sizes of the debonded silicon particles on the fracture surfaces are significantly larger than the average size in the bulk population. In overall population, these large particles constitute negligible percentage (less than 0.1%). Therefore, the fracture behavior of the cast Al-Si alloy depends significantly on the "tail" of the size distribution of the Si particles and not on the average size.

11:00 am

STRENGTH, DUCTILITY AND STRUCTURES IN HOT WORKING OF 7075 Al₂O₃ PARTICLE COMPOSITES: H.J. McQueen, E.V. Konopleva, G. Avramovic-Cingara, Q. Qin, Mechanical Engineering, Concordia University, 1455 de Maisonneuve Blvd. W., H-549-34, Montreal, Quebec H3G 1M8

In deformation by torsion in the ran,2e T=250-540°C at strain rate =0.14 s^-l, the strength of the 15% A1₂O₃/7075 A1, generally higher than that of 10% Al₂O₃/7075 A1, decreased gradually as test temperature rose and declined; but the difference is quite small at higher temperatures. The sinh Arrhenius equation is suitab1e asit was for 6061 matrix with activation energies for 15% A1₂O₃ and 10% A1₂O₃/7075, being 312 and 304 kJ/mol respectively. The ductility-went through a maximum at 400°C. Optical and scanning: electron microscopic examination showed that particles became aligned and contained cracks decreasing from 300 to 500°C. All specimens tested at 300-500°C had short cracks propagating on the surface in the plane of maximum shear stress. At 400°C many cracks started from particles which contained fine cracks. The drop in ductility with few long cracks at 500°C has been related to the formation of large precipitates at grain boundaries. At 500°C, polarized optical microscopy revealed elongated grains with internal substructure and some areas of fine equiaxed crystallites.

11:20 am

A STUDY ON BRAZING OF CLAD 3004Al SHEET: Hong Li, Amit K. Ghosh, Department of Materials Science and Engineering, The University of Michigan, Ann Arbor, MI 48109

A study has been conducted to examine braze-bonding behavior of 3004A1 sheet with surface clad layers of Al-Si alloy. The sheets were fabricated by cold rolling from hot-rolled gauges to different thicknesses which also provided a variety of thicknesses of the Al-Si layer. Bonding experiments on stacked sheets were conducted in vacuum under a variety of temperature, pressure, and time conditions. The tensile strength and microstructure of bond interface were investigated as functions of bond temperature, pressure and time. Optical Microscopy, SEM, EDX and X-ray diffraction were used to analyze the microstructural and chemical compositional change near the interface, and silicon diffusion between core metal (3004Al) and cladding alloy (Al/Si) during braze bonding. In addition to the smooth surface sheets, bonding of sheets with parallel surface grooves were also examined with a view to minimizing crushing of the ligaments separating the grooves. A competition between creep of these regions and diffusion of Si at the interface pose a technically challenging problem requiring optimization of bond parameters.

BOUNDARIES AND INTERFACES IN MATERIALS: THE DAVID A. SMITH SYMPOSIUM: Session V

Session Chair: Christopher R.M. Grovenor, the University of Oxford, Oxford, UK

STRUCTURE AND PROPERTIES OF TRIPLE JUNCTIONS IN ANISOTROPIC SYSTEMS: A.H. King, Department of Materials Science and Engineering, State University of New York, Stony Brook, NY 11794-2275

We have studies the morphology of triple junctions in systems that exhibit anisotropic grain boundary energy. It is shown that the grain boundaries adopt symmetrical configurations adjacent to the triple junctions in a very large fraction of all cases. Up to 70% of the triple junctions in gold thin films, for example, comprise three symmetric-tilt grain boundaries. We consider the energy balance that leads to such configurations and show that there is a very common set of cases in which the Herring equation (relating the morphology to the interfacial energies) appears to fail. It is then demonstrated that the triple junctions are stabilized in a number of different ways in these cases, usually involving the inclusion of additional defects at the junction itself, or along one of the grain boundaries. Research supported by the National Science Foundation, grant number DMR 9530314.

9:00 am INVITED

THE ROLE OF INTERFACIAL STRUCTURE IN DIFFUSIONAL CREEP: J.B. Bilde-Sorensen, Materials Research Department, Riso National Laboratory, PO Box 49, DK-4000 Roskilde, Denmark

The absorption and emission of vacancies at grain boundaries during diffusional creep take place at grain boundary dislocations (GBDs). The GBDs are confined to the boundary plane and in the general case their movement will therefore involve a combination of glide and climb. The coupling of vacancy absorption and emission to the movement of GBDs manifests itself on a larger structural scale in negative grain boundary sliding and negative grain boundary climb at certain grain boundaries as has been observed in materials deformed in diffusional creep. It also explains why denuded zones often are formed on only one side of the boundary in particle-containing materials.

9:30 am INVITED

SIMULATION STUDIES OF INTERFACE DIFFUSION AND PHASE FORMATION: J.M. Rickman, Department of Materials Science and Engineering, Lehigh University, Bethlehem, PA 18015

Two kinetic processes associated with grain boundaries are discussed. In the first study, we examine quantitatively the impact of heterogeneous nucleation on the temporal evolution of a phase transformation with particular emphasis on the correlation of nucleation site distribution and product phase microstructure. This is accomplished by investigating spatial correlations in the transforming system via the calculation of nonequilibrium correlation functions and by characterizing product grain sizes and shapes. Computer simulations of transformations are employed in order to validate our theoretical description and to relate microstructural features of the evolving phase to relevant length and time scales in the problem. In the second study, we investigate the kinetics of grain boundary diffusion using a spatially inhomegeneous lattice gasmodel. It is found that atomic transport can be accurately described by a series of approximate rate equations and that one can ascribe a bias, in a certain sense, to tagged atoms.

10:00 am BREAK

10:10 am INVITED

REVIEW OF INTERFACIAL SEGREGATION STUDIES BY ANALYTICAL ELECTRON MICROSCOPY: A.P. Sutton, T.N. Todorov, Department of Materials, Oxford University, OX1 3PH, UK

Solute segregation to grain boundaries and interphase interfaces can have significant effect on the properties of those interfaces, particularly if they are moving, e.g. during discontinuous precipitation, diffusion-induced grain boundary migration or electromigration. Study of interface migration was an abiding interest of David Smith throughout his career and he made seminal contributions to the literature in this field. The technique of analytical electron microscopy, is the principal tool for the measurement of the composition profiles that develop during interfacial segregation and this paper will review the advantages and limitations of X-ray energy-dispersive spectrometry and electron energy-loss spectrometry studies of interfacial segregation using the David Smith's work as examples where possible.

10:40 am INVITED

GRAIN BOUNDARY DIFFUSION AND THE GROWTH OF PRECIPITATES: G.J. Shiflet, Dept. of Materials Science and Engineering, University of Virginia, Charlottesville, VA 22903

The growth kinetics of grain boundary precipitates are measured as a function of grain misorientation, isothermal temperature and reaction time. The data are analyzed taking into account the short circuit diffusion path for substitutional elements provided by the grain boundary. Growth kinetics are found to increase by about an order of magnitude as the misorientation varies from 20 to 50 degrees for a given heat treatment period. The kinetics are analyzed using various models including Brailsford and Aaron. Solute segregation measurements are obtained using a FEG-TEM.

11:10 am

MEASUREMENT OF Cu DISTRIBUTION IN AN Au-4wt.% Cu THIN FILM BY AEM: D.T. Carpenter, M. Watanabe, D.B. Williams, K. Barmak, D.A. Smith, Department of Materials Science and Engineering, Lehigh University, Bethlehem, PA

The addition of small amounts of copper to aluminum interconnects used in microelectronics may increase their lifetimes by several orders of magnitude. While the exact role of copper is not well understood, there is some qualitative evidence that enhanced lifetimes are related to increased copper segregation at the grain boundaries. The object of the current work is to quantitatively describe the distribution of Cu at grain boundaries in a typical interconnect material. A 300 kV FEG-STEM will be used to measure the distribution of Cu in a thin (100 nm) film of Al-4wt.% Cu using spatially resolved EDS. Analytical results from reduced raster scans and line profiles across a large number of grain boundaries to give information about both the amount and the distribution of Cu at each boundary will be presented.

11:30 am

INTERACTIONS BETWEEN GRAIN BOUNDARY SLIDING AND SOLUTE SEGREGATION: J.S. Vetrano, C.H. Henager, E.P. Simonen, S.M. Bruemmer, Pacific Northwest National Laboratory, Richland, WA 99352

The presence of solute atoms along sliding interfaces can have a profound effect on the climb and glide processes of extrinsic grain boundary dislocations. We have studied these effects by selected solute additions to aluminum and measurements of grain boundary dislocation stability and grain boundary sliding (GBS). Scanning Auger Microprobe measurements of thermally treated Sn- containing alloys showed a strong segregation of Sn to grain boundaries. Measurements of grain boundary sliding in these alloys also indicated that Sn altered the thermal stability of extrinsic grain boundary dislocations and increased the ability of the boundaries to slide. Also studied were solid solution Al-Mg alloys. When these alloys undergo GBS, the point defect and grain boundary dislocation motion combine to redistribute the Mg atoms heterogeneously along the boundary. This also indicates a strong interaction between solute atoms and the mechanisms of GBS. Implications for superplastically forming these alloys will be discussed. Work supported by the Materials Division, Office of Basic Energy Sciences, U.S. Department of Energy under Contract DE-AC06-76RLO 1830.

12:00 pm INVITED

INTERFACIAL DISLOCATIONS, INTERFACIAL STEPS, AND THE PEARLITE REACTION: Paul R. Howell, Department of Materials Science and Engineering, The Pennsylvania State University, University Park, PA 16802

For several decades, pearlite growth was considered to occur through the random attachment of atoms across the advancing interface. More recently, however, a ledge mechanism for pearlite growth has been proposed. Various linear defects in the pearlite.austenite and ferrite/cementite interfaces have been identified as either "growth steps/ledges" and/or "direction steps/ledges". The present paper will review data concerning the nature of the interfacial defects in pearlite with particular regard to the following (still controversial) questions: 1) How does a pearlite colony develop into the abutting austenite? 2) What is the importance (if any!) of crystallography in the pearlite reaction? Results from both ferrous and non-ferrous systems will be reported.

FATIGUE AND CREEP OF COMPOSITE MATERIALS: Session V

Session Chairpersons: Brian N. Cox, Rockwell Science Center, Thousand Oaks, CA 91360; David Davidson, Southwest Research Institute, San Antonio, TX 78228

DESIGNING PARTICULATE COMPOSITE MICROSTRUCTURES FOR FATIGUE RESISTANCE: David Davidson, Southwest Research Institute, San Antonio, TX 78228

Non-deforming particles channel deformation at the tip of a crack into the matrix regions between particles. Particles also constrain flow in the matrix, lowering the deformation that can occur. This effect has been quantified in several particulate composites; representative results will be shown. Fatigue cracks growing at near threshold values of stress intensity factor are not affected much by the presence of particles because of the small plastic zone of the crack. As stress intensity rises, constraint retards crack tip plasticity which prevents the slope of the da/dN vs. K curve from decreasing to the value found in the unreinforced matrix. Fracture toughness is also lowered because of reduced plasticity. To design tougher composites, constraint must be relaxed by using deformable particles.

9:00 am INVITED

MICROMECHANISMS OF FATIGUE CRACK GROWTH IN MoSi₂-BASED COMPOSITES: F. Ye, Y. Li, R.J. Lederich¹ and W.O. Soboyejo, Department of Materials Science and Engineering, The Ohio State University, 2041 College Road, Columbus, OH; ¹McDonnell Douglas, P.O. Box 516, St. Louis, MO 63166-0516

The paper presents the results obtained from recent studies of the effects of reinforcement geometry (particles, fibers and layers) on the fatigue crack growth behavior of MoSix/Nb composites. Particulate reinforcement is shown to promote relatively fast growth rates at low stress intensity factor ranges. Intermediate fatigue crack growth rates are demonstrated in Nb layer-reinforced composites. The beneficial effects of synergistic toughening are also demonstrated in hybrid composites reinforced with transformation toughened MoSi₂ and Nb layers. The trends in the fatigue crack growth behavior are rationalized by considering the combined effects of crack-tip plasticity and the crack-tip shielding mechanisms observed in the experiments.

9:20 am

MICROMECHANICS OF FATIGUE & FRACTURE IN LAMELLAR TiAl COMPOSITES: Bimal Kad and Robert J. Asaro, Dept. of Applied Mechanics & Engineering Sciences, University of California San Diego, La Jolla, CA 92093

Finite element based numerical procedures, incorporating physically based crystal plasticity models, are employed to study the evolution of non-uniform deformation, under monotonic, and cyclic loadings, in lamellar TiAl composite microstructures. The impetus for such efforts is to gather fundamental insight into microstructure sensitive deformation mechanisms, and to extract additional information, not obtainable from traditional mechanical property measurements. Computational efforts address constant strain, and stress, amplitude cycling schemes with variable R=0.5, -0.1, and -1.0 loadings. Results, initially presented for a maximum of 100 cycles, indicate that flat S-N curve response can be understood via the 'signature' hydrostatic stress evolution, which decreases with cycling. Such signatory patterns are significantly affected by intrinsic effects such as plastic anisotropy and microtexture. We will present several examples of experimentally observed, and numerically computed results, to identify hot spots for strain localization in monotonic and fully reversed loadings, and prescribe microstructural remedies to alleviate such effects. Numerical procedures extending these analyses to 10,000 cycles, and those incorporating total strain to life criteria will be described.

9:40 am

FATIGUE CRACK GROWTH BEHAVIOR OF NICKEL ALUMINIDE COMPOSITES: P. Ramasundaram, M. Li, F. Ye, Y. Li, N. Katsube, and W.O. Soboyejo, Department of Materials Science and Engineering, The Ohio State University, 2041 College Road, Columbus OH 43210-1179; Department of Aerospace Engineering, Applied Mechanics and Aviation, 155 W. Woodruff Avenue, Columbus, OH 43210

The effects of composite architecture on the micromechanisms of fatigue crack growth are elucidated for a range of model nickel aluminide based composites. The effects of reinforcement architecture are discussed for Mo-reinforced composites. The effects of layer thickness on the fatigue crack growth behavior are discussed for composites reinforced with ductile phase reinforcements such as Mo, Nb-15Al-40Ti and V. The observed trends in the fatigue crack growth behavior are rationalized by considering the combined effects of crack-tip shielding and crack-tip plasticity.

10:00 am BREAK

10:20 am INVITED

CREEP BEHAVIOR OF MOLYBDENUM DISILICIDE MATRIX COMPOSITES: A.K. Ghosh, Department of Materials Science and Engineering, The University of Michigan, Ann Arbor, MI 48109-2136

High temperature intermetallics such as molybdenum disilicide under steady state creep conditions deform by dislocation creep mechanism exhibiting stress exponent values of 3-4. These intermetallics have weak grain boundaries and are prone to intergranular fracture under tensile creep loading. The addition of SiC particulate reinforcements lead to grain refinement of the matrix while providing load sharing via higher levels of elastic modulus and higher strength levels of the ceramic particulates. Grain refinement brings about an alternative creep mechanism n these intermetallics such as grain boundary sliding accommodated by dislocation glide and climb. The reinforcing particles introduce local stress concentrations at grain boundaries and triple points and can produce an additional component of the matrix creep rate. A creep deformation model has been developed which includes the effect of grain refinement and strain concentration, and particle strengthening effects to predict possible strengthening and weakening effects in the composites as a function of particle volume fraction. It is found that significant volume loading is necessary to produce net strengthening effects in these composites.

10:40 am

CREEP BEHAVIOR OF PARTICULATE NiAl₃ INTERMETALLICS REINFORCED Zn-Al ALLOY: R. Zhang, Z. Wang, Department of Metallurgy & Materials Science, The University of Toronto, Toronto, Ontario

Creep tests were carried out on NiAl₃ intermetallics reinforced commercial Zn-Al Zamak 3 alloy and also the unreinforced reference material at 120°C for a stress range from 20 to 50 MPa. The creep resistance of the composite was found to be significantly higher than that of the unreinforced alloy. For both materials the prediction of the total time to rupture or the time to reach a given creep strain can be made by a single linear relationship between the time to rupture or the time to a specific creep strain and the minimum creep rate. Necking was observed in the unreinforced alloy but not in the composite material, although both materials had similar total creep strain. This observation indicates that the creep deformation in the reinforced material was rather uniformly distributed along the whole gage of the sample. The creep mechanisms and the fracture behavior of the two materials will also be extensively discussed.

11:00 am

COMPRESSION CREEP OF THE NiAl-W INTERMETALLIC FIBER COMPOSITE: T.A. Venkatesh, D.C. Dunand, Massachusetts Institute of Technology, Cambridge, MA 02139

The compressive creep behavior of aligned long fiber reinforced NiAl-W composites, processed by reactive infiltration, is investigated. Three different fiber deformation mechanisms are identified for the case of creeping fibers deforming in a creeping matrix, i.e., longitudinal contraction, lateral deflection and kinking. Simple analytical models are developed to demonstrate the dominant deformation mechanism to be largely determined by two parameters - temperature and initial fiber configuration. Experimental validation of the models is obtained by characterizing the microstructures and creep behavior of NiAl matrix, tungsten reinforcement and NiAl-W composites.

11:20 am INVITED

EFFECTS OF CHANGES IN GRAIN SIZE AND R-RATIO ON FATIGUE OF MONOLITHIC NIOBIUM AND NIOBIUM-BASED "IN SITU" COMPOSITES: William A. Zinsser, Jr., John J. Lewandowski, Dept. of Materials Science & Engineering, Case Western Reserve University, Cleveland OH 44106

Considerable work has focused on the fracture toughness of refractory based systems as well as both intermetallic and ceramic systems toughened with such reinforcements. Less work has focused on the behavior of such systems under cyclic conditions. The present work investigates the effects of changes in grain size and solid solution additions of Zr and Si on the fatigue crack growth behavior of monolithic niobium as well as the effect of cyclic loading on the "in situ" composites based on the Nb-Si system. The effects of changes in the R-ratio and K on the rate of fatigue crack growth and the fracture morphology will be covered and compared to other recent work on such systems.

11:40 am INVITED

ON THE MICROSTRUCTURAL DEPENDENCY OF FRACTURE TOUGHNESS AND CREEP RESISTANCE OF NICKEL/YTRRIA COMPOSITES: Lian-Cho Sun, Leon L. Shaw, Department of Metallurgy and Materials Engineering, University of Connecticut, Storrs, CT 06269

The current prime materials for turbine blade airfoils are nickel-base superalloys which have a density of about 8.2 gm/cm³ and are capable of operating at about 1000C for long-time service. However, further improvement in the operating temperature and density of turbine blades relies on either the employment of ceramics, intermetallics and their composites or metal matrix composites (MMCs). In this study, pure nickel and yttrium oxide were selected as a model system of MMCs to investigate the feasibility of processing three-dimensionally-reinforced nickel-based composites through a powder metallurgy approach and to examine the microstructural dependency of mechanical properties of these model composites. Composites with a volume fraction of yttria ranging from 20 to 70% were prepared through hot pressing. The density, microstructure, fracture toughness and creep resistance of these composites were evaluated as a function of the yttria volume fraction and processing conditions. It was found that the fracture toughness and creep resistance of the composites changed dramatically with the volume fraction of yttria, and this could be related to the morphology change of the yttria, reinforcement from discrete particles to a continuous 3-D network in addition to the effect of the yttria volume fraction. The insight to the dependency of the fracture toughness and creep resistance on the microstructure was explored in light of theories of fracture toughness and creep resistance of composites.

GEORGE R. IRWIN SYMPOSIUM ON CLEAVAGE FRACTURE: Session V: Aluminides and Ceramics

Session Chairpersons: D.A. Koss, Department of Materials Science and Engineering, Pennsylvania State University, University Park, PA 16802; D.M. Dimiduk, Wright Laboratory, Wright-Patterson AFB, Dayton, OH 45433

OPENING REMARKS: Kwai S. Chan, Southwest Research Institute, San Antonio, TX 78238

8:30 am INVITED

FRACTURE AND FRACTURE TOUGHNESS OF DIRECTIONALLY SOLIDIFIED TiAl-BASE TWO-PHASE ALLOYS: N. Akiyama, S. Yokoshima, D.R. Johni, K. Kishida, H. Inui, and M. Yamaguchi, Dept. Materials Science & Engr., Kyoto University, Sakyo-ku, Kyoto 606-01, Japan

The lamellar microstructures in two-phase TiAl alloys are of special interest since the alloys in the lamellar form exhibit good fracture resistance at temperatures up to 800°C. We have introduced a new approach for studying the lamellar microstructures in two-phase TiAl alloys by producing polysynthetically twinned (PST) crystals with the lamellar orientation over the entire crystal by means of directional solidification techniques, and have been working to understand the fracture behavior of the lamellar structure in two-phase TiAl alloys at a fundamental level. We recently extended the fracture studies on PST crystals to ingots composed of columnar grains with the lamellar structure aligned parallel to the growth direction. In this paper, fracture surface morphology, toughness and toughening mechanisms of PST crystals and ingots composed of lamellar columnar grains are discussed as a function of microstructure characteristics, loading rate and environment.

9:00 am INVITED

DEFORMATION AND FRACTURE IN GAMMA TITANIUM ALUMINIDE ALLOYS UNDER MONOTONIC AND CYCLIC LOADING CONDITIONS: Young-Won Kim, UES, Inc., Dayton, OH 45432; Dennis M. Dimiduk, WL/MLLM, Wright Laboratory, WPAFB, OH 45433

Room and elevated temperature deformation and fracture behavior of gamma TiAl alloys was investigated for fine-grained duplex (DP) and large-grained fully-lamellar (FL) microstructures, under tensile and cyclic loading conditions. FL materials resulted in lower ductility and strength, but greater strength retention at high temperatures, and higher brittle-ductile-transition temperatures (BDTT) than DP structures. The Hall-Petch relationship with an unusually high constant exists for FL structures, resulting in strength levels for finer FL microstructures far exceeding those for DP structures. For both loading conditions at temperatures below the BDTT, the ductility was limited by the formation of strain incompatibility induced microcracks which grow to a critical size, comparable to the grain size, leading to cleavage fracture. At high temperatures, intergranular and interlamellar fracture became predominant, although transgranular cleavage fracture was persistent in FL materials. The high temperature deformation under high-cyclic loading was observed to take place by both glide as well as creep. The critical crack size for fatigue failure was related to the grain size, indicating that fine-grained FL materials are favored for enhanced fatigue resistance. In this paper, detailed, comparative analyses will be presented for the deformation and fracture characteristics of gamma TiAl alloys.

9:30 am

CLEAVAGE FRACTURE IN GAMMA-BASED TITANIUM ALUMINIDE INTERMETALLICS: W.O. Soboyejo, C. Mercer, and K. Lou; Department of Materials Science and Engineering, The Ohio State University, 2041 College Road, Columbus, OH 43210-1179

The possible dislocation/microstructure interactions associated with the nucleation of cleavage fracture are elucidated for gamma alloys consisting of lamellar and equiaxed grains. Dislocation based models are thus proposed based on experimental (TEM) evidence of crack-tip deformation (by slip an/or twinning) prior to fracture by cleavage mechanisms. Cleavage fracture in duplex and lamellar gamma alloys is analyzed using classical initiation- and propagation-controlled fracture theories. The analyses and experimental evidence reveal the fundamental importance of plastic flow in the initiation of cleavage fracture. They also explain the Hall-Petch dependence of strength and fracture toughness that is observed in ductile alloys with controlled levels of Cr, Mn, and O.

9:50 am BREAK

10:00 am INVITED

FRACTURE IN B2 COMPOUNDS: P.R. Munroe, Materials Science and Engineering, University of New South Wales, Sydney, NSW205, Australia; and I. Baker, Thayer School of Engineering, Dartmouth College, Hanover, NY 03755

THIS PAPER HAS BEEN WITHDRAWN

10:30 am INVITED

MIXED-MODE CLEAVAGE FRACTURE OF AN IRON ALUMINIDE UNDER MONOTONIC LOADING: F. Robert Frasier, J.P. Hirth, School of Mechanical and Materials Engineering, Washington State University, Pullman, WA 99164-2920

Polycrystalline 3-point bend specimens of an Fe₃Al alloy were K tested in a desiccated Ar atmosphere. Specimens were taken from a warm-rolled plate of nominal 6mm thickness and were fabricated in the L-T orientation. The measured K_lccwas found to be 31 MPa·. Fractographic examination of mode I specimens revealed that fracture had occurred by a complex mixture of mode I transgranular cleavage and mode III intergranular failure. No substantial evidence of ductile rupture was seen on the fracture surfaces. These results are unexpected for this alloy tested in a moisture-free atmosphere. A model for the state of stress responsible for these observations will be introduced, and the implications of these results will be discussed.

11:00 am

CLEAVAGE OF CERAMIC/MINERAL SINGLE CRYSTALS: R.C. Bradt, Dept. Met. & Mat. Eng., Univ. of Alabama, Tuscaloosa, AL 35487-0202

The criteria for cleavage planes in ceramic and mineral single crystals is reviewed from the traditional geometric perspective involving the structural elements of the various crystals with definitive cleavage planes. Crystal planes which exhibit near or quasi-cleavage are similarly addressed. Other historical criteria such as the bond density, the minimum elastic modulus, and a minimum surface energy are similarly considered, thus developing a structural sense as to why planes with a preferential character to cleave should exist within any crystal structure. Next, the rock salt structure is considered with respect to the {100} to {110} cleavage plane transition within that crystal structure. Finally, experimentally determined cleavage surface energies are compared with theoretically calculated surface energies in the alumina or sapphire structure to illustrate the minor role of plastic flow in these cleavages.

11:30 am

TENSILE FAILURE OF CERAMICS AT ELEVATED TEMPERATURES UNDER STATIC AND CYCLIC LOADING: K.J. Hsia, N. Dey, and D.F. Socie, Departments of Materials Science and Engineering, Theoretical and Applied Mechanics, and Mechanical Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801.

Failure of ceramics is usually brittle accompanied by little plastic deformation even at elevated temperatures. However, other micromechanisms such as flow of grain boundary phase may contribute to the overall fracture toughness. The present work studies experimentally the tensile failure of a ceramic containing a grain boundary viscous phase at 1000°C under static and cyclic loading. Under static loading, two failure mechanisms were observed: at high stresses, fracture is dictated by slow growth of a dominant crack along grain boundaries; whereas at low stresses, nucleation, growth and coalescence of multiple microcracks were observed. Fracture under cyclic loading was caused by growth of dominant crack. Interestingly, the lifetime under cyclic loading was one to two orders of magnitude longer (by about a factor of 30) than that under static loading for the same maximum stress. Extensive crack surface bridging by grain boundary viscous phase was observed, which is believed to be the major contribution to the strengthening under cyclic loading.

11:50 am

CLOSING REMARKS: Kwai S. Chan

HIGH CYCLE OF FATIGUE OF STRUCTURAL MATERIALS: Session III: Advanced Materials

Session Chairs: Dr. George Yoder, Office of Naval Research, Washington, D.C.; Dr. Jim Larsen, Wright Laboratories, WPAFB, OH

ON THE GROWTH OF LARGE AND SMALL FATIGUE CRACKS IN DUCTILE AND BRITTLE MATERIALS: R.O. Ritchie, C.J. Gilbert, Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720

The mechanisms of fatigue-crack propagation are examined with particular emphasis on the similarities and differences between cyclic crack growth in ductile materials, such as metals, and corresponding behavior in brittle materials, such as intermetallics and ceramics. This is achieved by considering the process of fatigue-crack growth as a mutual competition between intrinsic mechanisms of crack advance ahead of the crack tip (e.g., alternating crack tip blunting and resharpening), which promote crack growth, and extrinsic mechanisms of crack-tip shielding behind the tip (e.g., crack closure and bridging), which impede it. The widely differing nature of these mechanisms in ductile and brittle materials and their specific dependence upon the alternating and maximum driving forces (e.g., DK and K_max) provide a useful distinction of the process of fatigue-crack propagation in different classes of materials; moreover, it provides a rationalization for the effect of such factors as crack size, load ratio and variable-amplitude loading.

8:25 am INVITED

EFFECT OF ADHESIVE LAYER-THICKNESS ON FATIGUE CRACK GROWTH ALONG POLYMER/METAL INTERFACE: J.K. Shang, Z. Xing, University of Illinois, Urbana, IL 61801

The effect of adhesive layer-thickness on fatigue crack growth along polymer-metal interface was examined by conducting fatigue experiments on the flexural peel specimens and by performing elastoplastic finite element analysis of the interfacial crack tip field. The crack growth rate along the polymer/metal interface was correlated with the total strain energy release rate, G, and obeyed the Paris law at the intermediate growth rates. The fatigue crack growth threshold was found to increase with adhesive thickness for small thicknesses but to decrease gradually with increasing adhesive thickness for large thicknesses. Elastoplastic analysis of the crack tip field and microscopic examination of the failure mechanism indicated that the thickness dependence of interfacial fatigue threshold could be explained by the variations in the crack tip loading condition and in the fracture mechanism with adhesive thickness.

8:50 am INVITED

EFFECT OF FATIGUE LOADING ON THE ADHESION AND PROGRESSIVE DELAMINATION OF POLYMER/METAL INTERFACES: R.H. Dauskardt, Dept. of Materials Science and Engr., Stanford University, Stanford, CA 94305

Bonding of metals using polymers has significantly increased in a wide range of modern applications including aerospace structures, microelectronic packages and bio-prosthetic components. The reliability of these structures are profoundly influenced by the interfacial fracture resistance (adhesion) and resistance to progressive debonding of resulting polymer/metal interfaces. In this study we examine such interfacial fracture properties of representative metal/polymer interfaces commonly found in microelectronic and biomedical applications. Specifically, interface fracture mechanics techniques are described to characterize adhesion and progressive debonding behavior under cyclic fatigue loading. Cyclic fatigue debond-growth rates were measured from ~ 10⁷ to 10^-4 mm/cycle and found to display a power-law dependence on the applied strain energy release rate range, DG. Fracture toughness test results show that the interfaces typically exhibit resistance-curve behavior, with a plateau interface fracture resistance, G_ss, strongly dependent on the interface morphology and the thickness of the polymer layer. Micromechanisms controlling interfacial adhesion and progressive debonding are discussed in terms of the prevailing deformation mechanisms and related to interface structure and morphology.

9:15 am INVITED

PREDICTIVE METHODOLOGIES FOR ELEVATED TEMPERATURE HIGH CYCLE FATIGUE OF STRUCTURAL COMPOSITE MATERIALS: K. Reifsnider, Virginia Polytechnic Institute and State University, Blackburng, VA 24061

Structural composite materials are heterogeneous, generally anisotropic, and often brittle, even at elevated temperatures. However, they have the remarkable characteristic under high cycle fatigue loading of changing their stiffness and strength by significant fractions (one half is not uncommon) before fracture. This creates special challenges for the analyst, as well as for the designer of components made from such materials. Over the last fifteen years or so, several concepts have evolved that are now being widely used to approach problems of this type. The present paper will discuss the combination of those concepts that are embodied in the code series MRLife, used by a number of major industries to predict remaining strength and life of composite components under combined mechanical, thermal, and chemical long-term applied conditions. Some of those concepts include the "critical element" method, damage evolution integrals, and damage accumulation analysis. Comparisons of predicted remaining strength and life under combined fatigue, creep, and stress rupture conditions will be made with measured results for some of the applications to industrial components that have been completed to date.

9:40 am INVITED

HIGH CYCLE FATIGUE AND CRACK GROWTH OF INTERMETALLICS: N.S. Stoloff, Rensselaer Polytechnic Inst., Troy, NY 12180

The great interest in the mechanical behavior of ordered intermetallic compounds over the past three decades has focused upon monotonic strength and fracture properties. Consequently, the literature on high cycle fatigue of these compounds has been scattered and difficult to interpret. Recently, increasing attention has been paid to cyclic properties, especially under stress controlled crack growth conditions. In view of the close association between fatigue lives and growth rates under stress control conditions, this review will address both phenomena. Emphasis will be on recent experimental results for nickel, titanium, niobium and iron aluminides as well as on MoSi₂ alloys. Particular attention will be directed towards microstructural and environmental effects on stress-controlled lives.

10:05 am BREAK

10:20 am INVITED

FATIGUE OF GAMMA TiAl BASED ALUMINIDES: P. Bowen, School of Metallurgy and Materials / IRC in Materials for High Performance Application, The Univ. of Birmingham, Edgbaston, Birmingham B15 2TT, UK

Where specific stiffness improvements outweigh their increased cost Gamma (TiAl) based titanium aluminides are poised for industrial applications in selected components for both automotive and aerospace sectors. Engineering concerns relate to their low ductility, modest fracture toughness and limited fatigue crack growth resistance in the presence of a sharp defect. The topics of total life and fatigue crack growth resistance will be addressed in detail in this presentation. The influence of test temperature, microstructure, lamellar colony size, and lamellar colony orientation on fatigue crack growth resistance and total life will all be highlighted. In addition, effects of environment (air versus vacuum) will be considered for a range of test temperatures and cyclic frequencies. Particular emphasis throughout the presentation will be placed on the fully lamellar and/or near (fully) lamellar microstructures. The importance of both aligned lamellar colonies and randomized lamellar colonies on fatigue crack growth resistance and total life will be addressed. In particular, the problem of premature interlamellar failure under cyclic loading, and its engineering significance, will be explored. The concept of stress-sampling volume correlation's will be considered to rationalize the total life of gamma based aluminides under cyclic loading observed for both plane sided and notched test pieces. Such correlation's will be essential and must be quantified if the performance of large scale components is to be predicted (and approved) on the basis of tests carried out on relatively small scale test pieces. The general challenges of designing against fatigue failure in such brittle materials will also be outlined.

10:45 am

FOREIGN OBJECT DAMAGE AND FATIGUE BEHAVIOR OF GAMMA TiAl: T. Harding, J.W. Jones, Univ. of Michigan, Ann Arbor, MI 48109; T.M. Pollock and P. Steif, Carnegie Mellon University, Pittsburgh, PA 15213

A study is underway to examine the relationship between foreign object damage (FOD) and the fatigue behavior of gamma titanium aluminide. Axial fatigue specimens fabricated from cast Ti-47.9 Al-2Cr-2Nb alloy were impacted under controlled conditions with various indentor shapes to simulate FOD. The damage was quantified and related to impact parameters. A measure of the ambient temperature fatigue strength in the damaged specimens was obtained by standard fatigue testing employing a step-loading approach. Fractographic studies were performed to differentiate impact damage from subsequent fatigue crack growth and to elucidate the mechanisms responsible for the dependence of fatigue strength on FOD. A threshold-based fracture mechanics analysis of crack advance from damage zones, and its use in fatigue life prediction, will be described.

11:05 am

HIGH-CYCLE FATIGUE CRACK GROWTH IN THE PARIS AND THRESHOLD REGIME AT ULTRASONIC FREQUENCIES: Stefanie Stanzl-Tschegg, University of Agriculture, Türkenschanzstraße

Measurement of fatigue limit, fatigue crack growth thresholds and life times under constant or varying amplitudes needs high numbers of cycles and thus long testing times. This means that time and costs can be saved, if the testing frequency is increased up to ultrasonic frequencies. Ultrasound fatigue testing equipments have achieved high technical standard during the last years, so that, besides scientific investigations, reliable results of such measurements for industrial application can be obtained. Improvements of the ultrasound fatigue technique have been obtained mainly in the control accuracy and the development of several new testing possibilities. Testing, for example is possible now at various positive and negative axial mean loads. Similarly, low frequency loads or static mode II or mode III loads may be superimposed to the axial ultrasonic fatigue load. Recently, also torsional loading at ultrasonic frequency has been developed. In addition, not only constant amplitude, but also multi-step or random loading sequences can be performed. The techniques of these developments, the results on material testing with the ultrasound technique as well as comparison with results obtained at conventional frequencies are reviewed in this paper.

11:25 am INVITED

FATIGUE BEHAVIOR OF CONTINUOUS FIBER-REINFORCED CERAMIC-MATRIX COMPOSITES (CFCCS) AT AMBIENT AND ELEVATED TEMPERATURES: P.K. Liaw, M. Miriyala, C.J. McHargue, Dept. of Materials Science and Engineering, The Univ. of Tennessee, Knoxville, TN 37996; L.L. Snead, Oak Ridge National Laboratory, Oak Ridge, TN 37831

Fatigue tests were performed at room temperature in air and at 1000C in an argon environment, on two continuous fiber-reinforced ceramic-matrix composites. Both composites were reinforced with Nicalon fiber-fabrics, with alumina and silicon carbide being the respective matrix materials. Using four-point bend specimens, loads were applied either parallel or normal to the laminate plies to study the effects of fabric orientation on the mechanical behavior. The fatigue behavior of the Nicalon/Alumina composite was significantly affected by the laminate orientation at room and elevated temperatures, while the effects were insignificant in the Nicalon/SiC composite. Creep was observed in both composites at The damage mechanisms responsible for the differences in the fatigue behavior of the two composites, with regard to fabric orientation and test temperature, will be elucidated. Some results of finite element analysis (FEA) also will be presented to explain the effects of fabric orientation on the flexural behavior of laminate composites. Research supported by DOE under a subcontract from Lockheed Martin Energy Research Corporation (No. 11X-SV483V), and by the National Science Foundation, under Contract No. EEC-9527527 with Mrs. Mary Poats as a contract monitor.

11:50 am INVITED

INFLUENCE OF DUCTILE PHASE REINFORCEMENT ON THE CYCLIC FATIGUE BEHAVIOR OF AN OXIDE DISPERSION STRENGTHENED COPPER ALLOY: T.S. Srivatsan, Dept. of Mechanical Engr., The Univ. of Akron, Akron, OH 44325; J.D. Troxell, OMG Americas (formerly SCM Metal Products Inc., Research Triangle Park, Raleigh, NC 27709)

A study has been made to understand the cyclic stress response characteristics, fatigue properties and fracture behavior of an oxide dispersion strengthened copper-niobium composite. The composite specimens were cyclically deformed over a range of stress amplitudes at both ambient and elevated temperatures. Under strain-amplitude control the composite specimens displayed combinations of hardening and softening to failure. The cyclic stress-strain characteristics, fatigue properties and fracture behavior of the composite will be compared with the unreinforced alloy and observed differences rationalized in light of the competing and mutually interactive influences of cyclic strain amplitude and resultant response stress, cyclic stress amplitude, intrinsic microstructural effects, matrix deformation characteristics and macroscopic aspects of fracture.

MECHANICAL BEHAVIOR OF BULK NANO-MATERIALS: Session IV

Session Chair: Robert D. Schull, Metallurgy Division, Bldg. 223, Rm B152, NIST, Gaithersburg, MD 20899

DEPENDENCE OF MECHANICAL PROPERTIES ON SAMPLE IMPERFECTIONS IN NANOCRYSTALLINE METALS: J.R. Weertman, Northwestern University, Evanston, IL 60208

Because of grain refinement strengthening combined with possible extensive grain boundary sliding, nanocrystalline materials should exhibit interesting mechanical behavior. Measurements of mechanical properties have produced widely varying results, even on the same material. It has become evident that the observed behavior is highly dependent on imperfections in the samples. The evolution of mechanical properties in an extended study of nanocrystalline Cu and Pd will be described, and correlated with changes in sample imperfections. Research supported by U.S. Dept. of Energy under Grant DE-FG02-86ER45229.

9:00 am

STATISTICAL STUDY OF NANOSTRUCTURED MATERIAL PROPERTIES: J. Rawers, R. Krabbe, N.Duttlinger, D.Harlow¹, U.S. Dept. of Energy, Albany (Oregon) Research Center; Albany (Oregon) Research Center, 1450 Queen Ave. SW, Albany, OR 97321; ¹Lehigh University Bethlehem, PA

For applications, one needs to know both the reproducibility and reliability of a material. This study was conducted to characterize the statistical variance of nanostructured properties. There are several different steps in the production and consolidation of nanostructured materials and each contributes its own statistical spread to a material property. This study evaluated the reproducibility and ample variation that can be expected in nanostructure properties. Samples evaluated include (a) milled powder from multiple attrition runs in the same and different processing mills, and (b) consolidated compacts from numerous hot-press consolidation runs. Characterization includes nanostructure properties (grain size and strain of milled powders and compacts), and macroscopic properties (density, hardness, tensile/compression strength).

9:25 am

CRITERIA FOR STABLE HALL-PETCH NANOCOMPOSITES: R.L. Holtz, Geo-Centers, Inc., 10903 Indian Head Hwy, Ft. Washington, MD 20744; V. Provenzano, Materials Science & Technology Division, Naval Research Laboratory, Washington, DC 20375

Generic criteria for thermally stable metallic nanocomposites and the associated upper and lower bounds on the strength are examined. These considerations are applied to a two-phase Hall-Petch model with grain growth limited by Zener drag. It is shown that, in the context of this model, an optimal nanocomposite consists of nanoscale particles with a volume fraction of about 25% dispersed in a metal matrix with high Hall-Petch coefficient and grain sizes up to ten times larger than the sizes of the dispersed particles.

9:50 am

EFFECT OF STACKING FAULT ENERGY AND GRAIN SIZE IN FCC METALS ON STRAIN HARDENING REGIMES AND THEIR RELATIONSHIP TO MICROSTRUCTURE EVOLUTION: E.A. El-Danaf, S. Asgari, S.R. Kalidindi, R.D. Doherty, Department of Materials Engineering, Drexel University, Philadelphia, PA 19104

Constant true strain rate compression tests were conducted on annealed polycrystalline samples of different stacking fault energy (SFE) fcc metals Cu, 70/30 brass, 80/20 brass, 90/10 brass MP35N, and stainless steels to large strains. The strain hardening rate (d/d) plotted against both stress and strain, revealed four distinct regimes (or stages) in the low SFE metals: a recovery stage up to a true strain of about -0.07, a regime of approximately constant strain hardening rate up to a true strain of about 0.2, a regime of decreasing strain hardening rate up to a true strain of about 0.6, after which the strain hardening rate was observed to be constant but high compared to stage IV strain hardening rates observed usually in high SFE fcc metals. The concurrent evolution of microstructure in these strain hardening regimes was studied by optical and transmission microscopy.

10:15 am BREAK

10:30 am

GRAIN SIZE AND DUCTILITY OF NANOCRYSTALLINE MATERIALS: T.R. Malow¹, T.P. Smith¹, P.G. Miraglia², K.L. Murty,^2,1 and C.C. Koch¹, ¹North Carolina State University, Department of Materials Science and Engineering, Raleigh, NC 27695-7907; ²North Carolina State University, Department of Nuclear Engineering, Raleigh, NC 27695-7907

The main goal of this investigation is to determine the influence of grain size on mechanical properties and specifically the intrinsic ductility of nanocrystalline materials. Ball milled nanocrystalline Fe was consolidated by warm pressing at 1.5-3 GPa and 400-600°C. Compacts of near theoretical density were produced. Compaction parameters and annealing treatments resulted in a range of grain sizes for subsequent mechanical testing. The use of the miniaturized disk bend test, hardness and the automated ball indentation method have provided results which suggest marked differences in the ductility/mechanical properties of the Fe samples as a function of grain size. Melt spinning and controlled crystallization of CoZr intermetallic results in nanocrystalline ribbon. This processing route avoids compaction artifacts. The mechanical behavior/ductility of CoZr is also discussed.

10:55 am

Si₃N₄/SiC NANOCOMPOSITES: PREPARATION, CHEMISTRY, AND MECHANICAL PROPERTIES: P. Sajgalik,¹ P. Warbicher,² R. Riedel,³ and K. Rajan⁴; ¹Slovak Academy of Sciences, Bratislava, Slovak Republic; ²TU-Graz, Austria; ³TH-Darmstadt, Germany; ⁴RPI, Troy, NY

Nanocomposites consisting of nanoscale particles of silicon nitride in a silicon carbide matrix have been shown to have very promising mechanical properties. In this presentation we describe new processing methods for synthesizing such nanocomposites using a homogeneous nucleation based technique. The interfacial chemistry between the matrix and nano particulates has been studied using EELS imaging techniques. The effect of chemistry is interpreted in terms of its effect on chemical bonding and its subsequent role on properties such as fracture toughness.

11:20 am

GROUP DISCUSSION: Walter Milligan (moderator), Department of Metallurgical and Materials Engineering, Michigan Technological University, Houghton, MI 49931-1295

MICROSTRUCTURE EVOLUTION, CHARACTERIZATION, AND MODELING: Session V: Solid-Solid Systems II

Session Chairperson: M. Rappaz, Ecole Polytechnique Fdrale de Lausanne, DMX-G, CH-1015 Lausanne, Switzerland

MODELING OF PHASE TRANSFORMATIONS IN HYPOEUTECTOID FE-C STEELS DURING HEATING: A. Jacot¹, M. Rappaz¹ and R.C. Reed², ¹Laboratoire de mtallurgie physique, Ecole Polytechnique Fdrale de Lausanne, DMX-G, CH-1015 Lausanne, Switzerland; ²Department of Materials Science & Metallurgy University of Cambridge Pembroke Street, Cambridge, CB2 3QZ

Reaustenitisation from ferrite/pearlite microstructures in Fe-C steels is observed to occur in two steps. The pearlite dissolution which takes place just above the eutectoid temperature is followed by the ferrite to austenite transformation at higher temperature. At moderate heating rate these reactions are mainly governed by the diffusion kinetics of carbon in austenite. Since the diffusion scales and the subsequent transformation kinetics are very different, two separate models have been developed for the pearlite to austenite and the ferrite to austenite transformations. The dissolution of pearlite is described with a two-dimensional finite element approach using a deforming mesh and a remeshing procedure. The diffusion equation is solved in austenite () for a typical domain representative of the periodic structure of ferrite () and cementite () lamellae. The / and / interfaces are allowed to move with respect to the local equilibrium condition including curvatures effects via the Gibbs-Thompson coefficient. The model is able to predict the concentration field and the shape of the interface at different stages of the pearlite dissolution. Maps representing the steady state dissolution rate as a function of the temperature and lamellae spacing will be shown for small values of overheating. The appearance of a non-steady state regime at higher temperature will be discussed. The transformation of ferrite into austenite is described with a two-dimensional explicit finite volume technique. The diffusion equations are solved for a domain representative of the - grain structures, using a hexagonal grid and periodic boundary conditions. The discrete a/g boundary is represented by special interfacial elements which separate -elements from -elements. An -element always undergoes a transition to an /-element before becoming . This procedure allows to handle the displacement of the interface while respecting the flux condition at the interface. Simulated microstructures showing the dissolution of ferrite regions in the austenite matrix are presented at different stages of the phase transformation. Specifically, the influence of the microstructure scale and of the heating rate on the transformation kinetics and homogenization time has been investigated. Reverse TTT-diagrams calculated with this 2-D model are compared with experimental results from the literature.

9:05 am

THE EFFECTS OF INTERFACIAL STRUCTURE ON BOUNDARY MIGRATION IN Ni-Cr AND Fe-C ALLOYS: G. Spanos, R.A. Masumura, Naval Research Laboratory, Washington, DC 20375-5000; G. Chen, W.T. Reynolds, Jr., Virginia Polytechnic Institute, Blacksburg, VA 24061

The effects of growth ledges on interface migration and sympathetic nucleation of precipitates at interphase boundaries is examined with finite difference-based numerical calculations using the ledge growth model of Enomoto. Two alloy systems are considered: (1) Ni-45wt%Cr, and (2) Fe-C alloys. For the Ni-45wt%Cr alloy, direct measurements of ledge spacings and heights as a function of isothermal growth time and temperature are used as input into the model, and excellent agreement is found between experimentally measured and calculated lath thickening rates. These results are compared to similar calculations published previously for plate lengthening in the Fe-C system. Finally, the model is used to examine ledge nucleation and the driving force for sympathetic nucleation in Fe-C alloys. These calculations indicate that the return of super saturating at local regions of ledged ferrite: austenite interfaces should allow for plenty of driving force for sympathetic nucleation of ferrite crystals atop preexisting ferrite crystals, and the preferred sites for such nucleation are discussed. Results will be analyzed with the aid of color maps of the calculated solute profiles associated with ledged growth interfaces in both Ni-45wt%Cr and Fe-C alloys.

9:40 am

THREE DIMENSIONAL RECONSTRUCTION AND CLASSIFICATION OF CEMENTITE PRECIPITATES: M.V. Kral, G. Spanos, Naval Research Laboratory, Code 6324, Physical Metallurgy Branch, Washington, DC 20375-5343

The three dimensional nature of cementite precipitates in an Fe-1.34%C-13.0% Mn alloy has been revealed by two experimental techniques. A serial sectioning/computer reconstruction technique enables the visualization of entire grains of material from any perspective, and resolution is limited only by the thickness of material removed for each section and the number of sections that are taken. Scanning Electron Microscopy (SEM) of deep-etched specimens also shows excellent resolution of precipitates, but only near a specific plane of polish. The combined use of these methods on isothermally transformed specimens provides detailed information about the true three dimensional morphology, connectivity and growth of precipitates. Three-dimensional projections of various morphological types of cementite precipitates found in these materials are presented along with data on volume fractions and size measurements of all three dimensions. The resulting observations have enabled a more precise classification of precipitates than was previously possible with two dimensional microscopy techniques.

10:15 am BREAK

10:30 am

THE DEPENDENCE OF THE DEVELOPMENT OF RECRYSTALLIZATION TEXTURE ON THE STORED DEFORMATION ENERGY: Baolute Ren, Manufacturing Technology Laboratory, Corporate Research & Development, Reynolds Metals Company, 3326 East Second Street Muscle Shoals, AL 35661-1258

Among the many metallurgical and processing parameters which influence the development of recrystallization texture, it is generally believed that stored energy is the most important parameter. It determines the driving force for both nucleation and growth of crystal grains during primary recrystallization. In the present work, the development of stored energy is simulated on the basis of the Taylor/Bishop-Hill model. The simulation results indicate that the stored energy is not homogeneous. Therefore, the driving force for the development of recrystallization texture is not homogeneous. The effect of the stored energy in the forms of the grain boundary area, sub-boundary, and deformation zone on the recrystallization texture development is discussed.

11:15 am

DEVELOPMENT OF CUBE RECRYSTALLIZATION TEXTURE IN WARM PLANE DEFORMED ALUMINUM: EXPERIMENTS AND MODELING: Roger D. Doherty, Indradev Samajdarand Le Chun Chen, Department of Materials Engineering, Drexel University, Philadelphia, PA 19106

Heavily plane strain deformed fcc metals aluminum and copper show, in certain cases, a remarkable texture change from a stand rolling texture, containing only a small amount of cube texture, to a very strong cube texture with 75% or more of the volume occupied by recrystallized grains having a near cube texture. Recent experiments have shown that the development of strong cube texture arises from two features of the microstructural evolution during deformation: (i) The quasi stability of cube oriented material and (ii) The significantly lower stored energies of the deformed cube bands that are elongated in the rolling direction. A simple model for the development of recrystallized cube texture has been developed and successfully tested. Current studies are investigating the role of starting grain size, starting cube texture and deformation conditions on the input parameters of the model, which are the deformed cube band spacing, its thickness, and the frequency of cube grain nucleation from these bands in competition with nucleation of grains of other orientations.

11:40 am

NONCONVENTIONAL X-RAY DIFFRACTION TECHNIQUES FOR COATINGS: O.B. Girin, Yu. O. Proshenko, V.I. Bekerav, Dept. Physics, State Metallurgical Academy of Ukraine, Prospekt Gagarina 4, Dnepropetrovsk 320635, Ukraine

An improved package of nonconventional Xray diffraction techniques for coating characterization is described. The package permits (I) characterization of texture with due regard for anisotropy of crystal defects, (ii) characterization of substructure in oriented coatings, and investigation of substructure anisotropy in the various components of texture, and (iii) in situ studies of coating structure formation during deposition. These techniques combined with TEM, FIM and DTA revealed some effects throwing new light on formation of crystalline substructures and textures in electrodeposits. A mechanism and a model of texture evolution in electrodeposits is discussed. A model of electrodeposit structure formation involving liquid metal is addressed.

MICROSTRUCTURE PROPERTY RELATIONSHIPS IN / TITANIUM ALLOYS: Session II: Fatigue and Creep

Session Chairperson: Blair London, Materials Engineering Department, California Polytechnic State University, San Luis Obispo, CA 93407

THE INFLUENCE OF STRAIN RATE ON THE STRUCTURE/PROPERTY BEHAVIOR OF Ti AND Ti-ALLOYS: George T. (Rusty) Gray III, Los Alamos National Laboratory, Materials Research and Processing Science, Los Alamos, NM 87545

The high-strain-rate stress-strain response of titanium alloys is receiving continued attention related to design considerations for crash-worthiness and foreign-object damage in aerospace systems, ballistic and armor applications, high-rate forming, and high-rate machining. Interest in building more physically-based constitutive models to describe these processes utilizing Ti-alloys requires a knowledge of the coincident influence of temperature, strain rate, texture, and microstructure on the high-strain-rate mechanical response of Ti and Ti-alloys. While numerous experimental studies have probed the low-rate mechanical behavior of Ti and Ti-alloys, considerably less is known about the systematic effects of alloy content, interstitial content, microstructure, temperature, and texture on the high-strain-rate constitutive and fracture response of Ti-based alloys. In this paper the influence of strain rate, with emphasis on the high-rate regime, on the structure/property response of Ti and Ti-alloys will be reviewed. Substructure evolution in Ti and Ti-alloys as a function of strain rate will be discussed including the role of deformation twinning on plastic flow and fracture under high-rate, impact, and shock-loading conditions.

9:10 am

HIGH CYCLE FATIGUE PROPERTIES OF Ti-6Al-4V ALLOYS WITH EQUIAXED MICROSTRUCTURE: Svetlana G. Ivanova, Frederick S. Cohen, Materials and Mechanics Engineering, Pratt & Whitney, 400 Main Street, East Hartford, CT 01608; Ronald R. Biederman, Richard D. Sisson, Jr., Materials Science and Engineering, Worcester Polytechnic Institute, 100 Institute Road, Worcester, MA 01609-2280

The high cycle fatigue properties were experimentally investigated for both forged and cross-rolled Ti-6Al-4V with equiaxed microstructure. The cross-rolled material exhibited significant crystallographic texture while the forged material showed little texture. Constant fatigue life diagrams for both cross-rolled and forged materials lie below Goodman linear relationship. The HCF strength is similar for both cross-rolled and forged materials at stress ratios between R=0.1 and R=0.8. However, in reverse loading (R=-1) cross-rolled material has approximately 10% higher HCF. The effect of texture on HCF strength will be presented and discussed. SEM investigation of fracture surfaces revealed that failure always initiated at the surface for both forged and rolled materials at all loading conditions, including at tensile mean stresses (up to R=0.8 stress ratios).

9:30 am

MICROSTRUCTURE AND FATIGUE PROPERTIES OF b-CEZ: J.O. Peters, G. Lütjering, Technical University Hamburg-Harburg, 21071 Hamburg, FRG

The influence of the microstructure on the fatigue properties of the high-strength -CEZ alloy was investigated by comparing three distinctly different microstructures: a lamellar (-annealed), a bi-modal (+-processed), and a necklace microstructure (processed through the -transus). The fatigue tests were performed in vacuum and in air. Comparing the fatigue properties of the three microstructures at the same yield stress level of 1200 MPa, the bi-modal microstructure showed the best HCF and LCF properties. Concerning fatigue crack propagation, small surface cracks (microcracks) exhibited the slowest propagation rate in the bi-modal microstructures, with the tendency that the lamellar microstructure had a slightly higher resistance against macrocrack propagation. The results of fatigue properties are explained by differences in the b-grain sizes and by differences in the crack front profiles of the three microstructures.

9:50 am

FATIGUE CRACK GROWTH BEHAVIOR OF TITANIUM ALLOYS UNDER GASEOUS ENVIRONMENT AT HIGH TEMPERATURE: C. Sarrazin-Baudoux, Y. Chabanne, J. Petit, Laboratoire de Mcanique et de Physique des Matriaux, URA CNRS 863-ENSMA Site du Futuroscope, Chasseneuil du Poitou, B.P. 109, 86960 FUTUROSCOPE CEDEX

The deleterious influence of ambient air on the fatigue crack growth resistance of most of the metallic alloys is now well established at room temperature and has been clearly related to the presence of moisture in the surrounding environment. At higher temperatures the respective role of water vapor and oxygen is more disputed, and these species act differently according to the alloy. This paper aims to put in light general trends about microstructure (and composition) on the fatigue behavior of a selection of titanium alloys (Ti-6246, Ti-6242, IMI 834) tested at temperature levels extending from room temperature to 500°C in ambient air and high vacuum. From data performed under environments with controlled amounts of water vapor and oxygen, including high vacuum and with closure correction, critical conditions (partial pressure, frequency, load ratio, mean load, maximum load...) for the occurrence of water vapor assisted corrosion-fatigue, creep-fatigue and stress corrosion are explored and supported by microfractographic observations.

10:10 am BREAK

10:30 am

THE INFLUENCE OF THE MICROSTRUCTURE ON THE DWELL-TIME FATIGUE PROPERTIES OF Ti-6242: R. Faber, M.E. Kassner, Department of Mechanical Engineering, Oregon State University, Corvallis, OR 97331; Y. Kosaka, J.E. Kosin, B. Bristow, S.H. Reichman, OREMET Titanium, Albany, OR 97321

An increasingly important property of Ti-6242 is favorable dwell-time fatigue life. This study investigated the influence of the microstructural features associated with changes in the annealing temperature below the beta transus on the ambient temperature dwell-time fatigue life. Determination of the cycles (of complete unloading) to failure were performed for a variety of sustained stress levels for each set of specimens annealed at various temperatures below T. These results will be discussed along with comparisons to the conventional fatigue properties of the alloy.

10:50 am

CREEP BEHAVIOR OF Ti-6Al-2Sn-4Zr-2Mo: R.W. Hayes, Metals Technology Inc., 19801 Nordhoff Street, Northridge, CA 91324; E. Landers , B. London, Materials Engineering Department, California Polytechnic State University, San Luis Obispo, CA 93407

The creep behavior of the Ti alloy Ti-6Al-2Sn-4Zr-2Mo has been studied over the temperature range 510 to 538°C at initial applied stress levels ranging from 345 to 414 MPa. Two microstructures, a lath type and a fine equiaxed type, were studied. Both stress increase and stress reduction creep experiments were performed. From the stress increase experiments, the activation energies and the stress exponents for the minimum strain rates were obtained. The values of the activation energies and the stress exponents for both microstructures are consistent with a dislocation motion controlled recovery process which describes structure evolution under the present creep conditions. In addition to the minimum strain rates, the primary transient creep of the equiaxed and lath microstructures was also studied. An activation energy based upon the primary transient time suggests that recovery processes may be important in primary creep of this alloy. A brief discussion of the transient creep response following a stress reduction will also be presented.

11:10 am

ORIENTATION EFFECTS AND THE ROLE OF / INTERFACES IN ROOM-TEMPERATURE CREEP BEHAVIOR OF TITANIUM ALLOYS: S. Suri, T. Neeraj, V. Babu, G.S. Daehn, M.J. Mills, Department of Materials Science and Engineering, The Ohio State University, Columbus, OH 43210; D.-H. Hou, TEM Laboratories, Micron Technology, Inc., Boise, ID 83706

Room temperature creep at low stresses in two phase /Ti alloys has been widely reported in the literature. However, little fundamental understanding of the mechanisms associated with this creep behavior in the / alloys exists. In this study, single colony crystals of a near -Ti alloy have been grown using a float zone technique. The colony crystals have been oriented so the different prismatic slip systems have the highest resolved shear stress. Significant anisotropy in creep behavior has been observed. An attempt has been made to explain the anisotropy by investigating the micromechanisms of slip transfer across the / interface using scanning and transmission electron microscopy. The role of interfacial dislocations in the creep behavior has also been studied. These results will be presented and discussed. This work is supported by the Air Force Office of Scientific Research, Dr. Charles H. Ward, Project Monitor.

11:30 am

PHENOMENOLOGICAL DESCRIPTION OF CREEP BEHAVIOR IN AND / TITANIUM ALLOYS USING CONSTANT STRAIN RATE TESTS: T. Neeraj, S. Suri, B. Viswanathan, G.S. Daehn, M.J. Mills, Department of Materials Science and Engineering, The Ohio State University, Columbus, OH 43210; D.-H. Hou, TEM Laboratories, Micron Technology, Inc., Boise, ID 83706

Titanium alloys exhibit creep at low homologous temperatures and stress levels below the yield strength. In this work an attempt has been made to correlate constant strain rate tests and creep response of these alloys. At low homologous temperatures, creep curves in Ti alloys follow a power law: =At^a. An analytical solution describing the creep strains from constant strain rate tests data has been developed. Constant strain rate and creep tests have been performed on both single phase and two phase / alloys. The analytical solution has been used to predict the creep curves for several model binary Ti-Al alloy systems and Ti-6242 will be presented and discussed.

NON-LINEAR EFFECTS IN MATERIALS SCIENCE: Session III: Mechanical Behavior

Session Chairperson: A.J. Markworth, Dept. of Materials Science, The Ohio State University, Columbus, OH 43210-1179

A NEW MECHANISM OF DISLOCATION GENERATION AT FINITE TEMPERATURES: RELATION TO THE BRITTLE-TO-DUCTILE TRANSITION: V. Vitek, M. Khantha, D.P. Pope, Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, PA 19104

Unlike dislocation generation by Frank-Read sources, the mechanism described in this contribution is a thermally-driven, stress-assisted cooperative instability of many dislocation loops. The dislocation loops are formed by thermal fluctuations and the small plastic strain associated with them invokes an effective decrease of the moduli of the medium. The self-energy of such loops is proportional to these effective moduli and, as the temperature increases, the density of the loops increases and this in turn leads to the decrease of the effective moduli. This feedback ultimately triggers off a collective unstable expansion of many loops above a critical temperature, Tc. Without mechanical loading this mechanism corresponds to the Kosterlitz-Thouless model of defect-mediated melting transition but under large applied loads the instability occurs well below the melting temperature. Such loads can be attained near the crack tips and/or in dislocation free materials, such as whiskers. In both cases Tc represents the temperature at which the material becomes suddenly ductile and corresponds thus to the brittle-to-ductile transition temperature. This research was supported by the US Air Force Office of Scientific Research grant no. 95-1-0143.

9:00 am INVITED

BIFURCATIONS AND SINGULARITIES IN ICE BREAKING: Dale G. Karr, Department of Naval Architecture and Marine Engineering, The University of Michigan, 2600 Draper Road, Ann Arbor, MI 48109-2145

Several nonlinear effects prevalent in descriptions of the mechanical behavior of ice are discussed in this paper. Polycrystalline ice is a rate dependent material often subject to the nucleation and growth of microcracks which is a source of nonlinearity in the stress-strain relations. The types of bifurcations include localization, which is associated with a bifurcation of the homogeneous deformation, and general bifurcation, associated with lose of uniqueness of the stress-strain relations. Another nonlinear effect arises when structures dynamically interact with ice such as during the operation of ice breaking vessels or the impingement of ice fields against offshore platforms. The forces exerted on the structures are nonlinear and intermittent due to the breakage and clearing of the ice medium. These nonlinear effects lead to very complex nonlinear dynamic behavior exhibiting bifurcation of periodic response, multiple periodicity, and chaotic oscillation of structures. Singular conditions such as grazing and breaking thresholds play dominant roles in the development of response bifurcations and in bounding the nonlinear dynamic motions of the structure.

9:30 am INVITED

STATISTICAL MODEL FOR MECHANICAL FAILURE: Miron Kaufman, Physics Department, Cleveland State University, Cleveland, OH 44115; John Ferrante, NASA Lewis Research Center, Cleveland, OH 44135

In this paper we analyze an equilibrium statistical mechanics model of a solid. This model solid is made of "springs". We go beyond the Hooke law for harmonic "springs" by using the nonlinear energy versus atomic distance developed by Ferrante and his collaborators. If the energy of such a spring is larger than a threshold energy the "spring" is assumed to fail. Assuming that the relaxation times are short compared to the measurement time, we use equilibrium statistical mechanics to compute the various thermodynamic quantities. We find two transitions: (i) the softening transition corresponding to a thermodynamic instability when the isothermal derivative of the stress with respect to strain is zero; (ii) when the network of failed springs percolates the solid becomes brittle. In the temperature-stress phase diagram, the two transition lines intersect at a novel multicritical point. The model is extended to account for correlations between failed springs by using a mapping to the Potts model.

10:00 am INVITED

MECHANICAL BEHAVIOR OF SILICA ELEMENTS IN PHOTONIC STRUCTURES WITH CONSIDERATION OF THE NON-LINEAR ELASTICITY OF THE MATERIAL: E. Suhir, Bell Laboratories, Lucent Technologies, 600 Mountain Ave., Room 1D-443, Murray Hill, NJ 07974

Mechanical behavior of silica elements in various photonic (fiber-optic) structures is predicted, taking into account the non-linear elastic stress-strain relationship for the silica material. The following problems are considered: low temperature microbending of dual-coated (infinitely long) optical fibers; elastic stability of short bare and metallized fibers subjected to thermally induced compression; thermally induced stresses and strains in fused biconical taper (FBT) lightwave coupler experiencing thermal contraction mismatch with its base; free vibrations of glass fibers subjected to large deformations during "two-point bending". The analyses are carried out under an assumption that the non-linear stress-strain relationship obtained experimentally for the case of uniaxial tension (Krause, Testardi, and Thurston, (1979) holds also for other deformations. It is concluded that the deviation of the stress-strain relationship from Hooke's law can have a significant effect on the mechanical behavior of silica elements in photonic structures and should be accounted for.

10:30 am BREAK

10:40 am INVITED

CONSTITUTIVE BIFURCATION AND SHEAR BAND INITIATION OF RATE INDEPENDENT BRITTLE DAMAGE MATERIALS: Xin Sun, Research Scientist, Battelle Memorial Institute, 505 King Ave., Columbus, OH 43201

Continuum damage mechanics (CDM) has been widely used to model the mechanical behavior of brittle damage materials. In this paper, the stability issues of the CDM models are studied. The stress-strain relations derived from CDM theory are shown to possess bifurcation points associated with the loss of uniqueness of the constitutive configurations. It is shown that this nonlinear degradation of the material, despite restriction to infinitesimal strain, leads to the presence of multiple constitutive paths. The stability of the constitutive paths is analyzed by considering augmented dynamical systems. The effects of material degradation on failure modes of a brittle material are also investigated. Bifurcations from the homogeneous deformation mode in the form of shear bands are captured for loading conditions of plane strain compression and uniaxial compression. In contrast with previous research on shear band initiation, that all rely on plasticity and flow theory formulation, the present study finds it sufficient for the shear band to emerge in the regime of infinitesimal strain for brittle damage materials.

11:10 am

CHAOTIC EFFECTS IN ELECTRON DRAG PROCESSES: J.M. Galligan, Dept. of Metallurgy and Materials Engineering, University of Connecticut, Storrs, CT 06269-3136; L.N. Gumen, Departamento de Matematicas, Universidad Popular Autonoma del Estado de Puebla, Puebla, 72160 Mexico; I.V. Krivoshey, Kharkov State University, Kharkov, Ukraine; A.A. Krokhin and G.A. Luna-Acosta, Instituto de Fisica de la UAP, Puebla, 72570 Mexico

The non-linear trajectory type effect is applied to electron-dislocation interactions. It is shown, using chaotic criteria based on topological properties of the potential energy surface, that, in classically strong magnetic fields, chaotic states are formed near an edge dislocation. For this case the electron motion is treated in the diffusion approximation, resulting in a non-linear dependence of the electron drag upon the magnetic field. This theory is in good agreement with experimental data, obtained for a zinc crystal sample. Numerical simulations confirm the chaotic character of electron motion near dislocations.

11:30 am

YIELDING, MOBILE DISLOCATIONS AND CHAOS: J.M Galligan, T.J. McKrell, Dept. of Metallurgy and Materials Engineering and the Institute of Materials Science, University of Connecticut, Storrs, CT 06269-3136

A method of measuring the instantaneous mobile dislocation density has been used to examine what happens when lead alloy crystals deform. This measurement, which relates magnetic flux flow to the instantaneous mobile dislocation, shows that yielding occurs in bursts. During these bursts there is a large amount of correlation among the mobile dislocations. These bursts are followed by dislocation activity in which there is little or any correlation among the dislocations, interspersed with deformation regions with large, positive correlations. Such measurements show that mobile dislocation behavior has a random component superimposed on correlated motion of dislocations.

11:50 am

ELEVATED TEMPERATURE MECHANICAL PROPERTIES OF AN ACTIVE METAL VERSION OF THE 92Au-8Pd BRAZE ALLOY: J.J. Stephens, Materials Joining Department, Sandia National Laboratories, Albuquerque, NM 87185-0367

Calculations of residual stresses in brazed metal/ceramic assemblies are often limited by a lack of mechanical properties data for braze alloy(s) of interest. In recent years, a significant data base has been developed on the elevated temperature properties of both conventional and active metal alloys used in specific projects at Sandia. This talk will focus on a recently developed active metal version of the conventional 92Au-8Pd alloy, which was developed as a high temperature alloy for brazing to silicon nitride ceramics. The active element addition (2 wt.% V) serves to significantly increase the creep strength relative to the conventional braze alloy. For both alloys, a high temperature power law creep equation applies, followed by a transition to a Garofalo sinh equation at mid and lower temperatures. Microstructures and fracture behavior of both alloys will also be discussed. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy Contract number DE-AC0494AL85000.

P/M CURRENT RESEARCH AND INDUSTRIAL PRACTICES: Session I: Current Research I

Session Chairperson: Enrique V. Barrera, Rice University, Department of Mechanical Engineering and Materials Science, Houston, TX 77005-1892

OPENING REMARKS AND INTRODUCTIONS: Enrique V. Barrera, former Chair of the Powder Materials Committee

8:40 am INVITED Talk by Current Chair of the PM Committee

SHOCK-INDUCED AND SHOCK-ASSISTED SYNTHESIS OF MATERIALS: Naresh N. Thadhani, Kevin Vandersall, Shantanu Namjoshi, School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0245

Shock-compression of powders can lead to chemical reactions and phase transformations resulting in the formation of compounds with modified microstructures, as well as equilibrium and non-equilibrium phases. Two types of processes are possible and can be distinguished on the basis of their respective mechanics and kinetics. Shock-induced processes occurring during the shock-compression state before unloading to ambient pressure, can be utilized for synthesis of high-pressure phases and compounds not possible during ambient conditions. Shock-assisted processes are those that occur after unloading to ambient pressure, in an essentially shock-modified material, with subsequent thermal treatment. In this presentation, our results on "shock-induced" synthesis of carbon nitrides and "shock-assisted" processing of silicides with refined microstructures will be presented.

9:10 am INVITED

DIRECT SELECTIVE LASER SINTERING OF HIGH PERFORMANCE METALS FOR CONTAINERLESS HIP: S. Das, M. Wohlert, J.J. Beaman, D.L. Bourell, Laboratory for Freeform Fabrication, The University of Texas at Austin, Austin, TX 77281

A novel net shape manufacturing method known as SLS/HIP which combines the strengths of selective laser sintering and hot isostatic pressing is presented. Direct selective laser sintering is a rapid manufacturing technique that can produce high density metal parts of complex geometry with an integral, gas impermeable skin. These partscan then be directly post-processed by containerless HIP. The advantages of in-situ HIP encapsulation include elimination of a secondary container material and associated container-powder interaction, reduced pre-processing time, a short HIP cycle and reduction in post-processing steps compared to HIP of canned parts. Results from research conducted on Inconel 625 are presented. This research is funded by DARPA/ONR contract N00014-95-C-0139, titled "Low Cost Metal Processing Using SLS/HIP".

9:40 am

PHYSICAL MODELING OF THE EARLY STAGES OF METAL POWDER COMPACTION AND COMPARISON TO EXISTING ANALYTICAL MODELS FOR HOT ISOSTATIC PRESSING: Henry R. Piehler and David P. DeLo*, Department of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, PA 15213-3890

Several important mechanisms observed to operate during the early stages of HIPing of powders are not currently included in existing compaction models, which all tend to underpredict strain rates and densification, especially during the early stages of consolidation. A series of interrupted HIP experiments were performed on Ti-6Al-4V powders consolidated in thin-walled containers to characterize early stage behavior using both metallographic sections and stereo pairs of fractured partially consolidated powder compacts. Results using several approaches indicated that particle rearrangement or granular behavior occurred throughout the early stages of consolidation, accounting for somewhere between 20% and 50% of the total densification. These approaches included densification predictions from measurements of center-to-center particle distances on metallographic sections and contact areas on fractured samples, both of which consistently underpredicted the measured Archimedian density levels. Particle size effects, including rigid body motion of larger particles facilitated by the preferential deformation of contiguous small particles, also were observed to contribute to densification by rearrangement. Increases in particle coordination number with densification, again absent in all current consolidation models, were also observed. The inconsistency between the mechanisms and assumptions included in existing HIP models and the early stage consolidation behavior observed here requires new modeling approaches which incorporate this observed behavior in order to improve the accuracy of model predictions. *Now at Wright-Patterson Air Force Base, OH.

10:10 am INVITED

HIGH PRESSURE SINTERING OF NANOCRYSTALLINE CERAMICS: R.S. Mishra, A.K. Mukherjee, Department of Chemical Engineering and Materials Science, University of California, Davis CA 95616

In recent years, nanocrystalline ceramics powder have been synthesized by a number of techniques. Some of these techniques have matured into commercial production of well characterized powders. Sol-gel processing and inert gas condensation techniques are particularly attractive. However, the retention of nanocrystalline microstructure during consolidation of some ceramics phases has proved to be quite difficult. In particular, the processing of nanocrystalline alumina based ceramics has been challenging. One of the ways to circumvent this problem is to use high pressure sintering. We have processed a number of nanocrystalline alumina and alumina based composites at 1 Gpa pressure in a piston-cylinder apparatus. The experimental results agree quite well with the theoretical calculations based on densification by dislocation mechanisms. Use of amorphous precursors provide additional benefits. Amorphous Zirconia-alumina precursor has been sintered to high density at 7000°C. An overview of the advantages of using high pressure sintering in processing nanocrystalline ceramics is discussed.

10:40 am INVITED

CRACK PATH TRANSITIONS IN POWDER PROCESSED CERAMIC MATRIX COMPOSITES: B.R. Patterson, S. Wu, and *P. Bhargava, Department of Materials and Mechanical Engineering, University of Alabama at Birmingham, Birmingham, AL 35284-4461; *Center for Ceramic Research, Rutgers University, Piscataway, NJ

Studies have been performed to better understand the nature of the crack path in particulate reinforced ceramic and glass matrix composites produced by powder processing. A theory has been proposed predicting the combined influence of stresses from thermal and elastic mismatch on crack path. The theory predicts changes in the tendency for crack attraction to or avoidance of the reinforcements with crack tip stress intensity for different combinations of thermal and elastic mismatch. The predictions have been verified by stereological quantification of the crack path in model composite systems. For some systems with thermal and elastic mismatch of opposite sign, the crack path showed reversal in attraction/avoidance tendency from low to high crack velocity, due to a change in dominant mismatch control.

11:10 am INVITED

INFLUENCE OF REINFORCEMENT ACTIVATION ON SINTERED CHAR-REINFORCED Al-Si ALLOY COMPOSITES: J.U. Ejipfor, R.G. Reddy, Department of Metallurgy and Materials Engineering, The University of Alabama, P.O. Box 870202, Tusaloosa, AL 35487-0202

Coconut shell chars dispersed in hypereutectic Al-13.5Si-2.5Mg alloy were studied for lightweight, wear applications. The composites were fabricated by a low-cost, double-compaction reaction-sintering technique. Both primary carbonized and activated chars (under CO₂ atmosphere) were used as the filler phases. The wear, mechanical and thermal behaviors of the composites investigated revealed optimum properties at 0.02 volume fraction of char with improvements when activated char (AC) was used. The wear rate and coefficient of fraction of the alloy reduced by 79 and 55 percent, respectively, when activated char was dispersed. While their tensile properties marginally declined from those of the alloy and the composite containing primary carbonized char (PC), the coefficient of linear thermal expansion fell from 11.7 x 10-6(0C)-1 for PC to 9.8 x 10-6(0C)-1. At 2.5 wt.%Mg, a good interfacial bonding was achieved, and it appeared to be unaffected by activation.

THERMOMECHANICAL PROCESSING AND MECHANICAL PROPERTIES OF HYPEREUTECTOID STEELS AND CAST IRONS: Session I: Processing of Ultrahigh-Carbon Steels and Cast Irons

Session Chairperson: Ray Decker, USP Holdings, 717 E. Huron, Ann Arbor, MI 48104; Gordon Geiger, Qualitech Steel Corporation, 301 Merchant Bank Bldg., 11 So. Meridian St., Indianapolis, IN 46204

INTRODUCTORY REMARKS: Donald Lesuer, Lawrence Livermore National Laboratory, Livermore, CA

8:35 am KEYNOTE, INVITED

THE HISTORY OF ULTRAHIGH CARBON STEELS: Jeffrey Wadsworth¹, Oleg D. Sherby², ¹Lawrence Livermore National Laboratory, P.O. Box 808, L-342, Livermore, CA 94550; ² Stanford University, Dept. Materials Science & Eng., Stanford, CA 94305

The history and development of Ultrahigh Carbon Steels (i.e., steels containing between 1 and 2.1% C and now known as UHCS) are described. The early use of steel compositions containing carbon contents above the eutectoid level is found in ancient weapons from around the world. For example, both Damascus and Japanese sword steels are examples of hypereutectoid steels. Their manufacture and processing is of interest in developing an understanding of the role of carbon content in the development of modern steels. Although sporadic examples of UHCS compositions are found in the early part of this century, it was not until the mid 1970s that the modern study began. This study had its origin in the development of superplastic behavior in steels and the recognition that increasing the carbon content was of importance in developing that property. The compositions that were optimal for superplasticity involved the development of steels that contained higher carbon contents than conventional modern steels. It was discovered, however, that the room temperature properties of these compositions were of interest in their own right. Following this discovery a period of intense work began on understanding their manufacture, processing, and properties. The development of laminated composites containing UHCS was an important part of this history.

9:05 am INVITED

EFFECT OF Al-ADDITION ON MICROSTRUCTURAL REFINEMENT OF ULTRA-HIGH CARBON STEELS: Dong-Wha Kum, Korean Institute of Science and Technology, P.O. Box 131, Cheongryang, 130-650, Seoul, Korea

Small amounts of Al-addition to ultra-high carbon steels leads to finer grain size during thermomechanical processing, and is known to improve superplastic property of the steels. In order to understand the refining effect, 0.32-1.75 wt%Al were added to 1.2%C + 1.5%Cr steel and its effect during hot and warm working has been studied. The grain size of prior austenite after hot working decreased with Al-addition. The interlamellar spacing of pearlite was studied by isothermal transformation experiments at undercoolings of 50 to 130°C below the A_l temperature. The interlamellar spacing also decreased with Al contents at undercoolings of 100°C and 130°C, while it was independent of Al at the undercooling of 50°C. The role of Al will be interpreted by its partitioning in boundaries and ferrite.

9:25am

HOT WORKABILITY OF HYPEREUTECTOID TOOL STEELS: H.J. McQueen¹ and C. Imbert², ¹Dept. of Mechanical Engineering, Concordia University, 1455 De Maisonneuve Blvd. W., Montreal, Quebec H3G 1M8, Canada; ²University of West Indies, St. Augustine, Trinidad

The four tool steels M2, D2, A2 and W1 with a wide range of alloying additions have been subjected to torsion testing over the ranges 900-1200°C and 0.1 to 4 s^-1 in order to determine the variation of strength and ductility expressed by suitable constitutive equations. Microscopic examination clarified the role of alloy carbides and confirmed the occurrence of dynamic recrystallization resulting in grain size dependence on temperature and strain rate and correlation with the flow stress developed. In addition, simulation of multistage rolling was simulated and softening in interpass intervals was determined. After summarizing, the above characteristics of these hypereutectoid steels with quite high carbide content over the working range, comparisons will be made on one hand with HSLA, low and medium carbon steels and on the other hand with austenitic stainless steels strengthened principally by solutes.

9:45 am

MICROSTRUCTURES AND MECHANICAL PROPERTIES OF AN ULTRAHIGH-CARBON STEEL PROCESSED BY THE DIVORCED EUTECTOID TRANSFORMATION: B. Walser¹, T. Oyama², U. Ritter², O.D. Sherby³, ¹Sulzer Brothers, Inc., Winterthur, Switzerland; ²WESGO, General Telephone and Electric, 477 Harbor Blvd., Belmont, CA 94002; ³Stanford University, Dept. Materials Science & Eng., Stanford, CA 94305

Fine spheroidized structures in a 1.5%C ultrahigh carbon steel (UHCS-1.5C) were achieved by processing routes involving hot and warm working (HWW) and the divorced eutectoid transformation (DET). These procedures eliminate the need for isothermal working required in previous processing of UHC steels. The divorced eutectoid transformation (DET) is accomplished by heating for about 60 minutes at 790°C, followed by air cooling. If deformation accompanies the air cooling step, then it is a DETWAD (Divorced Eutectoid Transformation With Associated Deformation). Quantitative scanning and transmission electron microscopy studies were performed to evaluate microstructures. It is shown that HWW or HWW + DETWAD processing results in superplastic behavior of the UHCS-1.5C material at 700°C because of the fine ferrite grain size present (~1.5 µm). The HWW + DET processed material has a ferrite grain size of 6 µm and is not superplastic at 700°C. Its room temperature properties, however, are impressive, exhibiting a tensile strength of 790 MPa and 35% elongation to failure.

10:00 am BREAK

10:20 am

THERMOMECHANICAL PROCESSING OF AUSTEMPERED DUCTILE IRON: J.D. DeLa'o, C.M. Burke, D.J. Moore, K.B. Rundman, Michigan Technological University, Metallurgical Engineering, 1400 Townsend, Houghton, MI 49931-1295

Austempered ductile cast iron (ADI) derives its beneficial properties from an ausferrite matrix (stable austenite with fine acicular ferrite) formed in a transformation that is similar to bainite formation in austempered steels. The isothermal treatment of metastable austenite at the austempering temperature lends itself to an ausforming process. To explore the potential benefits of ausforming ADI, a rolling operation was introduced into the austempering schedule at a point following the quench but preceding any significant formation of ausferrite. Variable alloy chemistries, austenitizing and austempering temperatures, austempering times and degrees of deformation were investigated. Results indicate that ausforming provides significant kinetic and microstructural benefits leading to marked simultaneous increases in yield strength and ductility for all conditions studied. A preliminary study of the metal working parameters relevant to ausforming ADI was conducted and a means of fabricating ausformed ADI components is suggested. An overview of the work is presented.

10:40 am

THE EFFECT OF ALLOYING ADDITIONS ON THE FORMABILITY OF LEDEBURITE OF WHITE CAST IRONS: Yelena Pirogov, State Metallurgical Academy of Ukraine, Ukraine

The present investigation has been made to study the plastic behavior of white cast irons with respect to alloying additions of vanadium and chromium. Annealed cast iron flat bars were hot rolled at 1050°C by reduction range from 25% to 65% at strain rates of 30 s^-1. The deformed specimens were examined using microstructure analysis and X-ray diffractometry. Dynamic structure forming mechanisms of austenite and cementite have been studied. Evolution of eutectic cementite crystallographic textures as a function of alloying additions and degree of plastic working has been determined. Mechanism of enhanced formability of vanadium alloyed white cast irons has been found.

11:00 am

AUSTEMPERING OF Mo ALLOYED DUCTILE IRONS: S. Yazdani, R. Elliott, University of Manchester and UMIST, Grosvenor St., Manchester, M17HS, UK

Measurements of ultimate tensile strength, 0.2% proof stress, elongation, hardness and impact energy are reported during austempering at 400, 375, 320 and 285°C after austenitizing at 870°C for 1-4320 minutes for ductile irons with the chemical composition 3.55% C, 2.72% Si, 0.25% Mn, 0.25% Cu and variable Mo content in the range 0.13-0.45%. X-ray diffraction and optical microscopy were used to determine the high carbon austenite content, carbon content of austenite, untransformed austenite volume and stage I and stage II austempering kinetics. A microstructural model was used to define the heat treatment processing window and the variation of the window with austempering temperature was established. The effect of Mo on the kinetics of austempering and mechanical properties are reported. The austempering processing window is shown to be open for all the austempering temperatures studied and the optimum properties correspond with the defined window. It is shown that excellent ductility combined with high strength can be attained well in excess of the ASTM A897M:1990 standard by controlling the heat treatment parameters for a low Mo content composition.

11:20 am

THERMOMECHANICAL PROCESSING AND PROPERTIES OF A DUCTILE IRON: Chol K. Syn¹, Donald R. Lesuer¹, Oleg D. Sherby²; ¹Lawrence Livermore National Laboratory, P.O. Box 808, L-342, Livermore, CA 94550; ²Stanford University, Dept. Materials Science & Eng., Stanford, CA 94305

Thermo-mechanical processing of ductile irons is a potential method for enhancing their mechanical properties. A ductile cast iron containing 3.6%C, 2.6%Si and 0.045% Mg was continuously hot-and-warm rolled or one-step press-forged from a temperature in the austenite range (900°C-1100°C) to a temperature below the A₁ temperature. Various amounts of reduction were used (from 60% to more than 90%) and then a short heat treatment at 600°C was given. The heat treatment lead to a structure of fine graphite in a matrix of ferrite and carbides. The hot-and-warm worked materials developed a pearlitic microstructure while the press-forged materials developed a spheroidite-like carbide microstructure in the matrix. Tensile properties including tensile strength and total elongation were measured along the directions parallel and transverse to the rolling direction and along the direction transverse to the press-forging direction. The tensile ductility and strength both increased with a decrease in the amount of hot-and-warm working. The press-forged materials showed higher strength (645 Mpa) than the hot-and-warm worked materials (575 MPa) when compared at the same ductility level (22% elongation).

11:40 am

A COMPARISON OF MECHANICAL BEHAVIOR IN PEARLITIC AND SPHEROIDIZED HYPEREUTECTOID STEELS: Eric M. Taleff¹, Chol K. Syn², Donald R. Lesuer² and Oleg D. Sherby³, ¹The University of Texas at Austin, Dept. of Aerospace Eng. & Eng. Mechanics, CO600, Austin, TX 78712; ²Lawrence Livermore National Laboratory, , P.O. Box 808, L-342, Livermore, CA 94550; ³Stanford University, Dept. Materials Science & Eng., Stanford, CA 94305

Hypereutectoid steels can exhibit remarkable mechanical properties at room temperature, principally because of the ability to develop fine microstructures through thermomechanical processing. Use of the divorced eutectoid transformation (DET) allows the development of fine, equiaxed microstructures with spheroidized carbide particles. The hypereutectoid austenite-cementite to pearlite transformation provides a thermal processing method which can produce pearlitic microstructures with very fine interlamellar spacings. A combination of DET and thermal processing is shown to create extremely fine microstructures containing controlled amounts of both spheroidized carbides and pearlite with controlled interlamellar spacings. Tension tests have been conducted on such microstructures, and the resulting strengths have proven to be predictable based on several microstructural parameters. Most remarkable is that both spheroidized and pearlitic microstructures are shown to obey the same predictive relation.