Materials Week '97: Monday PM 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 Monday afternoon, September 15, during Materials Week 1997. To view other programming planned for the meeting, go to the technical program contents page.

ALLOY DESIGN AND SOLDERING TECHNOLOGIES FOR LEAD-FREE AND LEAD-BEARING SOLDERS: Session II

Session Chair: S. Jin, Bell Laboratories, Lucent Technologies, Murray Hill, NJ 07974

ENERGETICS AND KINETICS OF DISSOLUTIVE WETTING PROCESSES: F.G. Yost, P.A. Sackinger, E.J. O'Toole, Sandia National Laboratories, Albuquerque, NM 87185-1411

Dissolutive wetting occurs when a liquid wets and spreads over a solid surface with simultaneous dissolution of the solid into the liquid. It is shown that this process initially yields a metastable equilibrium and a compact model for the kinetics of approach to this metastable state is described. The technique for constructing these kinetics stems from the early work of Onsager and begins with a relationship for the entropy production. From this, a coupled set of nonlinear, ordinary differential equations can be written directly. The equations are solved numerically for the wetted area and compared with experimental data. The model captures many of the subtle complexities of dissolutive wetting such as multiple metastable states. The main weaknesses of the model are the assumptions of fully-stirred liquid and a spherical solid/liquid interface which were made to keep the analysis simple. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy Contract number DE-AC04-94AL85000.

2:25 pm INVITED

THROUGH-HOLE SOLDER JOINT FILLET LIFTING PHENOMENON AND ITS POSSIBLE REMEDIES: C. Handwerker, NIST, Gaithersburg, MD 20899; Y. Zhu, Delco Electronics, Kokomo, IN 46904-9005; T.-Y. Pan, H.D. Blair, J.M. Nicholson, Ford Motor Company, Dearborn, MI 48121-2053

Fillet lifting failure, observed immediately after wave soldering and hand soldering with certain tin-silver based lead-free alloys, is characterized as separation between the copper pad and intermetallic layer of the solder joint. This observation raised serious concerns of usability of these lead-free solder alloys. Considerable efforts have been made to investigate the root cause of the failure and the possible remedy for the problem. A study of using Design of Experiment (DOE) was conducted to determine the effect of alloy composition, board thickness, pad size, and surface finish on the cracking problem. Results from this DOE study will be discussed.

2:50 pm INVITED

EFFECT OF SURFACE FINISH ON SOLDER FLOW OVER FINE LINE FEATURES: F.M. Hosking and C.L. Hernandez, Sandia National Laboratories, P.O. Box 5800, MS1411, Albuquerque, NM 87185

The rapid advancement of interconnect technology has necessitated the development of new surface finishes to replace the standard, technology-limiting copper and Sn-Pb solder coatings. As part of the effort, new solderability test methods have been developed to measure the effectiveness of the new metallic and organic coatings. Of particular interest is how solder flow over fine line features can be "engineered" through a science-based understanding of the wetting process. Sandia National Laboratories' Center for Solder Science and Technology has designed a capillary flow test geometry that addresses this surface flow issue. The presentation will examine the effects of circuit geometry, surface finish, aging, and solder processing on solder flow over the test surfaces. The results, and their significance to advanced interconnect technology, will be discussed.

3:15 pm INVITED

MICROSTRUCTURE AND MECHANICAL PROPERTIES OF ULTRAFINE-GRAINED SOLDERS: H. Mavoori, S. Jin, and T.H. Tiefel, Bell Labs, Lucent Technologies, Murray Hill, NJ 07974

A fine grain size is desirable in solders due to the improved fatigue resistance, strength and possible superplastic behavior. This leads to better reliability, durability during handling, and accommodation of strains without damage accumulation. In this work, ultrafine-grained lead-free and lead-containing solders have been obtained through the use of non-reacting, non-coarsening refractory dispersions. The processing involves powder metallurgy techniques followed by plastic deformation. Grain sizes were of the order of 2000Å and the microstructure was seen to be quite stable. This paper discusses the mechanical properties and microstructures of these solders with different dispersed particles such as Al₂O₃ and TiO₂ and contrasts them with a dispersion-free solder subjected to the same processing and testing conditions. The dispersions are seen to have a significant effect on the mechanical deformation characteristics with respect to tensile, fatigue and creep behavior of the solders and can be used to develop new solders with more desirable properties for various electronic packaging applications.

3:40 pm BREAK

4:00 pm INVITED

INVESTIGATION OF DEFORMATION MECHANISMS IN Sn-Ag AND Sn-Sb SOLDER ALLOYS USING TENSILE, CREEP AND ABI TESTS: K.L. Murty, H. Yang, P. Deane, P. Magill, G. Rinne, North Carolina State University, Raleigh, NC 2795; MCNC Electronic and Information Technologies, RTP, NC 27709

Excellent thermal fatigue characteristics of Sn-Ag and Sn-Sb solder materials made these alloys candidates for electronic applications to replace leaded solders such as Pb-5Sn and Pb-63Sn. A thorough understanding of these alloys is needed for developing reliable life prediction methodologies. Tensile and creep tests were performed on bulk Sn-3.5Ag and Sn-5Sb samples. In addition, automated ball indentation (ABI) tests were performed on the shoulder portions of the creep samples using a Stress-Strain Microprobe (SSM) at different indentation speeds at ambient temperature. Long-term creep data correlated with those derived using short-term ABI tests. Transitions in deformation mechanisms were observed with distinct values for the stress exponent, and in Sn-3.5Ag, the low-temperature dislocation climb due to dislocation core diffusion was identified at high stases. In addition, single lap shear tests were performed on 3333 solder bump array of Sn-3.5Ag which exhibited relatively more scatter.

4:25 pm INVITED

INFLUENCE OF INITIAL LAYER THICKNESS ON THE GROWTH OF Cu-Sn INTERMETALLIC LAYERS IN SOLDER-COPPER DIFFUSION COUPLES: M. Schaefer, R. Fournelle, Materials Science Program, Marquette University, Milwaukee, WI 53201-1881; J. Liang, Rockwell Automation, 1201 South Second Street, Milwaukee, WI 53204-2496

An experimental study was done to evaluate the growth kinetics of Cu-Sn intermetallic compound layers. Samples of solder (62Sn-36Pb-2Ag) on copper were created by reflow soldering for various times such that resulting intermetallic layers were 1, 2, 3, 6, and 12 microns in thickness. Samples were subsequently aged at 125 or 165°C for times ranging from 1 to 69 days. The growth rate of relatively thin layers was faster than for thick layers. For a 1 micron initial layer, growth followed a t^0.33 dependence on time, t. For increasing initial layer thicknesses, growth shifted toward a t^0.5 dependence on time. This behavior may be explained in terms of a growth model which assumes that growth of thin layers is related to grain boundary diffusion and growth of thick layers is controlled by volume diffusion through the intermetallic layer. Samples with an initially thin layer exhibited more rapid development of a Pb-rich zone adjacent to the intermetallic.

4:50 pm

MICROSTRUCTURES AND FATIGUE PROPERTIES OF Bi-CONTAINING SOLDERS: C. Zhang, and J.K. Shang, Department of Materials Science and Engineering, University of Illinois, Urbana, IL 61801

Effects of Bi-addition on fatigue properties of Sn-based solder joints were examined by introducing different amounts of Bi in the bulk solder or on surfaces of copper substrates. Flexural peel specimens were made by reflowing Bi-modified solder alloys between cleaned copper substrates and by reflowing unmodified Sn-Pb solder between Bi-coated copper substrates. Microstructures and fatigue properties of the Bi-modified solder joints were studied by optical and electron microscopy, and by conducting fatigue crack propagation along solder interfaces. The results were compared to the previous work on unmodified Sn-Pb/Cu joint. The relationship between interfacial microstructure and properties will be discussed.

ALLOY MODELING AND DESIGN: Session II: Short-Range Order and Phase Stability in Alloys

Session Chair: Prof. G. Ceder, Massachusetts Institute of Technology, 77 Massachusetts Ave., Cambridge, MA 02139

MEASUREMENT AND MODELING OF SHORT-RANGE CORRELATIONS IN METAL ALLOYS: Gene E. Ice, Cullie Sparks, Xiaogang Jiang, Ernest Epperson, Lee Robertson, Oak Ridge National Laboratory, Oak Ridge, TN 37831

Precision measurements of pair correlations in binary metallic solid solutions challenge the understanding of atomic size and bonding in local alloy structure. Measurements with tunable synchrotron X radiation can control the scattering contrast between atoms in a solid solution to produce a minimum and maximum contrast with elements nearby in the periodic table. We present measured pair-correlations for a series of binary Fe-Ni, Cr-Ni and Cr-Fe alloys. The response of the atoms to the local forces in terms or their displacements and ordering is discussed. Models are described which combine chemical correlations and static displacements. Theoretical calculations, of static atomic displacements, are needed to compare with observations. New efforts to statistically approach the modeling of SRO in alloys are also discussed. Research sponsored by the Division of Materials Sciences, U.S. Department of Energy under contract DE-AC05-96OR22464 with Lockheed Martin Energy Research Corporation.

2:40 pm INVITED

CALCULATION OF ALLOY PHASE DIAGRAMS BY CONTINUOUS CLUSTER VARIATION METHOD: K. Masuda-Jindo, Department of Materials Science and Engineering, Tokyo Institute of Technology, Nagatsuta 4259, Midori-ku, Yokohama 227, Japan; R. Kikuchi, Department of Materials Science and Engineering, University of California, Los Angeles, CA 90024-1595

A new formulation of the CVM which allows atomic displacement from lattice points is used to calculate the phase stability and phase diagrams of binary alloys. We formulate continuous atomic displacements around lattice points using the pair approximation of the CVM. We study the phase stabilities of binary alloy systems taking into account the local lattice distortions, with the use of the continuous displacement treatment of CVM, in terms of three kinds of pair probability functions g_ij and three pairwise potentials _ij. We focus our attention on model 2D alloys as well as fcc binary alloys, like Cu/Au systems. It is shown that the continuous displacement lowers the ordering temperature, and size mismatch in the constituent atoms increases the transition temperature. The local lattice-distortion effects in a disordered Cu-25 at. pct Au solid solution have also been investigated using the pair distribution functions determined by the continuous CVM.

3:20 pm BREAK

3:40 pm INVITED

FIRST-PRINCIPLES STUDY OF SITE OCCUPATIONS IN Ti-AlNb B₂ ALLOYS: M. Asta, D.D. Johnson, Sandia National Laboratories, Livermore, CA 94551

The characterization of any B2 structure in a ternary alloy requires, in addition to the average compositions, two independent order parameters. For Ti-Al-Nb alloys, these may be defined as _Al=(c_I^Al -c_II^Al)/2c_Aland _Nb = (c_I^Nb-c_II^Nb)/2c _Nb, in terms of the compositions on sublattices I and II. Recent experiments in Ti-Al-Nb have shown that the ratio _Al/_Nb is strongly concentration dependent. We investigate the structure of the B2 phase using two different electronic-structure-based methods. In the first, cluster-variation method calculations of _Al and _Nb are performed with interactions derived from the formation energies of ordered bcc compounds. In the second, based on the inhomogeneous coherent-potential approximation, an ordering susceptibility is calculated within the disordered bcc phase and used to determine _Al/_Nb as a function of composition. Both sets of calculated results will be compared and contrasted with experimental findings. Work is supported by the U.S. Department of Energy, OBES, Division of Materials, contract #DE-AC04-94AL85000.

4:20 pm INVITED

COMPARISON OF THE ELECTRONIC STATES OF ALLOYS FROM THE COHERENT POTENTIAL APPROXIMATION AND ORDER-N METHODS: J.S. Faulkner, Nassrin Moghadam, Alloy Research Center and Department of Physics, Florida Atlantic University, Boca Raton, FL 33431

The Coherent Potential Approximation (CPA) was derived using infinite-order perturbation theory and tested against one-dimensional models at a time when the calculation of the electronic states of disordered alloys was deemed impossible. Today, order-N methods such as the locally self-consistent multiple-scattering (LSMS) method are capable of obtaining the electronic structure of models made up of thousands of atoms, each described by self-consistent LDA atomic potentials. These models have been shown to be large enough to give realistic results for infinite alloys. We compare the predictions of the two approaches for the total energy, density of states, and other properties of disordered alloys. This research was sponsored in part by DOE grant number DE-FG05-89ER45392.

5:00 pm

X-RAY AND NEUTRON SCATTERING STUDY OF THE ORDER-DISORDER TRANSITION IN AlMnCu₂: J.J. Hoyt, Computational Materials Science Department, Sandia National Laboratories, Livermore, CA 94550; B.C. Chakoumakos, Solid State Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6393; S.T. Misture, High Temperature Materials Lab, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6064; R. McCormack, National Institute of Standards and Technology, Gaithersburg, MD 20899; M. Asta, D.D. Johnson, Sandia National Laboratories, Livermore, CA, 94551; J.D. Althoff, Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720

The long range order parameter in binary alloys, such as CuZn, can be measured as a function of temperature using x-ray diffraction. In ternary systems however, an extra degree of freedom exists such that two order parameters describe the transition and the measurement of both requires separate diffraction experiments using different incident radiations. In this study we have measured both order parameters in the B2-disorder transition of the Heusler alloy AlMnCu₂ using a combination of high temperature x-ray and neutron powder diffraction, analyzed by the Rietveld method. For temperatures just below the critical point (.88 < T/Tc < 1.0) it is shown that the concentrations of Al and Mn on a given sublattice are not equivalent. The results are compared with two separate first-principles, electronic structure calculations. This work is supported by the NSF under contract #DMR 93301220. ORNL is managed by Lockheed Martin Energy Research for the DOE under contract #DE-AC05-96OR22464. Work at Sandia was supported by the U.S. DOE, Office of Basic Energy Science, Division of Materials Science.

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

Session Chair: Robert C. Pond, The University of Liverpool, Liverpool, UK

INTERACTION OF DISLOCATIONS WITH INTERFACES IN A SOAP FROTH: M.E. Rosa, M.A. Fortes, Departmento de Engenharia de Materiais, Instituto Superior Tcnico, Lisboa, Portugal

The interaction of dislocations with interfaces was observed in tensile tests of mondispersed 2D foam samples between two parallel walls. A special device was used to prepare the samples and deform them. Dislocations are found to be reflected by the walls into the foam. When this happens, another dislocation is formed (Burgers vector conservation) that moves in the foam along the wall. Necking of the foam sample was observed in tension. Similar observations (i.e. reflection of dislocations) were made on bicrystalline foam samples (bubbles of two different diameters), but in this case there is some transfer of cells from one side to the other of the interface which eventually leads to an amorphous foam structure.

2:30 pm INVITED

STRUCTURE AND PROPERTIES OF DISLOCATIONS AND GRAIN BOUNDARIES IN SILICON: Matthew F. Chisholm, Metals and Ceramics Division, Oak Ridge National Laboratory, Oak Ridge, TN

With a scanning transmission electron microscope capable of forming a probe of atomic dimensions, a new approach to defect structure determination has become possible. Images formed using high-angle, elastically scattered electrons show strong atomic number (Z) contrast and reveal atomic column locations without the need for preconceived structure models. This experimental technique has been combined with simulations to study extended defects in silicon. This synergistic combination of experiment and theory has been used to determined the atomic and electronic structures of these important atomic configurations and has provided a remarkably self-consistent, atomic-scale picture of the segregation of As to silicon GBs.

3:00 pm INVITED

A THEORETICAL AND EXPERIMENTAL STUDY OF NON-PERFECT GRAIN BOUNDARY DISLOCATIONS: L. Sagalowicz, W.A.T. Clark, Department of Materials Science and Engineering, The Ohio State University, Columbus, OH 43210

Non-perfect grain boundary dislocations have been shown to appear in grain boundaries in a variety of materials. Recent theoretical approaches enable non-perfect grain boundary dislocations and grain boundary structures to be described, and utilize the power of group theory. Grain boundary dislocations may be divided into three classes: 1) perfect grain boundary dislocations, 2) imperfect grain boundary dislocations, and 3) partial grain boundary dislocations. Experimental transmission electron microscope evidence will be presented for boundaries in the diamond cubic structure, and it will be shown that imperfect and partial grain boundary dislocations play an important role in this system. The implications of these experimental observations for the description of grain boundary properties in terms of non-perfect dislocations will be discussed.

3:30 pm BREAK

3:40 pm INVITED

DISCONNECTIONS AS TRANSFORMATION DISLOCATIONS IN MARTENSITE TRANSFORMATIONS: R.C. Pond, Department of Materials Science and Engineering, The University of Liverpool, Liverpool L69 3BX, UK; J.P. Hirth, School of Mechanical and Materials Engineering, Washington State University, Pullman, WA 99614-2920

For phase transformations with well defined terrace planes, interface motion can occur by the motion of disconnections (defects with both step and dislocation character). Recent attention has focused on the role of such defects in diffusional transformations. The present work treats the role of disconnections in martensitic transformations. These defects are discussed in terms of symmetry theory. The symmetry imposes constraints that can influence the motion of the defects and hence the rate of the transformation.

4:10 pm INVITED

ANALYSIS OF MISORIENTATIONS AT THIN FILM INTERFACES BY THE PHENOMENOLOGICAL THEORY OF MARTENSITE CRYSTALLOGRAPHY: James M. Howe, Department of Materials Science & Engineering, University of Virginia, Charlottesville, VA 22903-2442; David A. Smith, Lehigh University, Bethlehem, PA 18015

It is quite common to find slight misorientations (tilts or rotations) associated with overgrowths on substrates in semiconductor multilayers and metal-ceramic systems. Several investigators have explained this phenomena using simple geometric analysis based on misfit accommodation at the interface. The purpose of this talk is two-fold: 1) to show that the misorientations at these interfaces can be readily explained by application of the phenomenological theory of martensite crystallography (PTMC) to the shape deformation between the overgrowths and substrates, and 2) to demonstrate that the simple geometric analyses that have been used to explain such phenomena are identical to the PTMC. Having made this comparison, it is apparent that the tilts and defects commonly observed in epitaxial layers may provide a number of examples with which to test the predictions of the PTMC in systems other than those having a martensite transformation. This work was performed during a visit with David Smith at IBM in May, 1992.

4:40 pm

HREM OBSERVATIONS OF MECHANICAL TWINNING IN Cu-Ti ALLOYS: STRUCTURE OF INCOHERENT AND COHERENT TWIN INTERFACES: T. Radetic, W.A. Soffa, V. Radmilovic , Department of Materials Science & Engineering, University of Pittsburgh, Pittsburgh, PA 15261; Department of Physical Metallurgy, University of Belgrade, 11001 Belgrade, Yugoslavia

The flow of Cu-Ti alloys hardened by coherent b' precipitates (Cu₄Ti, D1_a superstructure) is particularly interesting since these alloys show profuse twinning on multiple systems after small amounts of plastic flow by slip. In this paper, studies of the fine-scale structure of the mechanical twins which form in particle hardened Cu-Ti alloys are reported. The growth of deformation twins occurs via the motion of G-ledges at twin/matrix interfaces and via a coalescence of fine twins. The atomic structure of the coherent twin/matrix interfaces and associated single and multiple layer ledges are characterized. Incoherent interfaces at the tip of deformation twins generated by 90° and 30° twinning dislocations have been investigated. The twin/twin and slip/twin interactions and effect of grain boundaries on deformation twinning are also studied. Work supported in part by the MRC at the University of Pittsburgh sponsored by AFOSR and by the National Science Foundation (DMR) and by a Fellowship at the NCEM, Lawrence Berkeley National Laboratory, under Contract No. DE-AC-03-76SF00098.

5:00 pm

THE ROLE OF INTERFACES IN DEFORMATION TWINNING OF LAMELLAR TiAl CRYSTALS: L.M. Hsiung, T.G. Nieh, Lawrence Livermore National Laboratory, PO Box 808, L-370, Livermore, CA 94551-9900

Deformation twinning (DT) is known to play an important role in enhancing both room temperature ductility and high temperature creep resistance of lamellar TiAl alloys. The formation mechanism of DT in lamellar TiAl has recently been investigated using transmission electron microscopy. The (TiAl/Ti₃Al) interfaces in lamellar TiAl crystals provide additional nucleation sites for deformation twins, and promote the formation processes of DT in TiAl lamellae. It is suggested that DT in lamellar TiAl can be viewed as a stress-relief process for the pile-up of interfacial dislocations gliding along the lamellar interfaces during deformation. The deformation twins within TiAl lamellae are accordingly formed by a dislocation reaction based upon a stair-rod cross-slip mechanism.

DEFORMATION AND MICROSTRUCTURE: General Abstract Session II

Session Chair: Sriram Seshagiri, Wright Patterson Air Force Base, OH

X-RAY TOPOGRAPHIC STUDIES OF DISLOCATIONS IN ICE: I. Baker, X. Hu, D. Cullen, X. Pierron, Thayer School of Engineering, Dartmouth College, Hanover, NM 03755; M. Dudley, Dept. of Materials Science, State University of New York at Stony Brook, Stony Brook, NY 11794; D. Black, U.S. Dept. of Commerce, National Institute of Standards and Technology, Gaithersburg, MD 20899

THIS PAPER HAS BEEN WITHDRAWN

2:20 pm

DEFORMATION AND FRACTURE BEHAVIOR OF A BULK AMORPHOUS Zr-Ti-Ni-Cu-Be ALLOY: Peravudh Lowhaphandu, Lorie Ludrosky, John J. Lewandowski, The Case School of Engineering, Department of Materials Science and Engineering, Case Western Reserve University, Cleveland, OH 44106

The deformation and fracture behavior of a bulk amorphous Zr-TiNi-Cu-Be alloy were investigated. The effects of changes in stress state on the subsequent mechanical behavior were determined via a variety of test techniques, including compression, notched bend, and microhardness. SEM fractography was utilized to characterize the fracture surfaces, while X-ray diffraction and optical metallography were used to characterize the structure and evolution of deformation, respectively.

2:40 pm

FRACTURE TOUGHNESS AND FATIGUE-CRACK PROPAGATION IN A Zr-Ti-Ni-Cu-Be BULK METALLIC GLASS: C.J. Gilbert, R.O. Ritchie, Department of Materials Science and Mineral Engineering, University of California, Berkeley, CA 94720-1760; W.L. Johnson, Department of Materials Science, California Institute of Technology, Pasadena, CA 91125

The recent development of several alloy systems which readily form metallic glasses at low cooling rates (~10 K/s) has permitted novel measurements of both fracture toughness and fatigue-crack growth properties. Specifically, bulk plates of a Zr_41.2Ti_13.8Cu_12.5Ni₁₀Be_22.5 alloy were machined into compact-tension specimens with thicknesses and widths of ~7 mm and ~38 mm respectively. Fracture measurements on fatigue-precracked samples indicate that the fully amorphous structure has a fracture toughness of ~55 MPa. Heat treatments at 360°C for 12h and 450°C for 24h to form partially and fully crystallized microstructures, however, drastically reduce the fracture toughness to ~1.2 MPa and 1.0 MPa. Furthermore, the fully amorphous alloy is indeed susceptible to fatigue-crack growth, with a Paris-exponent and fatigue-threshold comparable to ductile crystalline metallic alloys such as high-strength aluminum or steel. Possible microstructural mechanisms for such behavior are discussed. Material supplied by A. Peker and M. Tenhover at Amorphous Technologies International.

3:00 pm

MECHANICAL AND STRUCTURAL PROPERTIES OF Mg ALLOY AZ31 IN HOT WORKING: A. Mwembela, E.V. Konopleva, H.J. McQueen, Mechanical Engineering, Concordia University, 1455 De Maisonneuve Blvd. W., H-549-34, Montreal, Quebec H3G 1M8, Canada

Alloys AZ31 (Mg-2.8Al-0.88Zn-0.OlMn) and AZ31-Mn (Mg-3.2Al-1.lZn0.34Mn) in torsion testing at temperatures T of 180 to 450°C at strain rates of 0.01, 0.1 and 1.0s^-1 exhibited flow curves with a peak and progression towards a steady state regime above about 300°C. The alloy AZ31-Mn exhibited higher ductility while alloy AZ31-Mn exhibited higher strength. The flow stress dependence could be described by the sinh relation and the T dependence by an Arrhenius function with activation energies of 125 and 138 kJ/mol for the two solute levels agreeing with those during creep in the range 300-400°C. When compared with die cast specimens of AZ91 tested earlier, these alloys exhibit less ductility, higher strength and activation energy. Optical microscopy revealed that dynamic recrystallization (I)RX) had nucleated at the grain boundaries and progressed more extensively above 300°C, but never completely replaced the original grains. The occurrence of DRX improved the ductility considerably. Below 300°C there was increasing evidence of twinning.

EXTRACTION AND PROCESSING OF TITANIUM: Session II: Processing of TiO₂ and TiCl₄

Session Chairs: Dr. R.G. Reddy, Department of Chemical and Metallurgical Engineering, MS 170, University of Nevada, Reno, NV 89557; Dr. J. N. Hryn, Argonne National Laboratory, 9700 S. Cass Avenue, Bldg. 362, Argonne, IL 60439-4815

WASTE METAL CHLORIDE PROCESSING AND CHLORINE RECYCLE: J.W. Reeves and R.G. Reeves, 3R Associates, 8 Wollaston Road, Wilmington, DE 19810

In the chlorination of titania ores to produce titanium tetrachloride, all minerals present are chlorinated except silica. These anhydrous chlorides are waste products that must be processed for disposal, recycle or sale. The ores chlorinated range in titania content from 50% to 96% with a resulting wide range of waste to be processed. More than 60% of the titania ores chlorinated are 85% titania or less and a large portion of this is less than 65% titania. Currently, none of the chlorine in this waste is recycled, less than 10% is processed for sale and the remainder is either treated for disposal, pumped into deepwells or dispersed into the sea. While there has been extensive research on chlorine recycle, no process has been found with sufficient economic potential. Instead, titanium tetrachloride producers have chosen to purchase higher titania ores when their additional cost is less than the waste processing cost. The metal chloride oxidation technology will be reviewed as well as chlorine recycled by other means such as electrolysis and pyrohydrolysis. Disposal technologies by neutralization and deepwell disposal that are currently practiced will be reviewed also. The prospects for improved chlorine recycle processes versus high grade ore cost will be discussed.

2:40 pm

NEW OPPORTUNITIES FOR THE TREATMENT OF CHLORIDE CONTAINING BY-PRODUCTS FROM THE TITANIUM PRIMARY INDUSTRY THROUGH PYROHYDROLYSIS: C. Kogler, F. Barhold, Process Development Division, Keramchemie GmbH, Postfach 1163, D-56425 Siershahn, Germany

During processing of different raw materials in the titanium/titaniumdioxide industry a lot of undesirable by-products can occur. One example is the production of synthetic rutile out of ilmenite. This titanium ore can be treated by leaching, to separate the rutile from accompanying elements. Through the leaching step a chloride solution with varying Fe, Mg, Al, etc. contents is produced. This waste liquid can be heated very efficiently by pyrohydrolysis in a fluidized bed reactor. The products are a regenerated acid and granular metal oxides. Pyrohydrolysis in fluidized bed reactors is a well known technology for the treatment of used pickling acids in the steel industry. The applicability of this technology for the titanium industry is addressed in this paper. Pilot plant trials have been carried out on different by-products of the titanium producers (solutions and slurries). Conclusions on how production plants for the treatment of those wastes can be operated are drawn from the results.

3:05 pm

REMOVAL OF IRON FROM MIXTURES OF IRON-TITANIUM OXIDES USING CARBOCHLO-RINATION PROCESS: F.C. Gennari, A.E. Bohe, D.M. Pasquevich, Centro Atomico Bariloche, CNEA, CONICET, 8400 Rio Negro, Argentina

Chlorination in the presence of carbon, carbochlorination, is considered one of the most attractive methods to separate metals, as chlorides, from low-grade polymetallic minerals. The carbochlorination of iron-titanium oxides mixtures was studied in the temperature range of 600 to 950°C using thermogravimetric measurements (TGA), XRD, SEM, EDS, chemical and neutronic activation analysis. The effect of kinetic variables, such as reaction temperature, chlorine flow rate, mass transfer in the solid bed and the iron-titanium oxides ratio, on the iron removal was investigated. It was found that iron quantitative separation occurs at low temperature. The carbochlorination showed the existence of different reaction stages which are strongly dependent on experimental conditions. Interactions between iron and titanium chlorides with the starting oxides were also observed, giving rise to an alteration of the mixture reactivity as a function of reaction conversion.

3:30 pm BREAK

3:45 pm

THE SCHEME OF PIGMENTARY TITANIUM DIOXIDE PRODUCTION BASED ON THE ORES OF TITANIUM-MAGNETITE ROCK DEPOSITS: N.A. Vatolin, L.I. Leontiev, S.V. Shavrin, Institute of Metallurgy, Ural's Division of Russian Academy of Sciences, 101 Amundsen Str., Yekaterinburg 620016, Russia

The usage of rock ores gives a rise to ecologic problems including much difficulties associated with the waste materials utilization. Technical solution of such problems depends on the character of ilmenite and titanium-magnetite dissemination in the ore and on their mutual inosculation with each other. The scheme of concentration of the ore containing from 7 to 17% of titanium dioxide was developed. This scheme eliminates the stage of flotation and provides for complete utilization of the waste materials in the forms of gravel. building sand and aggloporite. The economic efficiency is conditioned by obtaining of two merchantable high-grade concentrates - the iron-vanadium concentrate with 60-65% of iron and 0.8-1.1% of vanadium pentoxide and the ilmenite one with 42.5-45.0% of titanium dioxide. The total yield of concentrates depends on the ore's composition and structure and varies between 20-51%. The iron-vanadium concentrate is utilized by vanadium-making industry and the ilmenite one is used after special pretreatment for smelting of titanium slags. The pretreatment includes pelletization or pelletization and metallization. In the second case besides the melting in electric furnaces some more ways are considered providing for coagulation of the metal with it following extraction up to 90% by magnetic separation. The main attention in pigmentary titanium dioxide production is paid to its quality. The latter corresponds to the international quality standards requirements.

4:10 pm

SELECTIVE DISSOLUTION KINETICS OF THE ILMENITE: Z. Jin, L. Wang, Z. Duan, Department of Metal Materials, Chengdu University of Science & Technology, Chengdu, China 610065

Sichuan province in China is endowed with abundant Ilmenite resources. Because of the low titanium content and high magnesium-calcium inpurity, it was difficult to yield commercially acceptable product from the ilmenite using general benefication method. This investigation has demostracted an effective way to select dissolve ferrous ion from the Ilmenite in dilute hydrochloric acid and to produce artificial rutile containing 90% TiO₂. The kinetics of selective dissolution reaction was studied. The effects of stirring, temperature, particle size, HCl concentration were investigated. The results indicated that the selective dissolution reaction was topochemical and fit a surface reaction control model. The activation energy was calculated to be 56.94 KJ/mol. The order of the reaction rate are 2 with respect to HCl concentration. It was found that a small amount of phosphoric content in the ilmenite would significantly decrease the selective dissolution rate. And a mechanism of selective dissolution ferrous ion from ilmenite was proposed.

4:35 pm

LOW TEMPERATURE TiO₂ PRODUCTION PROCESS: L.D. Smillie, M. Heydenrych, Mattek- CSIR, Division of Materials Science & Technology, P.O. Box 395, Pretoria 0001, South Africa

A unique process has been developed for selective chlorination of low-grade titanium-containing ores. The ore is first treated by carbo-thermal reduction in which the titanium component of the ore selectively forms a carbo-nitride. This species chlorinates at far lower temperatures (typically 350°C) than the conventional high-temperature chlorination process. The other oxides are unreactive at these temperatures - a relatively pure titanium tetrachloride product can be obtained in the low-temperature chlorination. The carbo-thermally reduced ore can be produced at a cost that is competitive with conventional feedstocks on a weight-for-weight basis of titanium. Low-temperature chlorination also presents many opportunities for cost reduction over the existing processes. The remaining unchlorinated ore is not environmentally harmful - the selectivity of this process gives it an edge over the existing processes in terms of environmental impact because unwanted oxides are not chlorinated.

FATIGUE AND CREEP OF COMPOSITE MATERIALS: Session II

Session Chairs: David P. Walls, United Technologies, Pratt & Whitney, West Palm Beach, FL A.H. Rosenberger, Wright Laboratory Materials Directorate, Wright-Patterson Air Force Base, OH 45433

STRESS RATIO AND TEMPERATURE EFFECTS ON FATIGUE CRACK GROWTH BEHAVIOR OF SCS-6/Ti-6Al-4V: R. John^*, J.R. Jira, J.M. Larsen, Wright Laboratory Materials Directorate, Wright-Patterson Air Force Base, OH 45433-7817; ^*University of Dayton Research Institute, Dayton, OH 45469-0128

Titanium alloy matrix composites (TMC) are targeted for use in many aerospace applications. Critical turbine engine and aircraft components fabricated from continuous fiber reinforced TMC will experience cyclic loads during service, and many of these components typically contain crack initiators. Hence, characterization of the fatigue crack growth behavior of [0]₈ SCS-6/Ti-6Al-4V was initiated by the USAF Wright Laboratory under the MMC Life Prediction Cooperative Program. The results of the experimental and analytical investigation of fatigue crack propagation in SCS-6/Ti-6Al-4V material will be presented. Automated fatigue crack growth tests were conducted using middle tension, M(T) specimens at 23, 177 and 316C with stress ratios of 0.1, 0.5 and 0.7. During some of the tests, the crack opening displacement profile was measured to verify the stress distributions predicted by the fiber bridging models. This presentation will also discuss the capability of the shear lag models to predict the crack growth life of SCS-6/Ti-6Al-4V.

2:30 pm INVITED

ELASTIC SHIELDING AT BI-MATERIAL INTERFACES DURING FATIGUE CRACK GROWTH OF TITANIUM MATRIX COMPOSITES: S.G. Warrier¹, B.S. Majumdar¹, D.B. Miracle, Wright Laboratory Materials Directorate, Wright-Patterson Air Force Base, OH 45433-7817; ¹UES, Inc., Dayton, OH 45432

A "weak" fiber-matrix interface can act as a debonding layer and facilitate fiber bridging, whereas a "strong" interface can assist in the transfer of load to the fibers and retard the crack as the crack approaches a fiber. The latter phenomenon, elastic shielding which is caused by a mismatch in the modulus of the two materials, has not received much attention in the past. In this study, controlled experiments were carried out using single-ply composites with several carefully selected interfaces possessing different normal and shear strengths. Results show that elastic shielding is dependent on the mechanical properties of the interface; the strongest interface offering the largest extent of crack retardation. The crack shielding mechanism was further examined using finite element analysis. Results of numerical simulation and experimental results provide a detailed understanding of the influence of interface properties on elastic shielding and crack growth rates in titanium matrix composites.

2:50 pm INVITED

EFFECTS OF SiC FIBRE VOLUME FRACTION ON FATIGUE CRACK GROWTH RESISTANCE IN UNIFORMLY REINFORCED Ti MMCS: A.L. Dore, H. Whitlow, X. Wu, P. Bowen, School of Metallurgy and Materials/IRC in Materials for High Performance Applications, The University of Birmingham, UK

Effects of fibre volume fraction on the fatigue crack growth resistance of Ti MMCs have been assessed for volume fractions in the range from 10 to 40%. Mode I bridged cracks result from unbridged defects growing perpendicular to the fibre axis, and the influence of fibre volume fraction on crack growth resistance can be quantified and predicted. Acoustic emission techniques and in-situ observations have been used to establish the critical role of local fibre fracture in controlling the crack growth resistance of such composites. Attention will also be given to the resistance to crack growth parallel to the fibre axis as a function of fibre volume fraction. For a fibre volume fraction of 10% the role of the partial debonding of individual fibre-matrix interfaces can be distinguished, and cracks deviate towards such regions of debonding. At a fibre volume fraction of 35% several areas of fibre matrix debonding are observed to occur simultaneously ahead of a growing fatigue crack and the rapid linkage of such damaged regions produces a marked acceleration of crack growth. Clearly, as the fibre volume fraction increases, for such transverse growth the resistance of the composite to cyclic loading now decreases sharply. Within the overall paper, the compromise required to obtain adequate crack growth resistance as a function of fibre volume fraction in biaxial stress fields will be outlined.

3:10 pm INVITED

FIBER DAMAGE MECHANISMS IN TITANIUM METAL MATRIX COMPOSITES: M.N. Tamin, H. Ghonem, Mechanics of Materials Laboratory, Department of Mechanical Engineering, University of Rhode Island, Kingston, RI 02881

This work examines the fatigue damage mechanisms of SiC fibers at different temperatures. Results show that static and fatigue strengths of SCS-6 fibers are unaffected by the test temperature of 650°C and below. In addition, the onset of fiber fracture is instantaneous. Temperature influences the fracture process of these fibers through the density of cracks in the outermost carbon-rich coating layer. The composite damage in terms of crack distribution along the fiber is examined using the finite element method. In support of these calculations, Ti-MMC specimens were fatigued to the same number of cycles as employed in the fiber residual strength tests at the respective temperature. Results indicate that the mechanism of crack initiation in the coating/interphase region of a composite and in the carbon-rich coating of a SCS-6 fiber specimen are similar. In addition, the interphase zone does not influence the strength of the fiber in the temperature range of up to 650°C.

3:30 pm BREAK

3:50 pm INVITED

EVALUATION OF THE MMC LIFE 3.0 CODE IN PREDICTING CRACK GROWTH IN TITANIUM ALUMINIDE COMPOSITES: David Harmon¹, Alonso Peralta², James A. Hall², James M. Larsen³, ¹McDonnell Douglas, St. Louis MO, ²Allied Signal Engines, Phoenix, AZ; ³Wright Laboratory Materials Directorate, Wright-Patterson Air Force Base, OH 45433

Crack growth and fatigue life predictions made with the MMCLIFE 3.0 code are compared to test data for unidirectional continuously reinforced SCS-6/Ti-24Al-11Nb (SCS-6/Ti-24-11) laminates. The MMCLIFE 3.0 analysis package is a design tool capable of predicting strength and fatigue in metal matrix composite laminates. The code uses a combination of micromechanic lamina and macromechanic laminate analyses to predict stresses and uses linear elastic fracture mechanics for crack growth. The crack growth analysis now includes a fiber bridging model to predict matrix flaws in 0 degree laminates and is capable of predicting the effects of interfacial shear stress and thermal residual stresses. The code has also been modified to include edge notch flaws in addition to center notch flaws. The model was correlated with constant amplitude, isothermal data from crack growth tests conducted on 0 degree and 90 degree SCS-6/Ti-24-11 laminates. Verification tests were conducted which included dwell times and frequency effects. Strengths and areas for improvement for the analysis are discussed.

4:10 pm INVITED

PREDICTIONS OF TOTAL LIFE IN SiC FIBRE REINFORCED Ti MMCS: J.G. Pursell, J. Liu, W. Ding, D.C. Cardona* and P. Bowen, School of Metallurgy and Materials/IRC in Materials for High Performance Applications, The University of Birmingham, UK; *Rolls-Royce plc.

In fibre reinforced regions of envisaged components the possibility of minor defects and damage occurring cannot be ignored in any lifting assessment. Such damage could plausibly arise from a local fibre failure and/or defects at the ends of fibres which may be impossible to eliminate fully in some envisaged components. This paper will address primarily the growth of such defects under cyclic loading in a local environment where crack bridging can occur. The crack bridging analyses and models have been extended to simulate and incorporate the influence of fibre failure and fibre strength distribution changes on the integrated crack life of such composites. Model predictions of integrated fatigue life will be compared with S-N curves obtained experimentally: for testpieces containing damaged surface fibres; for testpieces containing totally embedded fibres; and for testpieces containing fibre ends. Typical values of initial defect sizes required to match model predictions with S-N curves obtained experimentally will be shown to be in good agreement with the sizes of defects that are expected to occur for composites reinforced with such large diameter (100-140 mm) fibres. The paper will thus attempt to summarize the progress made to date in specifying integrated lives for such composites and will consider if further dramatic improvements in integrated lives are likely to occur in the near future for the composites currently under development.

4:30 pm INVITED

FIBER-MATRIX INTERFACE SLIDING IN CONTINUOUS FIBER SCS-6/Ti MMC: E.A. DeBartolo, B.M. Hillberry, Purdue University, West Lafayette, IN; G.T. Ward, Allison Engine Company

In modeling the fatigue and fracture of continuous fiber metal matrix composites, the behavior of the metal matrix interface has been elusive. Debonding at the interface plays an important role in the degree of fiber bridging that affects the crack growth rate. The interface shear stress in the debond region has been assumed to be constant, vary along the length or used as a fitting parameter in fatigue crack growth models. In this study the debond region is investigated using a fatigue loading stage inside the chamber of an Environmental Scanning Electron Microscope (ESEM). The SCS6/Timetal®21S unidirectional test specimens are masked and a 4 to 7 mm section of the matrix dissolved through the thickness leaving the four rows of fibers exposed and intact. The area in the dissolved region where the fiber enters the matrix is clearly visible in the ESEM and the fiber pull-out can be observed and readily measured. In the paper, load and distance that the fiber slips on the first cycle and periodically during fatigue cycling will be presented. The change in fiber pull-out during cycling, fiber surface roughness, and presence of wear particles will provide insight to the fiber-matrix behavior.

4:50 pm

CREEP BEHAVIOR OF POWDER METALLURGY SiC-Al COMPOSITES AND THEIR Al MATRICES: Farghalli A. Mohamed, Materials Science and Engineering, Dept. of Chemical and Biochemical Engineering, University of California, Irvine, CA 92697

The effect of stress and temperature on the creep behavior of silicon carbide particulate reinforced Al alloys, produced by powder metallurgy, has been studied over several orders of magnitude of strain rate. The experimental data of the composites are examined in reference to those of their Al matrices that were creep tested under similar experimental conditions.

GEORGE R. IRWIN SYMPOSIUM ON CLEAVAGE FRACTURE: Session II: Dislocation and Atomistic Theories

Session Chairs: V. Vitek, Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, PA 19104; David L. Davidson, Southwest Research Institute, San Antonio, TX 78238

2:00 pm INVITED

KINETICS OF THE CRACK-TIP-GOVERNED BRITTLE TO DUCTILE TRANSITIONS IN INTRINSICALLY BRITTLE SOLIDS: A.S. Argon, G. Xu, Massachusetts Institute of Technology, Cambridge, MA 02139; M. Ortiz, California Institute of Technology, Pasadena, CA 91125

Several activation configurations of dislocation embryos emanating from tips of (001) cleavage cracks in alpha iron at the verge of propagating, have been analyzed in detail by the variational boundary integral method as central elements of the rate controlling process of nucleation governed fracture transitions from brittle cleavage to tough forms, as is expected to be the case for BCC transition metals. The configurations include those on inclined planes, oblique planes and crack tip cleavage ledges. Surface ledge production resistance is found to have a very strong embrittling effect. Only nucleation of dislocation embryos on oblique planes near a free surface and at crack tip cleavage ledges are found to be energetically feasible to explain brittle-to-ductile transition temperatures in alpha iron, in the experimentally observed ranges.

2:30 pm INVITED

MODELLING CRACK TIP PLASTIC ZONES AND BRITTLE-DUCTILE TRANSITIONS: P.B. Hirsch, S.G. Roberts, Department of Materials, University of Oxford, Parks Road, Oxford, OX1 3PH, UK

Models based on crack tip shielding by arrays of interacting dislocations on a single slip plane, representing the plastic zone, have been developed to explain the sharp and gradual brittle-ductile transitions in different materials. Sharp transitions occur when the density of crack tip sources is small, and the transition temperature then depends sensitively on the source density. Gradual transitions occur when the source density is large. For both types of transition the strain-rate dependence is controlled by dislocation velocity (which is related to yield stress). The application of this model to experiments on the brittle ductile transition of a number of materials will be described. Extensions of the model to cases, such as steels, where the fracture event may occur ahead of the main crack will also be discussed.

3:00 pm INVITED

A NEW MODEL OF THE BRITTLE-TO-DUCTILE TRANSITIONS BASED ON A COLLECTIVE DISLOCATION GENERATION INSTABILITY: Robert H. Folk, II, Steven M. Labovitz, and David P. Pope, Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, PA 19104

We have proposed a model of the brittle-to-ductile (BDT) transition based on a new cooperative mechanism of dislocation generation. In the existing models, the BDT is assumed to be either controlled by the nucleation of a single dislocation or the mobility of a group of dislocations. The effects of temperature enter these models only via thermally activated generation or motion of dislocations. In contrast, the model advanced in this work suggests that the BDT corresponds to a combined thermal-and-stress induced cooperative instability of a large number of dislocation loops. The analysis is carried out in the framework of statistical mechanics and is closely related to the well-known Kosterlitz-Thouless type dislocation instability in two dimensions. The new model not only identifies the specific role of the crack tip in the BDT but also suggests that a sudden onset of plasticity at a certain temperature is possible in crack-free crystals.

3:30 pm BREAK

3:40 pm INVITED

INSIDE THE CRACK TIP: Michael Marder, Department of Physics and Center for Nonlinear Dynamics, The University of Texas at Austin, Austin, TX 78712

This talk will discuss experimental, numerical, and analytical studies of cleavage fracture, with special attention directed towards consequences at the macroscopic scale of atomic detail. I will show that a range of crack velocities is forbidden in brittle materials at low temperatures because of atomic effects, and that the ultimate speed of cracks is limited by dynamical instabilities. These claims will be illustrated in experiments on amorphous and crystalline materials, explained by exact solution of idealized models, and backed up by computer simulations designed to allow comparison of large and small scales.

4:10 pm

ATOMISTIC SIMULATIONS OF FRACTURE: Diana Farkas, Department of Materials Science and Engineering, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061

Embedded atom interaction potentials are used to simulate the atomistic aspects of the fracture process. Simulations are presented for the behavior of cracks in pure metals and intermetallics, near the Griffith condition. The materials considered include Fe, Cu, Ni as well as Fe, Ni, Co and Ti aluminides. The work focuses on the comparative study of fracture behavior in the different materials. The role of the atomic relaxation at the crack tip and of lattice trapping phenomena is analyzed.

4:30 pm

EFFECT OF IMPURITIES ON CLEAVAGE FRACTURE: H, C, B, AND S IN Ni₃Al: Nicholas Kioussis and G. Lu, Department of Physics, California State Univ. Northridge, Northridge, CA 91330-8268; M. Ciftan, US Army Research Office, Research Triangle Park, NC 27709; A. Gonis, Lawrence Livermore National Laboratory, Livermore, CA 94550

The Ll₂ intermetallic Ni₃Al exhibits unique mechanical properties that make it attractive for high temperature structural applications. Understanding the origins in the electronic structure of the impurity-induced strengthening or impurity-induced environmental embrittlement in Ni₃Al is of practical importance. The effects of hydrogen, carbon, boron, and sulfur impurities on the ideal cleavage fracture properties of Ni₃Al under tensile stress are investigated using total-energy full-potential electronic structure calculations with a repeated slab arrangement of atoms simulating an isolated cleavage plane. Results for the stress-strain relationship, cleavage energies, ideal yield stress and strains with and without impurities are presented, and the electronic mechanism underlying the contrasting effects of impurities on the ideal cleavage of Ni₃Al is elucidated. *Supported by US Army Research under contract No. DAAH04-93-G-0427.

8:00 pm GEORGE R. IRWIN SYMPOSIUM BANQUET LECTURE: INVITED

GEORGE R. IRWIN: THE MAN WHO SHOWED ENGINEERS HOW TO PUT FRACTURE TO WORK: H.P. Rossmanith, Institute of Mechanics, Technical University Vienna, International Society for Technology, Law & Technical Insurance, ISTLI Secretariat General Vienna, Wiedner Hauptstr. 8-10/325,A-1040 Vienna, Austria

The lecture presents a short biography of the Father of Fracture Mechanics whose professional life has been entirely devoted to the development and application of the field of fracture research. First, the general scientific and engineering setting of the time of his conversion from "journalism" to engineering will be illuminated. His appointment to the Naval Research Laboratory and the early work on armor followed by the path-breaking early papers on the basics of fracture mechanics and ensuing pilgrimage of scholars from all over the world to the NRL and, later, to Lehigh University and the University of Maryland. Some of the key developments in fracture mechanics and the decisive contributions by Professor George R. Irwin will be highlighted in this presentation. The sheer incomprehensible task of being an outstanding scientist and engineer, a very much sought after consultant for more than 50 years, an exemplary and beloved husband, father and head of a large family, a helpful and advisory professional colleague, a partner in sports and leisure and a superb teacher and promoter of the young and students in engineering, can only be achieved and flourish in a family setting created and tended by his charming and wonderful wife Georgia and his extremely helpful and caretaking children. The presentation will close with a few bonmots collected by the friends of the jubilant.

MECHANICAL BEHAVIOR OF BULK NANO-MATERIALS: Session I

Session Chair: Walter W. Milligan, Department of Metallurgical and Materials Engineering; Michigan Technological University, Houghton, MI 49931-1295

Naresh N. Thadhani, School of Materials Science and Engineering, Georgia Institute of Technology; Atlanta, GA 30332-0245

2:05 pm INVITED

DEFORMATION CHARACTERISTICS OF NANOCRYSTALLINE METALS, INTERMETALLICS, AND COMPOSITES: Shankar Sastry, W.E. Buhro, R.L. Axelbaum, Washington University, St. Louis, MO 63130

Nanocrystalline powder compacts of metals- Cu, Ni, and W; intermetallics - NiAl and MoSi₂, and composites - Cu-TiB₂ and Ti-TiB_D, were prepared by hot pressing and hot isostatic pressing of nanoparticles produced by solution phase synthesis and gas-phase combustion synthesis. The compacts were characterized by microhardness measurements, compression tests, three-point bend tests, and indentation creep measurements. Nanocrystalline MoSi₂, NiAl, Ni, and Cu exhibit significant creep at T < 1/2T_m. Nanocrystalline copper exhibits superplasticity over a narrow temperature/strain rate window. Nano-grained tungsten exhibits anomalous strain rate dependence of flow stress and ductility. These unique characteristics will be discussed in terms of the extent of dislocation activity in nanograined materials, role of grain boundary diffusion and sliding, and adiabatic shear band formation at high strain rates. This research was conducted under NSF and AFOSR grants.

2:35 pm

FABRICATION OF NANO-GRAINED BULK NICKEL ALUMINIDES FROM MECHANICALLY ALLOYED PRECURSOR: T. Aizawa, Department of Metallurgy, University of Tokyo, 7-3-1 Hongo, Bunky-ku, Tokyo 113, Japan

Bulk mechanical alloying was first used to yield fine-grained powder mixture and compact; both aluminum and nickel powders are refined into lamellar structured mixture with the mutual distance less than 1 m. After densification of this compact by uniaxial compression at room temperature, this Ni-Al mixture is reactively sintered into nano-grained, bulk nickel aluminides. 3 Ni+1Al system is employed here to experimentally demonstrate that Ni₃Al billet and wire can be directly fabricated by the above procedure from the mechanically alloyed precursor.

3:00 pm

MECHANICAL BEHAVIOR OF CONSOLIDATED, ATTRITOR MILLED, NANOCRYSTALLINE Fe, Fe Al, AND Fe - C ALLOYS: J. Rawers, R. Krabbe, N. Duttlinger, U.S. Dept. of Energy, Albany (Oregon) Research Center, 1450 Queen Ave. SW, Albany, OR 97321

Although numerous studies have been conducted and reported on the production and characterization of nanostructured powders, there have been few studies that have characterized the macroscopic properties of consolidated nanostructured materials. Much of this difficulty results from the limited nanomaterial production capability and from the difficulty in retaining the nanostructure during consolidation. In this study, attritor milled Fe, Fe-Al and Fe-C powders were consolidated by hot-pressing. The effect of differing processing conditions (time, temperature, and pressure) on the resulting compact and nanostructure are presented. Compacted samples were of sufficient size that macroscopic properties such as density, hardness, and tensile and compression strengths could be evaluated. Relationships between (a) the milled and the consolidated microstructure, and (b) between nanostructure and macroscopic properties are explored and explained.

3:25 pm BREAK

3:40 pm

BULK ULTRAFINE GRAINED MATERIAL OBTAINED BY INTENSE PLASTIC STRAINING: P.B. Berbon,¹ N.K. Tsenev,² R.Z. Valiev,³ M. Furukawa,⁴ Z. Horita,⁵ M. Nemoto,⁵ and T.G. Langdon;^{1 1}University of Southern California, Los Angeles; ²Ufa State Petroleum University, Ufa, Russia; ³Ufa State Aviation University, Ufa, Russia; ⁴Fukuoka University of Education, Munakata, Japan; ⁵Kyushu University, Fukuoka, Japan

Using the Equal-Channel Angular Pressing (ECAP) technique, a large bulk piece of material can be transformed into Ultrafine Grained (UFG) structure through the introduction of intense plastic deformation. This paper describes the microstructural characteristics and mechanical properties of pressed Al-Mg alloys and a commercial Al-Mg-Li-Zr (01420) alloy having UFG structures. Mechanical testing was performed at temperatures up to 603 K and at strain rates from 10^-4 to 10^-2 s^-1. The strength was found to be stable for the 01420 alloy and it increased for the Al-Mg alloy. The elongation to failure was improved in all cases after ECAP, and there was a very substantial improvement in the ductility of 01420 alloy. The microstructure of 01420 was also found to be remarkably stable upon annealing.

4:05 pm

CHARACTERIZATION OF Cu-BASED MULTILAYERED STRUCTURES PRODUCED BY ELECTRODEPOSITION: F. Ebrahimi, Q. Zhai, D. Kong, Materials Science and Engineering Department, University of Florida, Gainesville, FL

Copper-silver and copper-nickel multilayered structures with nano-size layer thickness were produced using electrodeposition techniques. Low temperature heat treatments were conducted to investigate the microstructural stability of the layers. The microstructure was evaluated using electron microscopy techniques. Mechanical properties were studied using tensile testing at room temperature. Electrical resistivity measurements and x-ray diffraction analysis were performed for evaluation of defect structure and crystallographic texturing, respectively. The results of this study indicate that very high strength levels can be obtained in these structures. The strength and fracture mechanism were found to be a complex function of residual stresses, continuity of the layers, and interface structure.

4:30 pm

MECHANICAL PROPERTIES OF ELECTRODEPOSITED NANOCRYSTALS: D. Clark,¹ G. Palumbo,² K.T. Aust,³ and U. Erb,^{3 1}Nanometals Corporation, Queen's University, Kingston, Ontario, Canada K7L 3N6; ²Ontario Hydro Technologies, 800 Kipling Ave., Toronto, Ontario, Canada M8Z 5S4; ³Department of Metallurgy and Materials Science, University of Toronto, Ontario, Canada M5S 3E4

The mechanical properties of fully dense bulk nanostructured metals and alloys produced by electrodeposition will be reviewed. Particular emphasis will be on the transition from regular to inverse Hall-Petch behavior which is typically observed in hardness measurements on these materials for grain sizes less than 30 nm. The effect of annealing on the hardness of some of the alloys will also be discussed.

MICROSTRUCTURE EVOLUTION, CHARACTERIZATION AND MODELING: Session II: Liquid Solid Systems II

Session Chair: J.A. Dantzig, University of Illinois, Dept. of Mech. & Industrial Eng., 1206 W. Green St., Urbana, IL 61801

MODELING OF THE GROWTH AND INTERACTIONS OF EQUIAXED DENDRITES ON A MESOSCOPIC SCALE: B. Kauerauf¹, I. Steinbach¹, C. Beckermann² and J. Guo¹, ¹ACCESS e.V., D-52056 Aachen, Germany; ²The University of Iowa, Iowa City, IA 52242-1527

The interactions between multiple equiaxed dendritic grains during diffusion-controlled growth into the undercooled melt of a pure substance are modeled using a novel mesoscopic simulation technique. The mesoscopic scale is of the order of the diameter of the dendrite envelopes, which is large compared to the interdendritic spacings. In the model, the calculation of the temperature field in the undercooled liquid is coupled with a modified stagnant film model for dendritic growth, and the evolution of the internal solid fraction inside the grain envelopes is predicted. Three-dimensional numerical results are presented for the transient growth of multiple grains in the presence of strong thermal interactions.

2:30 pm

MESOSCALE MODELING OF CONVECTIVE EFFECTS DURING SOLIDIFICATION: S.P. Marsh and S.G. Lambrakos, Code 6320, Materials Science and Technology Division, Naval Research Laboratory, Washington, DC 20375-5343

A stereology-based modeling approach is presented that combines macroscopic fluid flow with microscopic solidification mechanisms. This mesoscale method allows mass balances arising from buoyancy-driven flows to be coupled directly to local solidification phenomena. Simulation results describing the effect of fluid flow on cellular spacings will be presented and compared with experimental data. This work is being supported by NASA under Grant NAG8-1272.

3:00 pm

DEVELOPING MICRO-SEGREGATION MODELS FOR MULTI-COMPONENT ALLOYS: Vaughan R. Voller, Saint Anthony Falls Laboratory Mississippi River at 3rd Ave., SE University of Minnesota, Minneapolis, MN 55414

During the solidification of an alloy mass transport on the local scale of the secondary dendrite arm spaces is controlled by diffusion. This process is referred to as micro-segregation. The standard micro-segregation models focus on closed systems in binary alloys solidifying in a prescribed fashion. The term closed system implies that the mixture concentration within a given Representative Elementary Volume (REV) of the mushy region remains fixed during the solidification. In the context of a complete solidification model this feature is not realistic. In real solidification systems the mixture concentration in the REV will change with time--Macro-segregation. Furthermore, the alloy may not be binary and the solidification path may not be known a-priori. The object of this paper is to investigate models that can be used to model micro-segregation in multi-component alloys in open REV's (i.e., in a regime with time varying mixture concentrations). The focus will be on: (1) an investigation of the possible numerical schemes for modeling the micro-segregation (e.g., explicit vs. Implicit and fixed grid vs. deforming grid) and (2) a discussion on the coupling between micro and macro scale models, i.e., looking at the question -What REV values should be used to control the micro-segregation model?

3:30 pm

CRYSTALLOGRAPHIC EVOLUTION IN DIRECTIONALLY SOLIDIFIED MICROSTRUCTURES: Krishna Rajan, Jeffrey Trogolo, Department of Materials Science and Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180-3590

In this presentation we describe the crystallographic characterization of directionally solidified nickel-based superalloys using electron diffraction techniques. It is shown that there appears to be some maximum level misorientation that is prevalent in these systems. Transmission electron microscopy of the low angle boundaries in the region of the microstructure near the higher levels of misorientations shows high levels of twisting of the grain boundary plane along the axis of solidification growth. The relationship between the evolution of grain boundary plane misorientation and the mechanism of misorientation evolution of the overall sample is discussed.

4:00 pm

MICROSTRUCTURE EVOLUTION DURING SOLIDIFICATION OF A VIGOROUSLY STIRRED MELT: J. Roplekar, J.A. Dantzig, University of Illinois, Dept. of Mech & Industrial Eng., 1206 W. Green St., Urbana, IL 61801

Production of metallic parts using semi-solid forming techniques has become a commercially viable process. The process uses feedstock which is characterized by rounded primary phase, surrounded by solute-rich regions. During reheating, these solute-rich regions melt at lower temperature, resulting in a mixture which is relatively easily deformed into complex shaped parts. In this work, we describe the MHD-DC casting process, wherein a rotating electromagnetic field is used to impart rotation to the melt during solidification. Studies are described to correlate the observed microstructures with experimental conditions.

4:30 pm

THE EFFECT OF IRON CONCENTRATION ON POROSITY FORMATION IN 319 CAST ALUMINUM ALLOY: J.W. Zindel, Ford Motor Company, MD 3182 SRL, P.O. Box 2053, Dearborn, MI 48121-2053

Iron is a ubiquitous element in aluminum alloys. Sources for iron include impurities in the bauxite ore, contamination in the recycling stream, intentional additions to reduce the propensity for die soldering in the die casting process, and poor molten metal handling practices. Iron has been attributed to increasing the propensity for microporosity formation in castings. This work studied the effect of increasing the iron content in a 319 type alloy on microporosity formation in a well fed casting. The casting was a wedge shape with a chill placed at the thin edge of the wedge to generate parallel isotherms progressing towards a large riser and a solidification times which ranged from 15 to 2150 seconds. Four iron concentrations were studied: 0.38, 0.59, 0.83, and 0.95. Density measurements were used to determine porosity levels. The calculated porosity is a strong function of the absolute density of the material and the various techniques used to determine the absolute density will be discussed. The absolute density of the material did increase with iron concentration but the porosity did not appear to be a function of iron concentration in this casting configuration.

5:00 pm

EFFECT OF GRAVITY ON THE MICROSTRUCTURAL EVOLUTION OF TUNGSTEN HEAVY ALLOYS: A. Tewari, A.M. Gokhale, School of Materials Science & Engineering, Georgia Institute of Technology, Atlanta GA 30318

Tungsten heavy alloys are usually manufactured by liquid phase sintering of tungsten powder along with Ni and Fe powders. This results in a two phase material having grains of almost pure tungsten embedded in a matrix of W, Ni, Fe alloy. The evolution of microstructure during LPS also depends on gravity. To understand the role of gravity on the evolution of microstructure, LPS experiments were performed in normal gravity and under micro-gravity conditions of space shuttle. The microstructure of these sintered alloys were quantitatively characterized in detail using Digital Image Analysis to gauge the effect of gravity on the process of LPS. This article presents differences found in the evolution of microstructures of liquid phase sintering under gravity and micro-gravity environment.

MODELING CARBURIZING, NITRIDING AND RELATED DIFFUSIONAL SURFACE TREATMENTS: Session III: Microstructure and Properties I

Session Chair: M.A. Howes, IIT Research Institute, Chicago, Illinois

A HISTORICAL OVERVIEW OF THE MODELING AND PREDICTING OF CONCENTRATION AND HARDNESS PROFILES IN CARBURIZED AND NITRIDED STEELS: R.D. Sisson Jr., R.R. Biederman, M.M. Makhlouf, Worcester Polytechnic Institute, 100 Institute Rd., Worcester, MA 01609

The progress in the ability to model, predict, and control the hardness and composition profiles in carburized and nitrided steels will be presented from an historical perspective. Starting with the rituals of the ancient blacksmiths, progressing through a variety of graphical and empirical techniques, to the use of sophisticated computer models and the improvements in the ability to model, predict, and control to carburization and nitriding of steels will be discussed.

2:00 pm

FINITE ELEMENT SIMULATION OF THE "CORNER EFFECT": J.E. Morral, University of Connecticut, Dept. of Metallurgy and Materials Engineering, Storrs, CT 06269-3136; B. Dupen, WIX, P.O. Box 1967, Gastonia, NC 28053-1967

When compared with flat surfaces, the case depth of carburized parts is greater at external corners and smaller at internal corners. A finite element model that simulates vacuum carburizing was applied to corners in a plain carbon steel treated at 1040°C. The results showed that when the radius of curvature of a corner was similar in dimension to the case depth, the "corner effect" was negligible. Also, the model showed that the corner effect was amplified by two step, boost-diffuse heat treatments yielding significant variations in case depth regardless of corner radius.

2:30 pm

SIMULATION OF CARBURIZATION IN SECONDARY HARDENING STEELS: C.J. Kuehmann, BIRL Industrial Research Laboratory, 1801 Maple Ave., Evanston, IL 60201; J.P. Wise, C.E. Campbell, G.B. Olson, Northwestern University, Materials Science and Engineering Dept., 2225 N. Campus Dr., Evanston, IL 60208

The ThermoCalc and DICTRA codes have allowed the simulation of carburizing behavior of advanced secondary hardening gear and bearing steels. These simulations used in conjunction with materials systems design techniques and other computational tools, produced high-performance gradient structures for gear and bearing applications. Process simulations consider the 1-D multicomponent diffusion and simulataneous carbide precipitation from local equilibrium conditions during carburizing. Using the simulated carbon profile and a model for the heterogeneous precipitation of coherent M2C carbides, coupled with calculations for the precipitation strengthening behavior, the resulting hardness profile has been calculated and compared with experimental observations. Due to the high strengthening efficiency achieved during secondary hardening, these steels can attain surface hardness of greater that 60 Rockwell C at carbon levels below 0.6 wt%. Additionally, these properties can be achieved without the presence of primary carbides in the case microstructure. Preliminary bending fatigue results indicate this can significantly increase bending fatigue life.

3:00 pm BREAK

MODELING CARBURIZING, NITRIDING AND RELATED DIFFUSIONAL SURFACE TREATMENTS: Session IV: Microstructure and Properties II

Session Chair: R.D. Sisson Jr., Worcester Polytechnic Institute, Worcester, MA

MODELING THE KINETICS OF NITRIDING/NITROCARBURIZING OF IRON: M.A.J. Somers, E.J. Mittemeijer, Delft University of Technology, Laboratory of Materials Science, Rotterdamseweg 137, NL-2628 AL Delft, The Netherlands

The kinetics and the morphological and compositional development of the compound layer during nitriding and nitrocarburizing of pure iron has been investigated for various temperatures and various combinations of imposed nitrogen and carbon activities. The results indicate that no local equilibrium occurs at the gas/solid interface during nitriding/ nitrocarburizing, due to the slow kinetics of ammonia dissociation, the development of N₂ and the fast initial carbon uptake. The kinetics of gaseous nitriding of pure iron can be adequately described by a model that considers local equilibrium at the solid/solid interfaces and a composition-weighted intrinsic diffusion coefficient of nitrogen in the Fe-N phases. For nitrocarburizing modelling appears not possible yet, due to changing compositions at gas/solid and solid/solid interfaces and the lacking of diffusion coefficients of C in Fe-N-C phases.

4:00 pm

ELASTIC LAYERED MODEL OF GAS/PLASMA NITRIDED FE-CR SYSTEM BY ACOUSTIC SPECTRO MICROSCOPY: T. Aizawa, University of Tokyo, Dept. of Metallurgy, 7-3-1 Hongo, 113 Tokyo, Japan; H. Kuwahara, Institute of Applied Science, Kyoto 606, Japan

An Fe-Cr alloy for various mole content of Cr was gas/plasma nitrided with different temperature and processing time. The nitrided specimens were analyzed by XRD and observed by microscope to investigate the microstructure of iron and chromium nitrides distributed from the surface in the direction of thickness. Acoustic spectro-microscopy was further utilized to measure the surface wave velocity dispersion and to determine the variation of elastic constants in thickness. The effect of nitridation conditions on the layered elastic model was discussed both for gas and plasma nitridation.

4:30 pm

MODELING THE FORMATION, BUILD-UP AND GROWTH OF CARBONITRIDE PHASES AND LAYERS IN THE CARBONIT PROCESS: R. Roussev, S. Malinov, P. Petrov, Technical University of Varna, Studentska str. 2 dep. MTM, 9010-Varna, Bulgaria

Abstract not available.

PROCESS AND MATERIAL BEHAVIOR MODELING SYMPOSIUM: Session II

Session Chair: Glenn S. Daehn, Department of Materials Science and Engineering, The Ohio State University, Columbus, OH 43210

EFFECTS OF STRAIN LOCALIZATION ON SURFACE ROUGHENING DURING SHEET FORMING: R.C. Becker, Alcoa Technical Center, 100 Technical Drive, Alcoa Center, PA 15069; H.R. Piehler, Dept. of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, PA 15213

Numerical simulations of evolving surface roughening in sheet have been performed to determine the influence of microstructure and mechanical properties. The model accounts for the grain structure near the sheet surface with the behavior of the grains being characterized by a constitutive model which accounts for deformation by crystallographic slip and for rotation of the crystal lattice with deformation. In addition to the known linear dependence of surface roughening on strain and grain size, it was determined that small scale strain localization at the surface plays a significant role. Consequently, factors which affect strain localization, such as strain hardening, texture, and material homogeneity, also affect surface roughening. The results also show patterning of the strain localization which is induced by the material inhomogeneity inherent in a polycrystal.

2:20 pm

THE INFLUENCE OF ENGINEERED SURFACE TEXTURE ON THE FORMABILITY OF ALUMINUM SHEET: G.W. Jarvis, Alcoa Technical Center, Alcoa Center, PA 15069; H.R. Piehler, Dept. of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, PA 15213; R.C. Becker, L.G. Hector, Alcoa Technical Center, Alcoa Center, PA 15069

To date the major impact of engineered surface texture has been its influence on tribological characteristics during sheet forming. However, the presence of the engineered surface texture may also influence the formation of strain localizations during forming. These localizations may either increase formability by spreading strain or decrease formability by leading to early localized necking. The results from large strain experiments on flat sheets with four different electron-beam textures are presented. These sheets were subjected to four different strain states ranging from drawing to plane strain using the CMU sheet-metal deformation simulator. Finite element modeling of the deformation of these surface-textured aluminum sheets was used to provide additional insights into the effect of surface texture on the development of diffuse and localized necking during forming.

2:40 pm

VERIFICATION STUDY ON THE DENSITY DISTRIBUTION PREDICTIONS OF A POWDER COMPACTION MODEL: A. Casagranda, Concurrent Technologies Corporation, Johnstown, PA 15904

A combined experimental/computational study was carried out to verify the predictions of a powder compaction simulation program. The compaction experiments were performed with a 316L stainless steel metal powder on a fully instrumental production press. The operating conditions were systematically varied to produce controlled variations in local density gradients within the compacts. The tooling loads and displacements were also monitored. Several methods were employed to measure density gradients to ensure accuracy. The powder compaction simulation was then used to predict the density gradients and tooling loads from the measured press displacements. A summary of the experimental results and comparisons to the model predictions will be presented. This work was conducted by the National Center for Excellence in Metal-working Technology, operated by Concurrent Technologies Corporation under Contract No. N00140-92-C-BC49 to the U.S. Navy as part of the U.S. Navy Manufacturing Technology Program.

3:00 pm

THE PROPERTIES AND STRUCTURE OF Al-TiC COMPOSITES: Hongping Dong, Xianglin Dong, Ferrous Metal Research Institute, Section 9, China Science Academy, 72 Wenhua Road, Shenghei District, Shenyang, China

Abstract was not available.

3:20 pm

MODELING AND VERIFICATION OF LASER CUTTING AND LINKING TECHNOLOGY IN MICROELECTRONIC DEVICES: Ampere A. Tseng, Guo-Xiang Wang, Arizona State University, Tempe, AZ 85287-6106

A series of in-situ Al-TiC composites has been developed based on ingot metallurgy and rapid solidification technology. By optimizing material composition, processing parameters and systematic microstructural analyses, this series of in-situ composites has high Young's modulus and high tensile strength both at room and high temperatures. Based on the experimental analysis, a thermodynamic criterion for in situ synthesized TiC in the Al melt and a mathematical model for computing the TiC in-situ synthesized process have been established. A theoretical basis for designing the composites has been provided based on the relations between the TiC particles (size, distribution and volume fraction), material composition and processing parameters. Based on the experimental results and comparative nucleation dynamics it has been found that newly formed TiC dispersoids could act as nuclei for a-Al following rapid solidification. The mathematical relationship between the volume fraction of TiC particles and a-Al grain size has been established.

3:40 pm

A COMPUTER SIMULATION OF FLOTATION TREATMENT PROCESS FOR MOLTEN ALUMINUM: M. Maniruzzaman, M.M. Markhlouf, Aluminum Casting Research Laboratory, Department of Mechanical Engineering, WPI Worchester, MA 01609

The quality of finished aluminum products largely depends on melt treatment prior to casting. One of the widely used treatment processes in the aluminum casting industry is flotation of aluminum inclusions using gas purging. To optimize this process it is very important to understand the basic mechanisms underlying this process. During flotation, flow behavior in the melt reactor is very complex, mainly due to turbulence in the flow field. Unfortunately, with the available equipment, it is not possible to visualize the flow pattern inside the melt reactor. In this study, a computer simulation model for flotation treatment process has been developed based on turbulent flow field calculations. Predicted inclusion trajectories and streamlines along with the analysis of governing parameters will be presented.

4:00 pm

ACTIVITY COEFFICIENT OF INFINITE DILUTE SOLUTION AND INTERACTION PARAMETER IN METALLIC MELTS: Xueyong Ding, Pong Fan, Wenzhong Wang, Qiyong Han, Department of Ferrous Metallurgy, Northeastern University, Shenyang 110006, China

The models for calculating the activity coefficient at infinite dilution and interaction parameters in metallic melts were established. The values from the models are in accordance with those from the experiments on the whole, the ratio of same sign of data between calculation and experiment reaches 95.7% and over 80% for the activity coefficient at infinite dilution, and interaction parameters in liquid Fe-base alloys at 1873K respectively. The results reveal that the higher the reliability of experimental data, the more the ratio of same sign. The values between the models and experiments are in same quantity order in general.

4:20 pm

QUANTITATIVE CHARACTERIZATION AND MODELING OF SPATIAL ARRANGEMENT OF FIBERS IN COMPOSITE: Sichen Yang, Arun M. Gokhale, School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, GA 30332

In the unidirectional fiber reinforced composites, the spatial arrangement of fibers is often non-uniform. These non-uniformities are related to the processing conditions, and the composite properties are in turn affected by the non-uniformities. In this paper, digital image analysis is used to quantify the non-uniform spatial arrangement of Nicalon fibers in a glass ceramic matrix composite (CMC). The quantitative data are utilized to develop a computer simulated microstructure model that is statistically equivalent to the microstructure of the CMC. The simulated microstructure model can be modified according to the variation of the processing conditions to reflect the real microstructure of the composite. Further, the simulated model can be used as an input to predict the mechanical behaviors of the composites as a representative volume element in numerical method, such as finite element analysis.

PROCESSING TECHNOLOGIES: General Abstract Session

Session Chair: Eric Rodeghiero, Cornell University, Ithaca, NY 14853

INVESTIGATION OF PROCESSING PARAMETERS FOR LASER DIELESS SUPERPLASTIC FORMING: N. Hu, C.W. Chen, T.R. Bieler, R. Averill, K. Mukherjee, High Energy Processing Laboratory, Dept. of Materials Science and Mech., Michigan State University, East Lansing, MI 48824

The feasibility of the Laser Dieless Superplastic Forming has been demonstrated. This novel laser processing technique employs a CO₂ laser generator and a numerically controlled x-y moving table, for superplastic forming of an aluminum alloy sheet mounted on the top of a vacuum chamber. Some processing parameters such as the laser power, laser spot diameter, scanning speed and cooling conditions have been investigated. These parameters have critical influence on the temperature distribution of the deformed area, the total strain and the strain rate. An evaluation of the microstructure and the thickness variance in cross section of the specimen and an analysis of the stress distribution and heat flow appropriate to the process will be presented.

2:20 pm

MATHEMATICAL MODELING OF LASER BEAM MACHINING AND DUAL LASER BEAM WELDING OF METAL MATRIX COMPOSITE: S. Kudapa, V. Barnekov, K. Mukherjee, Department of Materials Science and Mechanics, Michigan State University, East Lansing, MI 48824

A simplified analytical model of laser beam machining will be presented to determine the maximum depth of cut for a given traverse speed and laser power. The correlation between the analytical and experimental results will be discussed. Variation in the absorptivity of the material to the laser beam was also accounted in the mathematical model. The collinear dual laser beam welding (CD-LBW) technique pioneered by High Energy Laser Material Processing Laboratory at MSU, was also modeled using a commercial FEM package and the results were compared with the experimental values.

2:40 pm

SYNTHESIS AND CHARACTERIZATION OF RAPIDLY SOLDIFIED POWDERS OF A HYPOEUTECTIC CAST IRON: S.N. Ojha, Dept. of Metallurgical Engineering, Banaras Hindu University, Varanasi-221005, INDIA; Presently at Dept. of Chemical Engineering, SH470, Cleveland State University, Cleveland, OH 44115

A confined gas atomization process is described to synthesize rapidly solidified powders of a hypoeutectic cast iron melt. The process basically utilizes the melt-gas interaction at the tip of a flow tube concentric to an annular gas flow channel to promote an efficient atomization of the melt. The effect of gas-metal flow ratio on the size and size distribution of powder particles are discussed. The results of X-ray diffraction analysis are used to show a large volume fraction of retained austenite in different size range of powder particles indicating considerable departure from their equilibrium solidification conditions. The microstructural features are presented to show a typical cellular and dendritic morphology of the primary austenite phase depending on the size of powder particles. The cooling rates estimated from the microstructral features and also from the heat flow model are presented and compared to show that these are well within rapid solidification regime for a wide size range of powder particles.

3:00 pm BREAK

3:10 pm

CONTROL OF SILICON CARBIDE/METAL REACTIONS: J.S. Park, K. Landry, J.H. Perepezko, Dept. of Materials Science and Engineering, University of Wisconsin-Madison, 1509 University Ave., Madison, WI 53706

The diffusion pathway and kinetics governing the reaction between SiC/metal combinations have been examined to identify systematic behavior. With respect to the metal components, two separate reaction modes were identified formation of carbides and development of silicides or formation of silicides and free carbon (periodic morphology). In each case, the diffusion pathway is dictated by the formation of carbon or carbides, and mass balance requirements. The analysis of the separate reaction modes was confirmed by experiments of SiC/Ni, SiC/Cu/Ni and SiC/Cr/Ni reactions. For the SiC/Ni reaction at 1123K, the chemical potential of Ni and Si decreased but that of carbon increased and decreased along the reaction path with formation of silicides and free carbon. The analysis method offers general guideline for the control of in-situ composite synthesis reactions The support of ONR (N0001492-J-1554) is gratefully acknowledged.

3:30 pm

IMPROVEMENT OF CYCLIC OXIDATION RESISTANCE OF CARBON-CARBON COMPOSITES BY Si IMPREGNATION: Y. Sato, S. Ohtani, Y.-C. Zhu, N. Iwamoto, Ion Engineering Research Ins. Corp., 2-8-1 Tuda-Yamate, Hirakata, Osaka, 573-01 Japan

To improve high temperature oxidation resistance of carbon-carbon composites following surface treatment was performed before CVD SiC coating. The surface of C/C composites were exposed by SiO gas for forming thin SiC layer. Then Si impregnation at high temperature was performed. The pores existing at the surface region of C/C composites were almost infiltrated with SiC. SIG-C composite zone and furthermore smooth surface were formed.

The surface modified C/C composites were SiC coated by using conventional CVD method and then cyclic oxidation tests at 1500°C in air or in methane-combusted atmosphere were performed. SiC coated C/C composites after Si impregnation showed improved oxidation resistance although C/C composites directly covered with CVD SiC showed a greater mass loss.

3:50 pm

NONDESTRUCTIVE EVALUATION OF MECHANICAL AND ENVIRONMENTAL DAMAGE IN A CONTINOUS SWIRL GLASS REINFORCED POLYURETHANE COMPOSITE: H.M. Herring, D.C. Worley II, R.S. Benson, P.K. Liaw, Department of Materials Science and Engineering, University of Tennessee, Knoxville, TN 37996; J.M. Corum, W.A. Simpson, Jr., Oak Ridge National Laboratory, Oak Ridge, TN 37831

This paper describes research into the methods by which a glass reinforced polymer composite may be examined nondestructively for fatigue and creep damage. Glass fiber reinforced polyurethane composite samples were mechanically tested in standard conditions and after exposure to two distinct environments, and studied using various nondestructive test methods. Specifically, continuous swirl glass mat (CSM) polyurethane samples were fatigue and creep tested in a standard environment and after being subjected to distilled water and windshield washer fluid. Virgin specimens were examined prior to testing, using C-scan ultrasonics and light-transmission microscopy (LTM). After mechanical testing, the samples were again examined with C-scan ultrasonics and LTM to document the damage sustained. Additionally, any specimens that failed during the fatigue or creep tests were further examined by scanning electron microscopy (SEM) to evaluate damage on any existing fracture surfaces. Subsequent analysis of the C-scans and LTN images displayed a general tendency toward increased attenuation in the post-teat samples, believed to be the result of sustained damage. In the C-scans, changes in the increase in attenuation of ultrasonic waves appeared to be more attributable to the different environments. Increased attenuation of transmitted light in the LTM images, however seems to be due more to mechanical damage, such as fiber/matrix interfacial debonding. Selected C-scans, LTM, and SEM images are presented, and a comparison is made of the images taken prior to mechanical testing and those taken afterwards. Some advantages and disadvantages of using these techniques to determine damage in this composite are discussed.

TESTING AND INSPECTION TECHNIQUES FOR STRUCTURAL INTEGRITY EVALUATION AND POWER PLANT LIFE EXTENSION: Session II: Structural Metals--Radiation and Thermal Effects

Session Chairs: Mr. Fahmy M. Haggag, Advanced Technology Corporation, 661 Emory Valley Road, Suite A, Oak Ridge TN 37830; Prof. In Sup Kim, Department of Nuclear Engineering, Korea Advanced Institute of Science and Technology, S. Korea

EFFECT OF LONG-TERM THERMAL AGING ON THE FRACTURE TOUGHNESS OF AUSTENITIC STAINLESS STEELS: H. Huang, ICF Kaiser Hanford Company, MSIN X3-85, Richland, WA 99352

Austenitic stainless steels (SS) are know to exhibit adequate strength, excellent resistance to sodium, and high fracture toughness. Because of these advantages, the steels are used in the components of nuclear reactors. Compact tension specimens taken from Fast Flux Test Facility (FFTF) primary piping materials (Type 316 stainless steel and 16-8-2 SS weld metal) and from reactor vessel materials (304 SS and 308 SS weld metal) were heated in laboratory furnaces for 100,000 hours. Fracture toughness testing was performed on these specimens, which are 7.62- and 25.4-mm thick, respectively, at the aging temperature (482 and 427C). Results were analyzed with the multiple-specimen method. Thermal aging continues to reduce the fracture toughness of FFTF component materials. Results show that thermal aging has a strong effect on the toughness degradation of weld metals, particularly for 16-8-2 SS weld whose aged/unaged Jc ratio is only 0.31 after 100,000-hour aging. The fracture toughness of the 308 and 16-8-2 SS weld metals fluctuated during 20,000 to 50,000-hour aging but deteriorated as the aging time increased to 100,000 hours. The need to consider fracture control based on a fracture mechanics approach in design and safety analyses for operation exceeding 100,000 hours is discussed.

2:20 pm

SMALL SPECIMEN TECHNIQUES USED TO MEASURE THE EFFECT OF IRRADIATION ON THE MECHANICAL PROPERTIES OF STRUCTURAL MATERIALS FOR THE ACCELERATOR PRODUCTION OF TRITIUM PROJECT: A. Maloy, W.F. Sommer, MST-4, Los Alamos National Laboratory, Los Alamos, NM 87545; D.J. Alexander, K. Farrell, R. Klueh and M.L. Grossbeck, Oak Ridge National Laboratory, Oak Ridge, TN 37831; M.L. Hamilton, Pacific Northwest Laboratories, Richland, WA 99352

For determining the lifetime of target/blanket components for the accelerator production of tritium project, the mechanical properties are being measured after irradiation in prototypic proton and neutron fluxes produced using a 800 MeV, 1mA Gaussian proton beam (where 2 sigma is 3 cm). Small specimen techniques must be used to obtain a uniform fluence on the specimens at a prototypic temperature. Thus, small-scale specimens are being used to measure tensile, bend, creep, fracture toughness, and stress corrosion cracking properties after irradiation. The properties measured with these small-scale specimens will be compared with those measured with larger specimens.

2:40 pm

USE OF THE ABI TECHNIQUE TO ASSESS RADIATION EMBRITTLEMENT AND QUANTIFY TOUGHNESS RECOVERY FOLLOWING THERMAL ANNEALING OF NUCLEAR PRESSURE VESSELS: Fahmy M. Haggag, Advanced Technology Corporation, 661 Emory Valley Road, Suite A, Oak Ridge TN 37830; K. Linga Murty, North Carolina State University, Raleigh NC 27695-7909

Abstract not available.

2:55 pm

SMALL PUNCH AND TEM DISC TESTING TECHNIQUES AND THEIR APPLICATION TO CHARACTERIZATION OF RADIATION EMBRITTLEMENT: J. Kameda, Ames Laboratory, Iowa State University, Ames, IA 50011; X. Mao, Department of Mechanical Engineering, The University of Calgary, Calgary, Alberta, Canada T2N 1N4

The present paper summarizes the development of miniaturized small punch (SP) and TEM disk testing techniques and shows their applicability in strength, difficulty and fracture toughness were empirically estimated by analyzing the deformation and fracture properties observe in miniaturized specimen tests. A correlation between the ductile-brittle temperature (DBTT) determined from the static SP and dynamic Charpy V-notched specimens tests has been theoretically and experimentally presented. The problems of cracking detection and data scattering often observed in the miniaturized specimen tests are discussed in terms of heterogeneous embattlement behavior of materials. It has been shown that miniaturized testing miniaturized testing techniques have capability of evaluating changes in the mechanical properties of ferritic and vanadium alloys caused by neutron irradiation and post-irradiation annealing. This work was supported by USDOE, Office of Basic Energy Sciences, Div. of Materials Science under contract no. W-7405-ENG-82.

3:15 pm BREAK

3:25 pm

A NOVEL TECHNIQUE TO MEASURE THE MATRIX CARBON COMPOSITION WITH AGEING OF 2.25Cr-1Mo BOILER STEEL: C. Orchard, B.J. Diak, S. Saimoto, Department of Materials and Metallurgical Engineering, Queen's University, Kingston, Ontario, Canada K7L 3N6

The classic work of Baker and Nutting developed a carbide precipitation map using time and temperature during aging of quenched 2.25Cr-1Mo-0.15C steel by the examination of precipitates in the transmission electron microscope (TEM). Since that time, analogous studies using higher resolution analytical TEM techniques have conformed this gradual transformation of carbides towards the stable M6C. In this study, a similar alloy was quenched and the thermodynamic response during tensile strain rate change tests were correlated with the aging time at specific temperatures. The strain rate sensitivity tests performed at 200C showed a solute drag contribution which could be attributed increasing when the M6C regime was approached. The calibration and sensitivity of this technique will be illustrated.

3:45 pm

EFFECT OF SPECIMEN THICKNESS ON NEAR-THRESHOLD FATIGUE CRACK PROPAGATION OF SA106 GR.C NUCLEAR MAIN STEAM LINE PIPE WELD JOINTS: E.S. Kim, I.S. Kim, Department of Nuclear Engineering, Korea Advanced Institute of Science and Technology, 373-9, Mabuk-ri, Kusong-gu, Taejon 305-701, Korea

Characteristics of near-threshold fatigue crack propagation have been examined in SA106 Gr.C nuclear main steam line pipe and its weld to evaluate the effect of specimen thickness. Tests were performed for load ratios of 0.1 and 0.5 at room temperature in ambient air. Near-threshold fatigue crack growth rates decreased and threshold values increased with increasing specimen thickness. The proposed concept of stress state and microstructure influence on crack closure explains the effect of specimen thickness on near-threshold fatigue crack propagation behavior.

4:05 pm

POST-IRRADIATION ANNEALING OF MICROSTRUCTURAL AND MICROCHEMICAL CHANGES IN PROTON IRRADIATED 304L STAINLESS STEEL: J.T. Busby, J. Gan, G.S. Was, Department of Nuclear Engineering and Radiological Sciences, University of Michigan, Ann Arbor, MI 48109-2104

Abstract not available.