Room: 315B
Session Chairperson: TBA
1:30 pm
ELECTRODEPOSITED Ni-Al PARTICLE COMPOSITE COATINGS: D.F. Susan, K. Barmak, A.R. Marder, Department of Materials Science and Engineering, Lehigh University, Bethlehem, PA 18015-3195
Electrodeposited metal matrix/metal particle composite coatings (EMMC's) were deposited, heat treated, and characterized. Nickel matrix/aluminum particulate coatings deposited on a nickel substrate was chosen as a model system. The microstructure of the as-plated and heat treated coatings was characterized using light optical microscopy (LOOM), SEM, quantitative image analysis (QIA), and EPMA. A schematic model is presented for coating morphological development during codeposition of small, electrically conducting particles. The heat treated coatings consist of a two phase mixture of y solid solution and Ni3Al (gamma-prime) with small Kirkendall voids also present. The hardness of the heat treated coatings increased with increasing Ni3Al content and remained constant with extended heat treatment time. Oxidation tests were performed isothermally in air at 800, 900, and 1000°C for up to 900 hrs. and the depth of the oxidation attack was determined. Throughout the investigation, the Ni-Al coatings were compared to conventional pure Ni coatings deposited under similar conditions.
1:50 pm
FORMATION OF THIN COLORED FILM ON STAINLESS STEEL: G.S.Gupta, Dept. of Mining, Minerals and Materials Engineering, The University of Queensland, St.Lucia, Queensland - 4072, Australia
Formation of colored film on type 304 stainless steel has been studied using chromic acid and potassium dichromate solutions. Permanent colors have been obtained using immersion and electrolytic (potentiostatic) methods. The effect of different parameters on the coloring process including temperature, coloring methods, solution composition, stirring, surfactant etc., have been noted. Moderate temperature of the solution, electrolytic method, use of surfactant and stirring of the solution were found to enhance the coloring process. A coloring solution which is strongly oxidized and acidic in nature is recommended. It is advisable that after the extended use of the coloring solution, fresh solution should be used.
2:10 pm
TRICATIONIC PHOSPHATE COATING ON GALVANNEALED STEEL SHEET: Carlos Nelson Elias, Mila Marques Justo, Escola de Engenharia Metaurgica de Volta Redonda, Av dos Trabalhadores 420, 27260 740 Volta Redonda, RJ - Brazil
The galvannealed compared to galvanized steel sheet has excellent paint adhesion, weldability and corrosion resistance after painting. However, the coating tends to cause exfoliation when undergoing severe deformation. This coating is used in the automobile industry. Since the mass loss is very large, it accumulates on the metal mold and results in severe damage to the surface quality of the formed panel affecting press workability and damaging the tool. Nevertheless, phosphating is the most widely used pretreatment, since it serves as an excellent paint base, increases corrosion resistance to metal paints, aids in cold forming of steel and increases the lubricant power of oil applied to the pieces subjected to friction and wear. The influence of galvannealed morphology on the phosphate crystal feature is not well understood. In this work, a tricationic phosphate coating was applied to a surface of a low carbon steel and an interstitial free steel with galvannealed coating. Our paper will discuss the structural and mechanical properties resulting from this treatment.
2:30 pm
PROGRESS TOWARD THE DEVELOPMENT OF DENDRIMER-BASED PROTECTIVE COATING: Larry A. Milco, Donald A Tomalia, Michigan Molecular Institute, 1910 W. St. Andrews Road, Midland, MI 48640-2696
There is a considerable demand for thin film coatings to which other substances will not stick. Coatings are required to protect surfaces from soils, stains, ice, graffiti, insects, oils, corrosion and chemical and biological contaminants. Moreover, as legislative pressures to remove volatile organic solvents from coating formulations persist, the demand for alternative water-borne systems continues to increase. The incorporation of dendritic polymers in non-stick coating formulation is a highly viable approach to overcoming the limitations associated with conventional polymer systems. Non-stick coatings were prepared by crosslinking dendrimers with linear polymers. The high advancing and receding contact angles and low contact angle hysteresis of water and hexadecane on the coatings are indicative of the non-stick (low adhesion) properties of these surfaces.
3:05 pm
PULSED LASER DEPOSITION OF COLLAGEN AND APATITE/COLLAGEN COMPOSITE BIOCOMPATIBLE THIN FILMS: J.A. Conklin, C.M. Cotell, Surface Modification Branch, Code 6671, U.S. Naval Research Laboratory, Washington, D.C. 20375
Thin films comprising either collagen or a composite of hydroxyapatite (HA) and collagen were deposited by pulsed laser deposition (PLD) on various substrates. Films were deposited using either a KrF (248nm) or ArF (193nm) excimer laser and the substrates were maintained at room temperature. Collagen films were characterized by SEM, Fourier transform infrared (FT-IR) spectroscopy and gel electrophoresis. The functional groups on the collagen molecule remained intact through the laser ablation process, but there was evidence for some loss of secondary structure. In addition, the surface morphology was a function of the laser fluence and the gas environment during PLD. Less denaturation of the molecule was observed when the ablation was at 193nm than at 248nm. For the composite films, the target material consisted of a 70/30 (by weight) mixture of apatite/collagen, the same ratio found in bone. The resulting composite films were a homogenous mixture of amorphous collagen and highly crystalline apatite.
3:40 pm BREAK
4:00 pm
PROTEINS FROM OYSTER SHELL AS MODELS FOR BIODEGRADABLE COMMERCIAL POLYMERS: A.P. Wheeler, Department of Biological Sciences, Clemson University, Clemson, SC 29634
Virtually all minerals formed by organisms are composites, containing an organic phase (matrix) which enhances the mechanical properties of the inorganic phase and which presumably controls mineral growth. The matrix of the CaCO3 shell of oysters contains soluble, highly anionic proteins which in vitro adsorb to mineral and in so doing alter mineral growth. Based on the surface reactive properties of these proteins, synthetic analogs have been made which have properties that make them candidates as inhibitors for mineral scale and corrosion and as dispersants. Synthesis of a simple matrix protein analog, polyaspartic acid, is possible on a significant scale using dry thermal polymerization and aspartic acid as the only reactant. Thermal polyaspartic acid is highly biodegradable when compared to polymers currently used for commercial surface reactive applications.
4:20 pm
THE APPLICATION AND SUBSEQUENT CROSS-LINKING OF A BIOLOGICAL POLYMER ON ALUMINUM METAL: Douglas C. Hansen, Naval Research Laboratory, Washington, D.C. 20375-5343
Biological polymers that are used by organisms in the formation of structural materials and protective coatings are widespread in nature. An example of such a polymer is the protein utilized by the blue mussel Mytilus edulis as a varnish coating on its byssal threads. This protein is of high molecular weight (>100,000) and is unique in containing the catecholic amino acid L-dopa. This protein exhibits high affinity for metals and acts as a substrate for the enzyme catechol oxidase, which catalyzes the two electron oxidation of catechol to o-quinone. This reaction is essential in the formation of a two component sclerotized resin. The adsorption of this protein onto pure aluminum (99.99999%) and subsequent cross-linking of the adsorbed protein results in a significant shift in the critical pitting potential and polarization resistance of the treated aluminum. Atomic Force Microscopy (AFM) measurements clearly show that treatment of the protein adsorbed onto Highly Ordered Pyrolytic Graphite (HOPG) with catechol oxidase results in a stable, varnish-like coating.
4:40 pm
ADSORPTION OF ADHESIVE PROTEINS FROM THE MARINE MUSSEL, MYTILUS EDULIS, ON POLYMER FILMS IN THE HYDRATED STATE USING ANGLE DEPENDENT XPS AND AFM: A.M. Baty, P.K. Leavitt, C.A. Siedlecki, B. J.Tyler, P.A.Suci, R.E. Marchant, G.G.Geesey. Center for Biofilm Engineering, Montana State University, Bozeman, MT 59717-0398
The adsorption of Mussel Adhesive Protein (MAP) from the marine mussel Mytilus edulis has been investigated on polystyrene (PS) and poly (octadecyl methacrylate) (POMA) surfaces using angle dependent XPS and AFM. XPS analysis was performed at LN temperatures without dehydrating the samples and at room temperature following dehydration. The analyses indicate that adsorbed MAP is stabilized on the surface of the PS through specific interactions preventing the protein from losing lateral spatial distribution across the surface upon dehydration. The adsorbed MAP on the POMA surface is representative of a loosely bound protein layer that is adsorbed through non-specific types of interactions allowing the protein to lose much of its lateral distribution when dehydrated. These data demonstrate that the surface chemistry of the polymer films influences protein-protein and protein-surface interactions, and that hydration plays a significant role in structuring protein-surface interactions.
Room: 340B
Session Chairpersons: Prof. Enrique J. Lavernia, Department of Mechanical and Aerospace Engineering, University of California, Irvine, CA 92717; Dr. Benji Maruyama, Wright Laboratory/NIST, 2230 10th ST STE 1, WPAFB, OH 45433
2:00 pm INVITED
PRESSURE INFILTRATION TECHNIQUE FOR SYNTHESIS OF ALUMINUM-FLY ASH PARTICULATE COMPOSITES: P.K. Rohatgi, R Q. Guo, H. Iksan, R. Asthana, Materials Department, University of Wisconsin, Milwaukee, WI 53201
Aluminum - fly ash composite was prepared by pressure infiltration technique. Loosely packed beds of cenosphere fly ash (above 55 vol% in the composite) can be successfully infiltrated by molten aluminum under pressure in the range of 0.3-0.7 MPa, and the density of the composite is 1.4 g /cm3 compared to the density 2.68g/cm3 for aluminum. Cenosphere fly ash particles are very light materials (density: 0.4 - 0.6 g/cm3). The microstructure of composite showed that there was uniform distribution of fly ash particles in the aluminum matrix. The mechanism of pressure infiltration and the reaction between aluminum and fly ash during infiltration and solidification will be discussed. Selected properties including hardness, microhardness, and compressive strength of aluminum - fly ash composites were determined and the results are presented in this paper.
2:30 pm
WETTABILITY OF NICKEL COATED GRAPHITE BY ALUMINUM: S. W. Ip, R. Sridhar, J.M. Toguri, Department of Metallurgy and Materials Science, University of Toronto, Toronto, ONT, Canada M5S 3E4; T. Stephenson, INCO Limited, J. Roy Gordon Research Laboratory, 2060 Flavelle Boulevard, Mississauga, ONT, Canada L5K 1Z9
Graphite is a very attractive candidate for aluminum MMC's. However, aluminum does not wet graphite. Thus the production of aluminum-graphite composite is difficult. Nickel is a material known to be wetted by aluminum. INCO Limited recently has developed a novel technique for coating graphite fibres with nickel. It was found that such coating promotes aluminum wetting on the fibres. To determine the wettability of aluminum on nickel coated graphite, the sessile drop technique along with a high temperature x-ray setup was used. Samples of graphite, electrolytic nickel, and nickel coated graphite were examined. Contact angles determined for these samples showed that graphite is non-wetting while the nickel coated graphite provided better wetting than electrolytic nickel. The nickel-aluminum interface of the samples were examined microscopically using SEM.
2:55 pm
MICROSTRUCTURAL INVESTIGATION OF INERT GAS ATOMIZED SiC/Al-Li-Cu-Mg-Zr ALLOY BASED MMC POWDERS: S. Özbilen, Gazi University, Faculty of Technical Education, Department of Metals Education, Teknikokullar, Arkara, Turkey
Variable amount SiC short fibres hardened & Al-Li-Cu-Mg-Zr alloy based MMC powders were produced by inert gas atomization under Ar, in a pilot plant down-draught atomizer with a Mannessman type nozzle. Melt temperature was 875°C and gas pressure was 1.85 MPa. Powder based alloy matrix CM was sieved dry. SEM and TEM investigation was used for microstructural characterization. Emphasis was given to the effect of PM processing route on the nature of bonding between matrix and second phase fiber particles.
3:20 pm
IN SITU PROCESSING OF TiB2/Cu-ALLOY COMPOSITES BY REACTIVE PRESSURELESS INFILTRATION (RPI): V. Shtessel, M. Koczak (deceased), Department of Materials Engineering; R. Mutharasan, Department of Chemical Engineering, Drexel University, Philadelphia, PA 19104
The process of Reactive Pressureless Infiltration (RPI) for in situ synthesis of Cu-Mn/TiB2 composites from elemental powders has been investigated. Primary factors, which alter phase composition and microstructure of the MMCs are enthalpy and relative free energy of reinforcement formation, initial powder size, processing temperatures and matrix wettability of the reinforcement. The Cu-Mn/TiB2 composite material was synthesized by matrix assisted displacement reactions, where the desired reinforcement forms gradually as a result of liquid infiltration and reaction. A weak boride former, e.g. Mn, is mixed with B in the bottom layer and infiltrated with Cu-Ti alloys. The process involves two consecutive steps: (1) formation of weak boride, e.g. Mn + 2B -> MnB2; (2) displacement reactions, e.g. MnB2 + Ti-Cu -> TiB2 + Cu-Mn. This route is advantageous due to improved wettability of Mn compounds by copper. Composites with 5, 7 10, 15, 20 and 30v/o of TiB2 have been produced and their microstructures and mechanical properties are being evaluated. Typical microstructures illustrate an even distribution of 0.5 - 5 micron particulates of TiB2. Possible mechanisms of the microstructure development during the process of reactive infiltration were analyzed. The study demonstrates that TiB2 forms by interfacial reaction: Ti + MnB2 = TiB2 + Mn. A kinetic model of the process has been proposed. The model enables us to determine the infiltration rate, the thickness of reacted layer as a function of time, the rate of reaction of TiB2 formation, the concentration profile of each component in the reactive layer. The model shows good agreement with experimental results. The RPI process is a promising technique for in situ, net-shape manufacturing of Cu-based MMCs. This research is supported by Office of Naval Research.
3:45 pm BREAK
3:55 pm INVITED
BREAKTHROUGH OF CONTINUOUS FIBRES REINFORCED CERAMIC COMPOSITES FOR INDUSTRIAL APPLICATIONS: M. H. van de Voorde, European Union, Joint Research Center, Institute for Advanced Materials, 1755 ZG Petten, The Netherlands
Long fibre Ceramic Composite Materials form the materials for the technology of the 21st century; this will be demonstrated on the hand of a number of industrial applications. A resume will be given of the ceramic composites used in space applications. An overview of the "inorganic fibres" and "continuous fibres ceramic composite materials" is planned with indications of their advantages and problem areas. This data base generation, and engineering properties as joining, machining and NDE. Developments will be highlighted to made cost competitive ceramic composites with high temperature stability, corrosion resistant and good mechanical properties for long duration. The breakthrough of mew technologies with CFCC's use will be sketched. The needs for Research, Development and Technology on CFCC are pinpointed and indications for a European programme given. The market trends are also resumed.
4:25 pm
PROCESSING OF DAMAGE-TOLERANT, ENVIRONMENTALLY-STABLE, ALL-OXIDE CERAMIC COMPOSITES: C.G. Levi, J.Y. Yang, B.J. Dalgleish, Materials Department, University of California, Santa Barbara, CA 93106
Ceramic composite systems based on all-oxide constituents are of interest for high temperature applications owing to their inherent oxidative stability but must also be designed to exhibit damage tolerance. One microstructural design concept relies on crack deflection through a porous matrix rather than at fiber/matrix interfaces, which is the more conventional approach to toughening in CMC's. The requisite matrix must have an optimum combination of toughness and strength which can be achieved by incorporating a minimum amount of fine, well distributed porosity, and must also be chemically and microstructurally stable at high temperatures. Composites based on this concept have been synthesized using vacuum infiltration of aqueous mullite-alumina slurries into woven polycrystalline alumina fiber preforms, followed by precursor impregnation and sintering. Initial evaluation of these materials shows promising behavior under tension and potential for notch insensitivity and thermal stress tolerance. Current understanding of the underlying mechanisms as well as microstructural design and processing issues relevant to the attainment of this behavior will be discussed. Research sponsored by DARPA under URI Grant N00014-92-J-1808.
4:50 pm
DENSE IN-SITU TiB2/TiN AND TiB2/TiC CMCS: REACTIVE SYNTHESIS AND PROPERTIES: I. Gotman, F. Olevsky, E.Y. Gurmanas, Department of Materials Engineering, Technion-Israel Institute of Technology, Technion City, Haifa 32000, Israel
In-Situ TiB2/TiN and TiB2/TiC CMCs were fabricated from fully dense BN-Ti and B4C-Ti powder blends with and without the addition of Ni powder. Three different methods were used: pressureless and pressure-assisted displacement reaction synthesis and thermal explosion under pressure. Application of a moderate external pressure (¾ 200 MPa) was shown to be sufficient to ensure full density of TiB2/TiN/Ni and TiB2/TiC/Ni composites. The addition of Ni powder allowed to significantly reduce the ignition temperature of thermal explosion due to the formation of the low temperature Ti-Ni eutectic phase. Thus, the preheating temperature or thermal explosion, as well as the processing temperatures of displacement reaction synthesis (¾ 1200°C) were considerably lower than those typical of current methods used for the processing/consolidation of CMCs. Microstructure and composition of materials obtained were characterized by x-ray diffraction and scanning and transmission electron microscopy (SEM and TEM). Mechanical properties were evaluated by measuring microhardness, fracture toughness and three-point bending strength. High fracture toughness of TiB2/TiN/Ni and TiB2/TiC/Ni CMCs was obtained indicating that fine Ni dispersions are effective in dissipating the energy of propagating cracks.
5:10 pm
FIBER FRAGMENTATION DURING PROCESSING OF METALLIC MATRIX COMPOSITES: Nicole M. Gorey, Donald A. Koss, John R. Hellman, Department of Materials Science and Engineering, Penn State University, University Park, PA 16802
Room: 230D
Session Chairman: D.J. Donaldson, Consultant, 18160 Cottonwood Road #158, Sunriver, OR 97707
2:00 pm
LAYERED DOUBLE HYDROXIDES FOR TREATMENT OF BAYER PROCESS LAKE WATER: A.J. Perrotta, Alcoa Technical Center, Alcoa Center, PA 15069-0001, F.S. Williams, Alcoa Alumina & Chemicals, L.L.C., Point Comfort, TX 77978-0101, L. Stonehouse, Alcoa of Australia, Kwinana R & D, Western Australia
Layered double hydroxides (LDHs) or the so-called anionic clays are the analog of the extensively studied family of cationic smectite clay minerals. The successful treatment of Bayer lake waters requires the removal of anions, both inorganic and organic; thus, the utilization of these anionic clays are ideally suited for this process. Insitu crystallization of layered double hydroxides have resulted in the removal of anions from Bayer lake water at two different refineries. Data on the removal of anions and their selectivity for intercalation in the layered double hydroxide indicates that the order of removal is CO3, C2O4, SO4. The monovalent chloride ion is least preferred but also can be extracted at significant levels at the higher level of treatment. The total organic carbon (TOC) is also lowered significantly. Measurements on succinate and malonate, both diacids, show a significant reduction at the higher levels of treatment. Successful regeneration of hydrocalumite show the thermal products of mayenite and calcium oxide which can then be recrystallized in lake water with concomitant further reduction in anion content. Additional experiments using ultrasonic radiation show an enhanced carbonate removal through increased calcite formation.
2:25 pm
TECHNICAL PECULIARITIES AND VIABILITY OF HYDROTHERMAL TREATMENT OF RED MUD: Károly Solymár, Janos Steiner, József Zõldi, ALTAK Consulting Ltd. H-1125 Budapest, Zsolna u.31-33. Hungary
The regeneration of the chemically bound caustic soda and alumina from red mud was studied in laboratory and pilot plant scale between 260 and 320°C as a function of caustic soda concentration and final A/C ratio of the liquor, lime dosage and retention time. The pilot plant tests have been carried out in a tube digester facility of 3.0 m3/h designed by HUNGALU ALUTERV-FKI Ltd. and the test work was performed in cooperation with the team from the Almásfuzitó Alumina Refinery. At 260-320°C iron-rich iron-aluminium-hydrogarnets (Ca-Al-Fe-hydrosilicates) were formed allowing 90-95% Na2O and 20-70% Al2O3 content of red mud to be recovered as a function of the final A/C ratio with a specific lime consumption of 2.5-3.0 kg CaO/kg regenerated NaOH. The tube digester was found as an adequate equipment. The viability of the process highly depends on the price ratio of the caustic soda and burnt lime. The latest development in red mud separation also increased the viability of the hydrothermal treatment of red mud.
2:50 pm
EXPLOITING THE RHEOLOGY OF BAUXITE RESIDUE IN TAILINGS DISPOSAL STRATEGIES: N. Pashias, D.V. Boger, Advanced Mineral Products Research Center and Department of Chemical Engineering, The University of Melbourne, Parkville, Victoria 3052, Australia
This paper demonstrates how an understanding of the complex rheological properties of bauxite residue suspensions can be exploited in the implementation of a dry disposal strategy. Comparative data are presented for bauxite residue generated from three refineries in Australia and from refineries in Jamaica, Suriname, and Brazil. The necessary properties for optimum handling of the residue at high concentrations and for dry stacking and/or slope deposition are established.
3:15 pm BREAK
3:35 pm
MOBILITY ON MUD: Danys Morin, Société d'électrolyse et de Chimie Alcan Ltée, Vaudreuil Works, Jonquiére, Québec
The Vaudreuil Alumina plants use deep thickeners to prepare their red mud for disposal in a 60 ft. high wet stack. Management of this stack requires four-seasons access to the pipework and occasional dust suppression procedures. From peak to foothills the stack can display all the phases of fluidity present in the mud. Vaudreuil has developed a family of vehicles to provide the required access from a 25 tonne amphibian to spread gypsum on the mud to control dusting, to a 1.5 tonne tractor-trailer for personnel transport. This paper describes the units in detail and explains how the characteristics of the mud can be used to give a good load-bearing capacity.
4:00 pm
OPTIMIZATION OF GRINDING OF ALUMINAS IN THE ROTARY-VIBRATION MILL AND THEIR EFFECT ON PHYSICAL PROPERTIES OF MOULDERS: Mariusz A. Wojcik, Tomasz Gajda, University of Mining and Metallurgy, Faculty of Materials Science and Ceramics, av. Mickiewicza 30,A-3, 30-059 Cracow, Poland; Ansgar Luttermann, Fachhochschule Munster, Fachbereich Chemieingenieurwesen Stegerwaldstrasse 39,d-48565 Steinfurt, Germany
The properties of high alumina ceramics are strongly influenced by the properties of the starting raw materials such as the grain size distribution and their morphology. Due to this, the grinding operation is an important step in the process of ceramic mass production. The investigations presented show that compared to other mills, the use of a rotary vibration mill will reduce time of grinding, the energy consumption and yield finer powders. The aim of these investigations was to obtain alumina powders with d50 approximately 3 um and d90 close to 10 um. Two different starting materials were tested. The process of grinding was performed according to various parameters such as filling ratio, amount of grinding media, alumina and water, ball set and frequency. The influence of ratios between the grinding media, alumina and water are shown. The influence of ball size, time and frequencies on the kinetics of grinding are shown. Finally the effect of alumina grinding on the physical properties such as green and fired density and shrinkage are shown and discussed in this paper.
4:25 pm
RESULTS OF TESTS OF THE FINE GRINDING OF ALUMINA IN THE ROTARY-VIBRATION MILL WITH LOW FREQUENCY OF VIBRATION: Jan Sidor, Mariusz A. Wojcik, University of Mining and Metallurgy, Faculty of Materials Science and Ceramics, av. Micklewicza 30, a-3, 30-059 Cracow, Poland
Results of the investigations of the fine grinding of alumina (below 0.5-2µm) were presented in this paper. Tests were carried out in the laboratory rotary-vibration mill with a low frequency of vibration below 16 Hz at two sets of work. In the first set, the mill chamber had both rotary and vibration movements, while in the second one, the mill chamber made only the vibration movement such as in the case of a classic vibration mill. Results show that it is possible to obtain a high purity alumina with low energy consumption in both sets.
Room: 230A
Session Chairperson: Dave Belitskus, Alcoa Technical Center, Alcoa Center, PA 15069
2:00 pm
IMPROVEMENT OF THE ALUMINIUM CELL LINING BEHAVIOUR BY USING SiC: M.M. Ali, A.A. Mostafa, R&D Management, The Aluminium Company of Egypt, Naga Hamadi, Egypt
Now, the finite element analysis techniques (FEA) are widely used in the aluminium reduction cells. For example, to select new lining materials or to better understanding of the aluminium cell behaviour, through the use of modelling concepts. One of the important studies is the thermal model of the aluminium reduction cells. A 2-D FEA model has been established, tested and verified to study the thermal balance of the cell. This paper presents a study for using a new lining material. The effect of this material on the ledge profile and thermal distribution were discussed.
2:25 pm
PROPERTIES OF A COLLOIDAL ALUMINA-BONDED TiB2-COATING ON CATHODE CARBON MATERIALS: H.A. Øye, Institute of Inorganic Chemistry, The Norwegian University of Science and Technology, 7034 Trondheim, Norway; V. de Nora, J.-J. Duruz, G. Johnston, MOLTECH S.A., 9 Route de Troinex, 1227 Carouge, Geneva, Switzerland
Laboratory studies of colloidal alumina-bonded TiB2 have been carried out. The following properties have been demonstrated: Thermal expansion somewhat higher than carbon materials, electrical conductivity of the same order as graphite, porosity of coating 
30%, resistance to sodium is high if the coating is protected by infiltrated aluminium, strong adherence of the coating to carbon materials even after thermal cycling, high abrasion resistance of the coating, wettability of the coating by liquid aluminium, the coating's action as a barrier to sodium penetration. The barrier action to sodium is the ability to form a stable liquid aluminium layer in the pores of the coating which will slow down the reaction: Al(l)+3 NaF(in electrolyte)=3 Na(in C cathode)+ AlF3 (in electrolyte) as the sodium stabilizing C is no longer in close contact with the bath.
2:50 pm
PENETRATION FORCE OF BATH COMPONENTS INTO POT LININGS DESCRIBED BY NONEQUILlBRIUM THERMODYNAMICS: Stig F. Johansson, Skamol a/s, Østergade 58-60, DK-7900 Nykøbing Mors, Denmark; Signe Kjelstrup, The Norwegian University of Science and Technology, Department of Physical Chemistry, N-7034 Trondheim, Norway
From principles of irreversible thermodynamics, a simple expression is developed for the driving force for transport of bath constituents across a non-isothermal matrix into the refractory lining in aluminium electrolysis furnaces. The expression shows that the transport outwards of aggressive substances is favoured by the temperature gradient. Possible applications are demonstrated for 3 penetrating gases with a variety of applied and proposed refractory materials. Common to all results is a possible reduction of the rate of penetration into barriers by a smaller temperature gradient due to the entropies of reaction. This can be accomplished by proper heat insulation below. A set of conclusions regarding the resistance of chemical barriers to penetration by bath components completes the paper.
3:15 pm BREAK
3:35 pm
FLUORIDE ATTACK ON ALUMINO-SILICATE REFRACTORIES IN ALUMINIUM ELECTROLYSIS CELLS: Jørn Rutlin, Tor Grande, Department of Inorganic Chemistry, Norwegian University of Science and Technology, N-7034 Trondheim, Norway
The phase relations in the system sodium fluoride - mullite are relevant to chemical reactions that take place in fireclay based refractory materials used in the bottom lining of aluminium electrolysis cells. The solid-liquid phase relations have been investigated by means of differential thermal analysis, powder X-ray diffraction analysis and microscopy. At sub solidus temperatures the four compounds sodium fluoride, cryolite, nepheline and 
-alumina have been found at excess amount of sodium fluoride. At lower concentration of sodium fluoride, corundum is formed at the expense of -alumina. The solidus temperature was observed at 857±5°C which corresponds well with results in systems with higher aluminium fluoride content. An apparent eutectic was observed at 12±0.5 mole% mullite. Nepheline and cryolite are the two main components of the melt crystallizing at the solidus temperature. The liquidus temperature at the solubility limit was observed at 943°C and at 7.5±0.5 mole% mullite. The present results are discussed in relation to the deterioration mechanism of alumino-silicate refractories.
4:00 pm
PYROHYDROLYSIS OF SPENT POTLINING: Vladimir Blinov, Tor Grande, Harald A. Øye, Department of Inorganic Chemistry, Norwegian University of Science and Technology, N-7034 Trondheim, Norway
Due to the toxic nature of spent potlining (SPL) this is becoming one of the major environmental concerns for the aluminium industry today. SPL also represents a major recovery potential because of its fluoride and energy content. The Elkem SPL recycling process is one of several technically feasible alternatives for treatment and recycling of SPL. In this process the fluorides can potentially be recovered by a pyrohydrolysis of the oxyfluoride silicate slag from the Elkem process. The thermodynamic aspects of this pyrohydrolysis are discussed in the present work. Chemical activity of fluorides in actual oxyfluoride silicate melts has been determined by means of high temperature mass spectroscopy and the Knudsen effusion method. Only sodium fluoride was observed as volatile species for dry melts. The chemical activity of sodium fluoride in several melts could therefore be determined by the Knudsen effusion method. Together with data on the activity of sodium oxide, the present activity data of sodium fluoride enable a calculation of the equilibrium pressure of HF during pyrohydrolysis of SPL.
4:25 pm
IDENTIFICATION OF NONLINEAR SWELLING PRESSURE DISTRIBUTION OF ALUMINUM REDUCTION CELL: H.S. Sayed, Structural Eng. Dept., Cairo University, Egypt; M.M. Megahed, Mechanical Eng. and Prod. Dept., Cairo University, Egypt; F. Dawi, S. Abdella, R&E Dept., EgyptAlum Co, Nagi Hammadi, Egypt
The swelling pressure has the most significant effect on determining the cathode life among other loads applied on the steel casing of the aluminium reduction cell. In previous research, an identification technique is applied to determine the average swelling pressure value that is exerted on the side wall of the cell. The identification technique made use of the measured deformation of the cathode steel casing at different elapsed times (after 88 and 615 days) on an existing cathode that belongs to EgyptAlum. A new set of measurements has been conducted at different elapsed time (after 1055 days). This new set of measurements has been conducted at different elapsed times (after 1055 days). This new set of measurements gives a unique opportunity to reassess the identified swelling parameters. On the other hand, the distribution of the swelling pressure along the cathode interface with the vertical steel walls of the cell, has a significant effect on the performance of the steel casing. In this paper, the distribution of the swelling pressure, the interface of the carbon blocks and the steel casing, are identified. A nonlinear elasto-plastic model for the steel casing using combination of shell and beam elements is developed. The carbon blocks are modeled using solid and truss elements. The model utilizes Dewing model that describes the swelling pressure as a function of the elapsed time and confined swelling strain rate to determine the swelling pressure distribution. The swelling pressure is initiated using a predetermined free and confined swelling rate. As a nonlinear analysis progress with item, the confined swelling rate of the previous time step is used to assess the swelling pressure at the interface for the subsequent steps. The swelling pressure distribution is then determined at any elapsed time.
4:50 pm
MODELLING OF DYNAMIC LEDGE HEAT TRANSFER: Chuck C. Wei, John J.J. Chen, Barry J. Welch, The University of Auckland, Chemical & Materials Engineering Dept., Auckland, New Zealand; Vaughan R. Voller, University of Minnesota, Dept. of Civil & Mineral Engineering, Minneapolis, USA; M.P. Taylor, New Zealand Aluminium Smelters Company Ltd., Invercargill, New Zealand
Cell disturbances such as anode effect and process operations such as feeding, anode changing and metal-tapping often cause variations in the heat balance of a reduction cell, resulting in deviations from the optimum operating conditions. A dynamic ledge heat transfer model built on an earlier one-dimensional dynamic simulation by the authors and based on the finite difference method was used to solve the transient heat conduction with phase change in the sidewall/ledge region. Fixed-grid and deforming-grid spacings were respectively superimposed on the sidewall and ledge region in order to track the moving front. Various aspects of the process dynamics with respect to the variation of ledge thickness and sidewall shell temperature were considered.
Room: 232A
Session Chairs: J.C. Malas, Wright-Patterson AFB, OH 45433-6533; V.L. Acoff, Dept. of Metallurgical and Materials Engineering, The University of Alabama, Tuscaloosa, AL 35487
2:30 pm
ELECTRONIC PROTOTYPING: TOWARDS VIRTUAL MATERIALS RESEARCH: S.R. LeClair, Wright Laboratory, Materials Directorate, Wright-Patterson AFB, OH 45433-7746; S.L. Thaler, PanAptics, Inc., 12906 Autumn View Dr., St. Louis, MO 63146-4331
The most significant first step toward electronic prototyping (EP) in the research community has been the pursuit of intelligent processing (i.e., closed-loop process control), effectively using in situ sensors to monitor both process parameters (energy input), material behavior (changes in bulk structure, temperature, etc.), and characterize materials at the atomic scale, wherein material properties such as strength, residual stresses, composition, phase, compressibility, modulus, etc., are monitored and controlled in real-time. Of interest are the identity of interrelationships between sensed variables which enable fault detection - discover and monitor, in real-time, an evolving taxonomy (a set of linear and other meaningful relationships) which can be classified as "faults"; process stability analysis - establish a measure of stability, in real-time, by assessing process response/noise in the context of energy input, apparatus actuation, and the control algorithm requisite variety; and response time - establish a measure of performance, in real-time, of "sensor-actuator" limits to affect specific processing requirements. Yet these capabilities are merely near-term - the more interesting, longer term capability is the pursuit of process discovery, i.e., how these relationships and measures might provide insight into explaining phenomena of interest and/or enable the development of a material or process model in situ.
2:55 pm
ACCURATE, RELIABLE CONTROL OF PROCESS GASES BY MASS FLOW CONTROLLERS: J. Hardy, T. McKnight, Instruments and Controls Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-6003
The thermal mass flow controller, or MFC, has become an instrument of choice for the monitoring and controlling of process gas flow throughout the materials processing industry. MFCs are used on CVD processes, etching tools, and furnaces and in the semiconductor industry are used on 70% of the processing tools. Reliability and accuracy are major concerns for the users of MFCs. Calibration and characterization technologies have been developed and implemented for mass flow devices. A test facility is available to industry and universities to test and develop gas flow sensors and controllers and evaluate their performance related to environmental effects, reliability, reproducibility, and accuracy. Additional work has been conducted in the area of accuracy. A gravimetric calibrator was invented that allows flow sensors to be calibrated in corrosive, reactive gases to an accuracy of 0.3% of reading. This paper will present possible sources of error in MFC process gas flow monitoring and control, and will present an overview of corrective measures which may be implemented with MFC use to significantly reduce these sources of error.
3:20 pm
MONOLITHIC MICRO-SPECTROMETER FOR LOW-COST LIQUID AND GASEOUS CONSTITUENT PROCESS MONITORING: S. Rajic, C.M. Egert, Engineering Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-8039
A monolithic miniature spectrometer suitable for a variety of sensing applications including industrial process monitoring has been developed. The device consists of a solid structure with a volume less than 6 cubic centimeters. All optical components of the spectrometer including two aspheric mirrors, a diffration grating, and entrance and exit surfaces are fabricated onto the surface of the structure. All light paths are internally contained within the structure. The result is a small, rugged spectroscopic sensor ideally suited for use as a process monitor. Due to its monolithic nature, the device requires no post-fabrication alignment; nor can it be knocked out of alignment during use. The prototype device discussed here was produced in PMMA by precision diamond turning; however, lower cost manufacturing approaches involving injection molding (using diamond turned molds) are under development to produce an affordable sensor. Due to its rugged monolithic design, small footprint, and low cost, the monolithic micro-spectrometer is ideally suited for distributed process monitoring applications often required for industrial processes. It is expected that specific sensor designs will be required for specific applications so as to maximize the performance and resolution over the operating range. The performance of this device as well as the design tradeoffs necessary to optimize performance for given applications will be discussed in the report.
3:45 pm
AN ULTRASONIC SENSOR FOR HIGH TEMPERATURE MATERIALS PROCESSING: Rollie E. Dutton, Materials Directorate, Wright Lab., WL/MLLM, Bldg 655 Ste 1, Wright-Patterson AFB, OH 45433-7817; David A. Stubbs, University of Dayton Research Institute Structural Integrity Division, 300 College Park, Dayton, OH 45469-0120
A sensor has been developed and tested that is capable of emitting and receiving ultrasonic energy at temperatures exceeding 900°C (1652 f) and pressures above 150 MPa (22,500 psi). The sensor works with standard ultrasonic pulser-receivers and has a demonstrated capability of measuring workpiece deformation during hot isostatic pressing (HIP). Details of the sensor design, performance, and the coupling of ultrasonic to the workpiece are described. Ultrasonic data acquired by the sensor, in situ, during HIP runs are presented.
4:10 pm BREAK
4:20 pm
TWO METHODS FOR EMBEDDING OPTICAL FIBERS IN METAL COMPONENTS: L.J. Talarico*, S.W. Allison1, C.A. Blue**, H.M. Meyer III***, L. Riester**; *Engineering Technology Division, Oak Ridge National Laboratory; **Metals and Ceramics Division, Oak Ridge National Laboratory; ***Development Operations, Y-12 Plant; Oak Ridge, TN 37831
Manufacturing or embedding optical fiber sensors in composite materials such as graphite-epoxy, cementitious, plastic or ceramic components is of growing importance to a variety of smart materials and structures applications. A variety of fiber optic sensors can measure strain, pressure, and temperature by observing changes in an output signal fluorescence spectrum or a shift in output signal peak wavelength. The ability to measure these parameters on the surface of, as well as within, a part could provide experimental data for evaluating models, perhaps identifying some inferior parts on-line, and monitoring operating behavior. Some examples are monitoring cure and assessing thermal damage in polymer matrix composites by measuring changes in the epoxy fluorescence spectrum and monitoring densification during powder pressing. Other potential applications include engine diagnostics, condition based maintenance of aircraft and vehicle components, and monitoring and control in the metal forming and materials processing industries.
4:45 pm
EXAMINATION ON THE USE OF ACOUSTIC EMISSION FOR MONITORING METAL FORGING PROCESS: A STUDY USING SIMULATION TECHNIQUE: W.M. Mullins, TMC, Inc., WL/MLIM, WPAFB, OH 45433-7746; R.D. Irwin, Dept of Electrical Engineering and Computer Science, Ohio University, Athens, OH; J.C. Malas III, S. Venugopal, Materials Process Design, Materials Directorate, Wright Laboratory, WPAFB, OH 45433-7746
Physical models for acoustic emission (AE) are introduced and expressions are derived to predict AE activity from such parameters as applied stress, strain, and strain rate. These models are then incorporated into a visco-plastic finite-element simulation program, and the acoustic emission event rate generated during metal-forming operations are predicted. Simulation results are presented for upsetting operations on a typical C-Mn type steel for various friction and die geometry conditions. The AE model predictions compare well with the available experimental results reported in the literature and demonstrate that AE signatures can be reliably simulated. The signatures predicted for metal forging are fairly insensitive to changes in certain processing parameters. This suggests that AE event rate monitoring may not be well suited for monitoring changes in microstructure and friction during forging operations. Some very preliminary studies of event spectral analysis have been performed which show that AE spectral features may be sufficiently dependent on structure and loading to warrant further examination for application to monitoring metal-forging processes.
5:10 pm
SHEAR MODE EMAT AS A SENSOR FOR COMPOSITE LAMINATE FABRICATION: David K. Hsu, Center for Nondestructive Evaluation, Iowa State University, Ames, IA 50011
The polarization direction of shear mode ultrasound interacts strongly with the fiber direction in a composite laminate. Using conventional piezoelectric ultrasonic transducers, we have shown that cross-polarized shear waves were effective for sensing ply orientation errors and ply sequence anomalies. In this paper, we extend the technique to use electromagnetic acoustic transducers (EMAT), particularly for green (uncured) graphite epoxy composite laminates. Preliminary results will be shown and potential applications for quality control will be discussed.
Room: 231A
Session Chairperson: A. Kasaaian, Elkem Metals Co., P.O. Box 299, Marietta, OH 45750
2:30 pm
RECOVERY OF BISMUTH AND ANTIMONY FROM THE ELECTROLYTE IN COPPER ELECTROREFINING TANKHOUSE: Kouji Ando, Naoyuki Tsuchida, Sumitomo Metal Mining Co., Ltd., Niihama Research Laboratories, Hydrometallurgical Research Center, 3-5-3, Nishibara, Niihama, Ehime 792, Japan
An ion-exchange process for recovery of antimony and bismuth from the copper electrorefining process has been studied on both a bench scale test and a pilot test. Recently, the ion exchange technique for the removal of antimony and bismuth from the electrolyte has been in a practice. In the existing technology, the metals recovered are required to process further. The separation of antimony and bismuth is sometime difficult and costly. Sumitomo Metal Mining has developed a noble process to recover antimony and bismuth metals directly from the eluent by an electrowinning. Furthermore, antimony and bismuth were successfully separated. The content of bismuth in antimony metal was less than 0.2%, while the content of antimony in bismuth metal was less than 0.4%.
2:55 pm
THE USE OF POLYMER CONCRETE IN ACID CONTAINMENT EQUIPMENT FOR ELECTROMETALLURGICAL PLANT PRACTICE: G.J. Karcas, Corrosion Technology International, Gent, Belgium
Various non-ferrous Metals such as Copper, Nickel, Lead, Zinc and Precious Metals are produced via electrorefining and electrowinning processes. Apart from controlling electro-chemical conditions for the production of these metals, there has always existed the challenge of controlling the severely deleterious effect of the aqueous environment which serves as the medium for the production of these metals. The practice of acid containment in electrometallurgical operations has entered the world of high technology with the advent of advanced acid resistant materials such as Vinyl Ester Resin which is either combined with concrete to form Polymer Concrete or used to coat or line various acid containing vessels and surfaces. The North and South American copper industry has indulged quite heavily in the use of polymer concrete cells, electrolyte storage tanks, and the use of Vinyl Ester Resin for corrosion protective systems. Electrolytic cell and electrolyte storage tank construction has undergone a transformation from a multiple part configuration to a single piece unit. The combination of a wood, steel, or concrete frame with a lining made of lead, asphalt, rubber or plastic has been historically successful in providing protection from chemical and mechanical attack in electrolytic cells and electrolyte storage tanks all over the world. These combinations have now entered the world of obsolescence as materials such as polymer concrete have proven to provide exceptional corrosion and mechanical protection with greatly reduced maintenance requirements. Polymer concrete cells and electrolyte storage tanks fall into the category of modern tankhouse equipment which includes anode preparation machines and stainless steel cathodes, all of which play a vital role in improving quality and productivity while lowering operating costs. Polymer Concrete Cells, and other modern equipment, are succeeding in the difficult, yet important, task of driving electrometallurgical plants into the next millennium.
3:20 pm
POSSIBILITIES TO LOWER ENERGY USE IN ELECTROWINNING BY MODIFICATIONS OF LEAD ANODES: O. Forsén , J. Aromaa, Helsinki University of Technology, Laboratory of Corrosion and Material Chemistry, Vuorimiehentie 2, FIN-02150, Espoo, Finland
Lead alloys are most common anodes in metal electrowinning processes from acid solutions, even though their oxygen overpotential is very high. New inventions like mixed metal oxide anodes are not widely used, the main reasons being possibly doubts on the anode operation and large capital investments on the existing lead anodes. The main reaction on lead anodes in sulphate solutions is oxygen evolution, and on the lead dioxide film it has a high overpotential. In this paper we have studied possibilities to decrease the energy consumption in metal electrowinning by lowering the anode overpotential. Samples from currently used anode material were coated with different lead alloys using plasma spraying and detonation coating. Also modifications on the anode microstructure by melting and controlled cooling and additions of certain ions to the electrolyte were studied. Most of the studied procedures were able to lower the anode potential about 100 - 150 mV at the operating current density of 500 A/m2. Rapid cooling during the anode casting was found to be the simplest and thus also the least expensive way to lower the anode potential.
3:45 pm
STUDIES RELATING TO THE DEVELOPMENT OF AN ANODE FOR LEAD ELECTROWINNING: D.J. Robinson, Dremco Inc., Phoenix, AZ; T.J. O'Keefe, University of Missouri, Rolla, MO; C. Bemelmans, Hazen Research, Golden, Colorado; R. George, South Coast Air Quality, Diamond Bar, California
The pyrometallurgical method of recycling car batteries suffers from a potential for air pollution violations, which prompts the desire for a viable hydrometallurgical alternative. Several Processes have been proposed, including leach-electrowinning routes using HBF4 and/or H2SiF6 based electrolytes, along with anodes of PbO2 on inert substrates. A study has been carried out to evaluate the proposed systems, with the intention of demonstrating a six month life of an electrolyte-anode combination. The work, which focused on the use of H2SiF6, has indicated that there was no acceptable anode, and has lead to the development of a technique to produce smooth, glassy, adherent coatings on graphite, and to the development of an anode composed of a series of copper cored graphite rods. The paper will present results on leaching and electrowinning experiments, as well as the description of a flowsheet for a pilot plant that will be used to produce one tone per day of refined lead from spent battery paste.
4:10 pm
APPLICATION OF ELECTROCHEMICAL MEMBRANE METHODS IN HYDROMETALLURGY OF COPPER: A.S. Mustafinova, A.A. Zharmenov, Chemical-Metallurgical Institute, Karaganda, Republic of Kazakstan
Production of electrolytic copper is characterized by large volumes of acidic spent solutions the utilization of which requires considerable expenditures of neutralizing agents. At present, an economically substantiated technology does not exist for treatment of sulfuric acid rinse solutions and spent copper electrolyte which will allow for repeated use of the acid and the metal salts. Therefore, for this aim the most advantageous is the use of electromembrane methods. In the paper there is stated the essence of developed technologies which are distinguished by low waste and provide the extraction from the solutions of metallic copper, nickel and sulphuric acid. There are also considered methods of copper electrolyte processing allowing to regulate its composition by that or another component (H2SO4, Cu, Ni, As). On the base of the detailed study of the behaviour of various types of ion- exchanger membranes in copper-, nickel-, zinc-, arsenic- bearing sulphate solutions there is realized the selection of the most suitable membranes (MK-401, MA-411, MB-2 and MKL-1/MAL-2). Some variants of electromembrane processing of spent solutions of the hydrometallurgy process have passed productions tests on a number of copper-electrolyte enterprises of the countries of CIS (cooperation of Independent States). These tests showed high efficiency of developed methods.
4:30 pm
ELECTROMETALLURGY, PAST, PRESENT, AND FUTURE: F. Habashi, Laval University, Department of Mining and Metallurgy, Quebec City, Quebec, G1K 7P4, Canada
While pyrometallurgy is an ancient art and hydrometallurgy can be traced to the Middle Ages, electrometallurgy is the most recent technology since it started only after the discovery of electric current in the ninteenth century. Davy, Faraday, and Bunsen laid the foundation of this technology which had a great impact on other areas of metallurgy, e.g., the liberation of aluminum for the first time from its salts. Once aluminum was available, it was used to liberate other metals from their oxides. Electrometallurgical processes, however, are capital intensive. Improvement of the present technology should be intensely investigated.
4:55 pm
FACTORS AFFECTING CO-DEPOSITION OF COBALT AND NICKEL: R. Luo, R.V. Kumar, Department of Mining and Mineral Engineering, University of Leeds, Leeds, LS2 9JT, UK; D.J. Fray, Department of Materials Science and Metallurgy, University of Cambridge, Cambridge CB2 3QZ, UK
Room: 230C
Session Chairperson: Jaffar G. Ameeri, Aluminium Bahrain, P.O. Box 570, Manama, Baharain
2:00 pm
TRACKING THE CARBONISATION PATHWAY OF DIFFERENT PITCHES AND PITCH COKE MIXTURES BY CRYSTALLITE HEIGHT (Lc) MEASUREMENT: N.R. Turner, Bitmac Limited, Scunthorpe Works, Dawes Lane, Scunthorpe, North Lincolnshire, DN15 6UR, UK
Pitches with a range of properties, e.g. Q.I. content, were carbonised to a final temperature of 1080°C. The same pitches were co-carbonised with finely divided petroleum coke. Lc has been measured at the end of the carbonisation cycles and at intermediate points to track the status of carbonisation for different pitches and pitch/filler coke combinations. The original pitch Q.I. content and pitch aromaticity appear to influence the carbonisation pathway of the pitch and the pitch-coke mixtures according to Lc data. The properties of the filler coke also appear to influence the ultimate size of the crystallites formed in the binder matrix at the end of the carbonisation cycle. It appears that there is a variable degree of interaction between the filler coke and the pitch, with extent of the interaction depending on the properties of the two components. The results so far reinforce the importance of consistency in raw materials preparation and in the baking process if uniform anodes are to be produced.
2:25 pm
ABSORPTION OF OXYGEN FROM AIR INTO COAL TAR PITCH: Jaromir Cibulec, Daniel Kostal, DEZA Corporation, P.O. Box 28, 757 28 Valasske Mezirici, Czech Republic
Changes in the properties of coal tar pitch during contact with atmospheric oxygen were studied. There were found to be changes in the content of toluene insolubles and the melting point. These changes are dependent on the quantity of the absorbed oxygen and on the temperature of the pitch during absorption. As the reaction of the air with the pitch is very rapid, reaction speed is controlled by the speed of absorption of oxygen into the pitch. When surface of the pitch is motionless, the rate determining step of the reaction is controlled by the rate of absorption of the oxygen into the pitch which is dependent on the diffusion coefficient of molecular oxygen in air. When the surface of the pitch is agitated, the reaction rate is controlled by the rate of reaction of the oxygen with the pitch. The mechanism of the reaction is described and a practical application is discussed.
2:50 pm
X-RAY AND MICROSCOPY INVESTIGATIONS ON THE CATALYST-CARBONIZED STRUCTURE OF COAL TAR PITCH--PETROLEUM COKE PILOT ANODES: Jilai Xue and Harald A. Oye, Institute of Inorganic Chemistry, the Norwegian University of Science and Technology, N-7034 Trondheim, Norway
The carbonization has important impacts on the energy consumption and the productivity in anode baking. It was found that the coke yield of the coal tar pitch binder in its mixtures with petroleum coke increased with addition of catalysts, such as S, AlF3, and FeCp2 (CO)4. Then the structures of the samples taken from the carbon pilot anodes with and without above additions were further investigated using X-ray diffraction and microscopic techniques. The remained S, Al and Fe in the carbons after heat treatments were also analysed. The coke yield, Lc value, and porosity will be presented against various heat treatment temperatures for different additions. The possibility of obtaining the carbon anodes by lower baking temperatures with the required structure will be discussed based on above results.
3:15 pm BREAK
3:35 pm
COAL-TAR/PETRO INDUSTRIAL PITCHES: E.R. McHenry, Koppers Industries, 1005 Wm. Pitt Way, Pittsburgh, PA 15238
Throughout the 20th Century coal tar has been the dominant raw material for producing industrial pitches in North America. However, the coal tar supply in North America has been decreasing due to closure of metallurgical and foundry coke-oven batteries. Because of lower carbon yield, petroleum-based pitches have only limited usage. To increase the raw material base and provide more flexibility in characteristics, coal tar/petro pitches are being developed. The evaluation of the coal-tar/petro industrial pitches is reported.
4:00 pm
PETROLEUM PITCH, A REAL ALTERNATIVE TO COAL TAR PITCH AS BINDER MATERIAL FOR ANODE PRODUCTION: Roger Marzin, Carola Acuna, Intevep S.A., Departamento de Refinacion, Apdo. 76343, Caracas 1070 A, Venezuela; Maria de Oteyza, Maraven, S.A., Nuevos Desarrollos, PO Box 829, Caracas 1010A, Venezuela; Raymond C. Perruchoud, R&D Center, Le Chable, P.O. Box 157, CH-3960 Sierre, Switzerland
A petroleum pitch, specially developed for aluminum anodes was produced, at commercial scale, as a real alternative to coal tar pitch. Using a highly aromatic refinery stream and well designed process conditions, it is possible to produce the right chemical composition needed for a high quality binder material for anodes. Through a Dynamic Process Optimization study, the paste formulation and parameters were adjusted to produce full size anodes with 100 % petroleum pitch. The performance of baked anodes were similar to coal tar pitch anodes, with the advantage of a much lower content of cancerogenic Polycyclic Aromatic Hydrocarbons (PAH). Comparative green and baked anode properties, PAH emissions and storage stability test for molten pitch are presented.
4:25 pm
REDUCTION OF POLYCYCLIC AROMATIC HYDROCARBONS (PAH) IN ANODES BY USING PETROLEUM PITCH AS BINDER MATERIAL: A COMPARAISON OF ANODE PROPERTIES AND ANODE BEHAVIOR OF PETROLEUM AND COAL TAR PITCH ANODES: Dr. Ulrich Mannweiler R&D Carbon Ltd. Winterthurerstrasse 92, CH-8006 Zürich, Switzerland; Raymond C. Perruchoud, R & D Carbon Ltd., Le Chable, P.O. Box 157, CH-3960 Sierre, Switzerland; Roger Marzin, Intevep S.A., Departamento de Refinacion, Apdo. 76343, Caracas 1070 A, Venezuela
Petroleum pitch as binder material of anodes has the great advantage, that the content of cancerogenic Polycyclic Aromatic Hydrocarbons (PAH) compared to coal tar pitch is practically inexistent. In applying the correct content of pitch and through optimization of all process parameters during anode manufacturing, full size anodes with coal tar pitch (CTP) and petroleum pitch (PP) have been prepared in an anode plant. While baked anode properties for coal tar and petroleum pitch anodes are very similar, the PAH content can be reduced by 98 %. In this paper the behavior of representative lots of coal tar and petroleum pitch anodes in 100 kA electrolysis cells and the properties of the resulting anode butts will be described.
4:50 pm
TESTS OF PITCH PRODUCED FROM THE MIXTURE OF COAL RESIN AND REMAINDERS AFTER OIL PYROLYSIS: M.L. Itskov, E.A. Yanko, Russian National Aluminium-Magnésium Institute (VAMI), 199026 St. Petersburg, Russia; V.V. Krukov, Central Investigations Laboratory, 101000, Moscow, Russia
Actually the main type of binding materials, used in the anode paste production is a coal tar pitch. This binder provides the necessary quality of anode, but its using is related to increased emissions of cancerogenous matters, especially, benzapyrene, into the anode paste and aluminium reduction plants. The cancerogenous danger could be considerably reduced if using, as a binding material in the anode paste production, pitch produced at the distillation of the mixture of coal resin and remainders after oil pyrolysis, called "modified pitch". This report gives the results of laboratory investigations and industrial tests of anode paste, produced with using above mentioned binder, in the aluminium reduction. The laboratory investigations and industrial tests on Soderberg pots have shown, that by using the modified pitch we succeeded to produce anode paste of satisfactory quality. In this case we achieve a considerable reducing of the emissions of cancerogenous matters, namely benzapyrene, from the surface of anode.
Room: 230B
Session Chairperson: Julian V. Copenhaver, NSA - A Division of Southwire, P.O. Box 500, Hawesville, KY 42348
2:00 pm
ON THE EFFICIENCY OF IN-LINE DEVICES TO CLEAN THE MELT: P. Pouly, E. Wuilloud, Alusuisse Technology and Management Ltd., Technology Center Chippis, 3965 Chippis, Switzerland
Rotating impellers for injecting gas into aluminium melt are an often used in-line device to remove impurities. However, the treatment efficiency is seldom investigated, as the precise effect of flotation is not well understood. Moreover, the interpretation of the results obtained by measuring the inclusion content is not always obvious. For these reasons, it is sometimes observed that rotors work in an erroneous way. This paper illustrates how the variation of some parameters can affect, either in a good or a bad way, the efficiency of rotating impellers. The problematic of measurement techniques with some of their limitations will also be exposed. A special emphasis will be laid on inclusion removal.
2:20 pm
THE FLUID MECHANICS IN THE HI10 HYCAST REACTOR: Bodil Hop, Frede Frisvold, Stein Tore Johansen, SINTEF Materials Technology, N-7034 Trondheim, Norway; Bjørn Rasch, Hydro Aluminium a.s R&D Centre, N-6600 Sunndalsøra, Norway
The flowfield in the HI10 HYCAST reactor is investigated both theoretically and experimentally. The paper demonstrates how fluid flow calculations can be used to interpolate between experimental velocities. The experiments are performed by a two component Laser Doppler-technique in a reactor which has one single baffle. In the 3-dimensional computation a rotating grid for the rotor is imposed on the stationary grid for the reactor body. The flow effects caused by the baffle are discussed in relation to metallurgical performance. The turbulent dissipation rate is translated into bubble sizes and the result is compared to experimental bubble sizes. The paper discusses the reasons for deviations between predicted and experimental velocities.
2:40 pm
IN-LINE FLUXING WITH HIGH SPEED MULTIPLE DISPERSER ROTORS: D.C. Chesonis, H. Yu, Aluminum Company of America, Alcoa Technical Center, 100 Technical Drive, Alcoa Center, PA 15069; M. Scherbak, Aluminum Company of America, Ingot Technology Group, 900 South Gay Street, Knoxville, TN 37902
The impurity removal efficiency of in-line fluxing equipment depends strongly on the interfacial area between the gas bubbles and the molten metal. High interfacial areas can be generated by direct shearing of the gas bubbles using rotating dispersers. Placing multiple dispersers on a single rotor shaft and injecting gas into the metal at each disperser increases this direct shearing, improving the efficiency of the in-line fluxing unit. Small dispersers can operate at high gas loading and high speed without vortexing. Small dispersers also allow smaller metal treatment units, reducing floor space requirements and metal holdup. This paper will summarize the theory behind the direct shearing approach and will present data illustrating the difference in trace element removal for single, double, and triple disperser rotors. Comparisons with other in-line fluxing equipment will also be presented.
3:00 pm
EVOLUTION OF THE ALPUR® ROTOR TOWARDS IMPROVED PERFORMANCE
REPLACEMENT Panel Discussion on Hydrogen Levels
3:20 pm
QUANTIFIED QUALITY AND EFFICIENCY IMPROVEMENTS AT SAPA, LTD.: Stephen J. Rose, SAPA Ltd., Saw Pit Lane, Tibshelf Derbyshire DE55 5NH, England; David W. Busch, Foseco Inc., 777 Old Saw Mill River Road, Tarrytown, NY 10591-6729
SAPA, Ltd. is a 3 press aluminium extrusion operation specializing in quality flat and hollow product with a small remelt facility for in house generated scrap. In 1994 a decision was made to improve the overall quality of the in plant remelt facility. The first capital improvement was the installation of a SNIF Sheer P-60U two nozzle degassing system with immersion heaters. This unit was started up in December of 1994. A new melter with an exhaust collection system was brought on line in September of 1995. This paper will describe how improved melt treatment produced measurable economic benefits downstream.
3:40 pm BREAK
3:50 pm
CHLORINATION OF TiB2 GRAIN REFINED ALUMINIUM: T. Gudmundsson, G. Saevarsdottir, T.I. Sigfusson, Science Institute, University of Iceland, Dunhaga 3, 107 Reykjavik, Iceland; D.G. McCartney, Department of Materials Engineering & Materials Design, University of Nottingham, United Kingdom
The effects of fluxing grain refined molten aluminium with Ar/Cl2 gas mixtures have been studied both in the laboratory and during experimental casthouse trials. Samples taken during gas fluxing were subjected to chemical analysis and examined in the scanning electron microscope (SEM). Detailed chemical analysis of agglomerates was undertaken in the SEM using energy dispersive X-ray analysis and the elements Na, K and F were found to be associated with TiB2 clusters. Results will be presented in detail and discussed in the context of previously published work on grain refiner characterization and wettability of TiB2 in liquid Al/salt mixtures.
4:10 pm
COALESCENCE BEHAVIOUR OF ALUMINUM DROPLETS UNDER A MOLTEN SALT FLUX COVER: K.J. Friesen, T.A .Utigard, Department of Metallurgy and Materials Science, University of Toronto, 184 College Street, Toronto, Ontario, Canada M5S 1A4; C. Dupuis, J.P .Martin, Arvida Laboratories and Development Centre, Alcan International Ltd., 1955 Mellon Blvd., Jonquière, Québec, Canada G7S 4K8
The formation of a suspension of aluminum droplets in salt fluxes used during the refining and recycling of aluminum, leads to decreased aluminum recovery. To enhance the coalescence and recovery, it is required to remove the oxide layer formed on the surface of these small aluminum drops. Coalescence behaviour of aluminum droplets under a molten salt flux cover was investigated using a hot stage microscope at a temperature of 740°C. Two salt systems were studied: i) an equal weight mixture of NaCl-KCl with chloride and fluoride additives and ii) the MgCl2-KCl system with fluoride additives. Chloride salts did nothing to enhance coalescence but fluoride additions to both salt systems removed the oxide film from the metal. NaF, KF and Na3AlF6 additives were the most effective salts at stripping away the oxide layer and promoting coalescence. As the concentration of MgCl2 increased, the ability of the droplets to coalesce decreased significantly.
4:30 pm
KINETICS OF MAGNESIUM REMOVAL FROM ALUMINUM ALLOYS BY CHLORINE FLUXING: Qian Fu, J.W .Evans, Dept. of Materials Science and Mineral Engineering, University of California, Berkeley CA 94720
Chlorine fluxing (sparging chlorine-argon mixtures into the melt) is a standard procedure for eliminating magnesium from Al-Mg alloys. The kinetics of the reactions/mass transport involved have not been extensively studied in the past. The paper describes an ongoing investigation of those kinetics. Much of the work to date has been using laboratory-scale melts with magnesium removal measured by analysis of samples. Techniques have been developed to measure both bubble frequency and bubble residence time in the melt, enabling estimation of the total interfacial area and, consequently, the intrinsic kinetics of the phenomena involved. It is observed that, above the melting point of MgCl2, there exists a "critical Mg concentration". Above this critical concentration the reactions are sufficiently fast that all entering chlorine is converted to MgCl2 and there are negligible chlorine emissions. Below this concentration, Mg mass transport in the alloy appears to be rate-controlling. The results are interpreted in terms of a mathematical model for fluxing. The paper also describes the preliminary results of experiments on a larger scale (few tens kg alloy) in which bubble probes are used to detect the dispersion of gas injected into the melt and measure the interfacial area.
4:50 pm
INTERFACIAL TENSION BETWEEN ALUMINUM AND CHLORIDE-FLUORIDE MELTS: A. Silny, Institute of Inorganic Chemistry, Slovak Academy of Sciences, SK-84236 Bratislava, Slovakia; T.A. Utigard, Department of Metallurgy and Materials Science, University of Toronto, M5S 3E4 Toronto, Canada
A technique was developed for the measurement of the interfacial tension between liquid metals and molten salts at elevated temperatures. The technique is based on the measurement of the capillary depression ocurring when a capillary, which is moved vertically down through the molten salt layer, passes through the salt/metal interface. The depression is measured by simultaneous video recording of the immersion height of the alumina capillary and the position of a liquid meniscus in a horizontal tube connected to the alumina capillary. The interfacial tension was measured in the following systems: aluminium and an equimolar melt of NaCl and KCl with several salt additions at 1000 K, aluminium and NaCl-NaF at 1123 K, and aluminium and NaCl-KF at 1123 K and aluminium and equimolar melt of NaCl-KCl with additions of NaF, KF, LiF, BaF2, SrF2, CaF2, Na3AlF6, MgF2 and AlF3. It was found that the interfacial tension decreases with increasing NaF, KF, LiF, CaF2, BaF2 and SrF2 additions, remains unchanged with AlF3 additions and slightly decreases with MgF2 and Na3AlF6 additions. Aluminium and equimolar melt of NaCl-KCl with additions of LiCl, AlCl3, BaCl2, CaCl2, MgCl2. All above chlorides slightly increase the interfacial tension in the system. Obtained data are compared with the ones found in literature and some conclusions for the aluminium refining and recycling were derived.
5:10 pm
A COMPARATIVE STUDY ON THE EFFICIENCIES OF Na2SiF6 AND AlF3 FOR DEMAGGING MOLTEN ALUMINUM BY SUBMERGED POWDER INJECTION: A. Flores-Valdéz, M.A. Hinojosa-SanMiguel, A.H. Castillejos-Escobar, E. Macias-Avila, F.A. Acosta-González, Centro de Investigación y de Estudios Avanzados del IPN, Unidad Saltillo, P.O. Box 663, 25000 Saltillo, Coahuila, México
A comparative study on the efficiencies of Na2SiF6 and AlF3 for demagging molten aluminum by submerged powder injection is reported. Experimental trials carried out at a scale of 250 Kg molten metal furnace capacity showed that the higher efficiencies, ~ 70%, were attained by the use of the Na2SiF6 powder. The basic parameter to establish the comparison between the two demagging agents was the powder size, maintaining constant the other important parameters, e.g., initial magnesium content, carrier gas flow-to-powder flow ratio, pressure of the carrier gas, temperature of the bath, the dimensions of the reactor and the shape and dimensions of the injection lance. The removal efficiency was monitored analyzing the magnesium content in both the metal and the slag. The fumes produced were also analyzed, having found that the Na2SiF6 powders have the inconvenience of giving place to the generation of highly toxic SiF4 fumes.
Room: 330C
Session Chairperson: En Ma, Louisiana State Univ., Dept. of Mechanical Eng., Baton Rouge, LA 70803
2:00 pm INVITED
COMBUSTION FLAME SYNTHESIS OF NANOPHASE MATERIALS: B.H. Kear*, G. Skandan***, N. Glumac**, and Y. Che*, *Ceramics, **Mechanical & Aerospace Engineering, Rutgers University, NJ, ***Nanopowder Enterprises, Inc. Piscataway, NJ
Non-agglomerated nanopowders are becoming increasingly important for a number of commercial applications including UV-scattering, chemical mechanical polishing (CMP), displays and catalysis, among others. We have developed a continuous nanopowder production process, called Chemical Vapor Condensation (CVC), which involves precursor pyrolysis and condensation in a reduced pressure environment. We have introduced a flat-flame combustor unit, operable at low pressures, as a heat source in place of the original hot wall reactor. The temperature profile is uniform across the entire face of the burner, therefore, the reactants experience the same processing history and the powder has a uniform particle size distribution. The modified process, called Combustion Flame-Chemical Vapor Condensation (CF-CVC), has been used to produce a range of non-agglomerated nanoparticles (3-50 nm average particle size) of single phase, multiphase, and multicomponent materials. Examples include Al2O3, SiO2, TiO2, Al2O3/SiO2 and Eu:Y2O3. The as-synthesized powder is fully pyrolyzed (characterized by TGA), has a high surface area (SiO2 > 300 m2/g; TiO2 > 80 m2/g), and is non-agglomerated (TEM and BET pore size distribution). In addition, when the superheated particles leaving the combustion flame impinge on a heated substrate, in situ sintering can occur. Nanoporous or dense films or multilayered thin film structures can be synthesized. We have demonstrated the scalability of the process by increasing the burner diameter. Design consideration, processing parameters, powder characteristics and the commercial potential for the powders will be discussed. This work is supported in part by the Office of Naval Research contract #N00014-95-C-0283.
2:30 pm
EVIDENCE OF SURFACE ROUGHNESS IN NANOSTRUCTURES: M. José Yacamán, Institute de Física, Universidad Nacional Autóma de México, Apdo, Postal 20-364, 01000 México, D.F., México
Nanostructured Materials present very unique properties which has allow several technological applications. In the case of chemical reactions nanostructured materials have been used as catalyst. It is know that reduction of the particle size to a few nanometers produce an increased catalytic activity. The origin of this activity has been attributed to the increase surface area. We will show in this paper that nanoparticles present an increased surface roughness. This roughness increases the number of kink which provide an excellent site for promoting chemical reactions. In order to characterized the roughness we have used a new technique which combines High Resolution Electron Microscopy with computer length, It will be also shown that in several cases, the particle size in the image do Dot correspond with the true particle size. A method to correct this problem will be discussed.
3:00 pm INVITED
NANOSTRUCTURED MATERIALS VIA CHEMICAL ROUTES: Kenneth E. Gonsalves, Department of Chemistry & Institute of Materials Science, University of Connecticut, Storrs, CT 06269
This talk will focus on the chemical synthesis and processing of nanostructured materials. The precursor chemistry for the synthesis of nanostructured metals, ceramics, polymers, biomaterials, semiconductors and nanocomposites will be outlined and selected examples presented. Issues such as material purity, homogeneity, agglomeration and scale-up will be addressed.
3:30 pm BREAK
3:45 pm
TEM AND HRTEM OF NANOSTRUCTURED M50 TYPE STEEL PREPARED BY HOT PRESSING OF CHEMICALLY SYNTHESIZED POWDERS: G. M. Chow1, C.R. Feng2, Naval Research Laboratory, Washington, DC 20375; S.P. Rangarajan, X. Chen, K.E. Gonsalves, Institute of Materials Science, University of Connecticut, Storrs, CT 06269; C.C. Law, Pratt & Whitney, United Technologies Corporation, East Hartford, CT 06108. 1Laboratory for Molecular Interfacial Interactions, 2Materials Science Division
Nanostructured M50 type steel materials were prepared by hot pressing the precursor powders chemically synthesized using two different techniques, namely, thermal decomposition and co-reduction. During the hot press process, the precursor powders were transformed to nanocrystalline phases with the precipitation of carbides. Simultaneously the crystalline powders were densified. The densified samples were studied using both conventional and high resolution transmission electron microscopy. The effects of hot pressing temperature, time and pressure on the evolution of nanostructures, grain growth and defects formation are discussed.
4:05 pm
SYNTHESIS OF NANOSTRUCTURE W/Cu/Co ALLOY BY THERMOCHEMICAL METHOD: Gil-Geun Lee, Gook-Hyun Ha, Dong-Won Lee, Byoung-Kee Kim, Korea Institute of Machinery & Materials, 66 Sangnam-Dong, Changwon, Kyungnam 641-010, Korea
Nanostructure W/Cu/Co alloy was developed by thermochemical processing method using metallic salt precursors as the starting material for improving thermal, electrical and mechanical properties. Nanostructure W/Cu/Co powder have loosely agglomerated homogeneous clusters of nanoscale size W( 50 nm), Cu. and Co particles. The density and microhardness of the sintered nanostructure W/Cu/Co increased with increasing of Co content from 0.1 to 0.7wt.%, but electrical conductivity drastically decreased with addition of Co. Full density for nanostructure W/Cu/Co was achieved when sintered at 1473K for 20 minutes in H2 atmosphere with addition of 0.5wt.%Co. It was proved that optimum Co content is under 0.5wt.% when based on the relations between electrical conductivity and density. It is shown that nanostructure W/Cu/Co have higher sinterability and better electrical conductivity than conventional W/Cu/Co.
4:25 pm
MÖSSBAUER EFFECT STUDY OF MECHANICALLY ALLOYED 
-Fe3Zn10 and 1-Fe5Zn21 CUBIC INTERMEDIATE PHASES: Oswald N.C. Uwakweh, Zhentong Liu, Materials Science & Engineering, University of Cincinnati, Cincinnati, OH 45221-0012
The Mössbauer effect measurements of as-ball milled mechanically alloyed Fe-Zn intermediate phases of 
-Fe3Zn10 and -Fe5Zn21 compositions exhibit characteristic spectra consisting of triplets. Each is characterized with an Fe-site with a high quadrupole splitting measuring 0.94 mm/s, together with three other doublets. In the aged states, both compositions show spectra free of the Fe-site with the large quadrupole splitting. This suggests that both have similar metastable states, while their separate transformation paths to stable equilibrium states yield distinct crystal structures as found in the literature.
4:45 pm
NEUTRON DIFFRACTION & PHASE EVOLUTION OF MECHANICALLY ALLOYED 
-FeZn13 INTERMETALLIC: Oswald N.C. Uwakweh, Zhentong Liu, Materials Science & Engineering, University of Cincinnati, Cincinnati, OH 45221-0012; Brian Chakoumakos, Stephen Spooner, Oak Ridge National Laboratory, Solid State Division, Oak Ridge, TN 37831-6393
High Energy Ball -Milling is used to synthesize 
-FeZn13 intermetallic. The mechanically alloyed phase in the as-millet state is determined to be metastable, while the characteristic stages associated with the stable equilibrium transformation are identified based on DSC measurements. The as-milled material is described in terms of mechanical mixture of the elemental constituents, while the equilibrium state is confirmed to have a C2/m space group, with lattice parameters of a=13.40995 Å, b=7.60586 Å, c=5.07629 Å, and =127° 18'. The atomic positions of Fe and Zn are compared with reported values.
Room: 340D
Session Chairperson: Martin Crimp, Dept. of Materials Science and Mechanics, Michigan State University, East Lansing, MI 48824
2:00 pm INVITED
TITLE TBA: James C. Williams, General Electric Aircraft Engines, 1201 Edison Drive, Cincinnati, OH 45216
2:30 pm
THE STRUCTURE OF MATERIALS ENGINEERING; A NEW MODEL FOR MATERIALS ENGINEERING CURRICULA: Blair London, Linda S. Vanasupa, Robert H. Heidersbach, Materials Engineering Department, California Polytechnic State University, San Luis Obispo, CA 93407
We have developed a new model for the materials engineering curriculum designed around a structure that must be carefully "built" during our program. The model clarifies the materials engineering curriculum, representing a new organizational structure that incorporates the broad concepts yet allows room for technological changes. We used four guiding principles: (1) the curriculum must impart certain core competencies to the students, (2) the courses are either four or five units (quarter system), many with a strong laboratory emphasis, (3) the sequence of courses is logical and can be clearly communicated, and (4) it must be accredited. The model can be summarized by the major physical features of the building. The foundation includes an introduction to materials engineering and a course on the structure of solids. The pillars represent the four basic parts of materials engineering (thermodynamics, kinetics, mechanical properties, electronic properties) that are common to, and "hold up", all other aspects of the field. The roof consists of basic courses in metals, ceramics, polymers, and composites along with several technical electives. Finally, the building shows the capstone Senior Project as the culmination of their undergraduate education.
2:55 pm
MATERIALS SCIENCE CURRICULUM: HOW WE SHOULD BE CHANGING IT: David Laughlin, Dept. Materials Science & Engineering, Carnegie Mellon Univ., Pittsburgh, PA 15213
The Materials Science curriculum in many Universities is a direct descendant of their Metallurgy or Metallurgical Engineering curriculum. As such, they often have in them "relics" of the past that seem to some to be outdated. One of the theses of this presentation is that some of the "relics" are indeed "holy relics" which if they are discarded by "reforming zealots" may lead to a crisis in the curriculum. A second thesis, is that indeed some of the elements of the curriculum must be changed in order to better prepare our students for the changing directions of the field. In this presentation the important elements which should remain are discussed as are some new themes that currently are not universally present in our curricula. Suggestions on how these "old" and "new" themes are to be integrated into the overall curricula will close out the presentation.
3:20 pm
UNDERGRADUATE ENGINEERING AND MSE EDUCATION AT THE UNIVERSITY OF MICHIGAN: J.W. Jones, R. Gibala, Dept. of Materials Sci. and Engineering, Univ. of Michigan, Ann Arbor, MI 48109-2136
The College of Engineering at the University of Michigan has begun an ambitious program to restructure its undergraduate curriculum. The new curriculum will increase freshman year participation in engineering and provide an active program for each student to learn communication skills throughout the curriculum, while increasing flexibility within all discipline-specific curricula. Furthermore the new approach calls for the restructuring of courses into 4 credit-hour units, thereby enabling students to complete their studies in 4 years by selecting 4 courses of 4 credit hours in each of 8 semesters for a total of 128 credit hours. The impact of this new curriculum template on Materials Science and Engineering education at Michigan will be described. Emphasis will be placed on proposed changes in laboratory and design courses to accommodate a strong and professionally administered engineering communications segment. The approaches used to provide breadth and depth in the MSE curriculum will also be described. Finally, the impact of these college-wide changes on MSE's role in providing an introductory materials education to non-MSE majors will be described.
3:45 pm INVITED
TITLE TBA: Harold Weinstock, Air Force Office of Scientific Research, 110 Duncan Avenue, Bolling AFB DC 20332-0001
4:15 pm
PANEL DISCUSSION: Participants to be announced. A summary of the symposium as completed by the Panel Discussion is planned to be submitted for publication in the JOM.
Room: 332
Session Chairpersons: Paul T. Vianco, Sandia National Laboratories, Department 1831, Mail Stop 0340, P.O. Box 5800, Albuquerque, NM 87185; James A. Warren, Metallurgy Division, Materials Science and Engineering Laboratory, National Institute of Standards and Technology, Gaithersburg, MD 20899
2:00 pm INVITED
THE NIST SOLDER INTERCONNECT DESIGN TEAM PROGRAM: James A. Warren, Carol A. Handwerker, Metallurgy Division, Materials Science and Engineering Laboratory, National Institute of Standards and Technology, Gaithersburg, MD 20899
The NIST Solder Interconnect Design Team has been found to address several pressing issues in the design and fabrication of circuit board. Having met frequently over the past three years, in partnership with academic and industrial researchers, the Team has established an agenda for solving solder joint shape, and the consequential thermal/mechanical properties of the formed joint. Our ultimate goal is to provide the industrial community with a suite of useful software tools for solder interconnect design, and to provide solved test problems (available electronically on the World Wide Web), that can be modified to suit the needs of the particular user. With this in mind we are actively supporting the development of software which will interface the public domain program Surface Evolver, which has been shown to be quite capable at computing equilibrium meniscus shapes. A discussion of the outstanding problems, as well as the software tools under development will be presented.
2:20 pm INVITED
MODELING NON-ISOTHERMAL INTERMETALLIC LAYER GROWTH IN THE 63Sn-37Pb SYSTEM: Paul T. Vianco, P.L. Hopkins, K.L. Erickson, D.R. Frear, R. Davidson, Department 1831, Mail Stop 0340, P.O. Box 5800, Sandia National Laboratories, Albuquerque, NM 87185
The reliability of mechanical and electronic systems can be acutely dependent on the integrity of the soldered joints used in their assembly. During product manufacture, intermetallic layers from in reaction zones between the dissimilar materials of solder joints. Thermal cycling during service can cause further growth of the intermetallic layer, which may jeopardize the mechanical integrity of the joint as well as its capacity for rework or repair later on. Models describing service-related changes to solder joint microstructure are essential to understanding and predicting the long-term mechanical reliability and serviceability of these interconnects. In previously published work (J. Electronic Materials, v. 23, No. 8, 1994, pp. 721-729) a model describing the diffusion-controlled growth of multiple intermetallic layers and the displacement of the interfaces between layers was developed and implemented in a one-dimensional computer code based on the method-of-lines. The model can accommodate cases involving: (1) finite initial layer thickness, (2) rate-limiting interfacial reactions, (3) multiple and variable diffusion coefficients, and (4) finite material boundaries. Additionally, the effects of nucleation can be modeled empirically. A transformation of spatial coordinate circumvented the need to remesh the growing and (or) shrinking layers. Results from the one-dimensional code were verified by comparing the numerical output with analytical solutions for simple systems involving two, three, and five layers. The computer code was then applied for analysis of intermetallic layer growth from solder aging experiments performed with 100Sn and 63Sn-37Pb solders. The analysis indicated that intermetallic layer growth was consistent with bulk diffusion mechanism involving Cu and (or) Sn and variable diffusion coefficients that reflect some enhanced diffusion during early growth. In this work, non-isothermal solder-aging experiments were done with the 63Sn-37Pb/Cu system using two temperature histories: a low frequency history consisting of 4 cycles per day between -50 and 170°C, and a high frequency history consisting of 72 cycles per day and the same limits. Thicknesses of both the Cu3Sn and Cu6Sn5 intermetallic layers were determined as a function of time for both temperature histories. An enhanced version of the previously developed model was used to predict the non-isothermal intermetallic layer growth for both temperature histories. Arrhenius expressions for diffusion coefficients in both the Cu3Sn and Cu6Sn5 layers were determined using experimental data from previous isothermal studies. This paper describes the non-isothermal experiments and a comparison of calculated and observed layer growth as a function of time. This work was performed at Sandia National Laboratories and supported by the U.S. Department of Energy under contract DE-AC04-94AL8500.
2:50 pm INVITED
SOLDER JOINT FORMATION, SIMULATION AND RELIABILITY PREDICTION: Xiaohua Wu, Kai Hu, Xinyu Dou, Gary Mui, Chao-pin Yeh, and Karl Wyatt, Applied Simulation and Modeling Research (ASMR), Corporate Software Center (CSC), Motorola Inc., 1303 E. Algonquin Road, Mail Stop: IL01/ANX2, Schaumburg, IL 60196
The fatigue-induced solder joint failure of surface mounted electronic devices has become one of the most critical reliability issues in electronic packaging industry. Solder joint reliability performance has been found to be highly dependent on the solder joint configuration, which, in turn, is governed by bond pad size, component weight, alloy material, and leadframe structure, as well as solder reflow characteristics. The objective of this work is to develop numerical models: 1) to simulate the solder joint formation during the reflow process; 2) to determine the stress/strain distribution within the joint; and 3) further predict the reliability (fatigue life) of the solder joints. The solder joint formation process during solidification stage can be simulated using the Surface Evolver software tool developed by University of Minnesota. The thermomechanical stress-strain analysis can then be carried out using ANSYS to study selected critical design/manufacturing parameters such as leadframe geometry, pad size and dimensions, solder paste volume, leadframe placement misalignment, etc. This effort also involves the development of interface linking Surface Evolver and ANSYS.
3:15 pm BREAK
3:25 pm INVITED
COMPUTATIONAL CONTINUUM MODELING OF SOLDER JOINT INTERCONNECTS: Steve Burchett, M.K. Neilsen, D.R. Frear, J.J. Stephens, Department 9117, Mail Stop 0443, P.O. Box 5800, Sandia National Laboratories, Albuquerque, NM 87185
The most commonly used solder for electrical interconnects in electronic packages is the near eutectic 60Sn-40Pb alloy. This alloy has a number of processing advantages (suitable melting point of 183°C and good wetting behavior). However, under conditions of cyclic strain and temperature (thermomechanical fatigue) the microstructure of this alloy undergoes a heterogeneous coarsening and failure process that makes prediction of solder joint lifetime complex. A visco-plastic, microstructural dependent, constitutive model for solder has been developed and implemented into a finite element code. With this computational capability, the thermomechanical response of solder interconnects, including microstructural evolution, can be predicted. This capability was applied to predict the thermomechanical response of various solder interconnects to determine the effects of variations in geometry and loading. In this paper, the constitutive model will be briefly discussed and response predicted by the constitutive model will be compared to material test results. Finally, the results of computational studies to determine the effect of geometry and loading variations will be presented. This work was performed at Sandia National Laboratories, and supported by the U. S. Department of Energy under Contract No. DE-AC04-94AL8500.
3:50 pm INVITED
AN ELASTOPLASTIC BEAM MODEL FOR COLUMN-GRID-ARRAY (CGA) SOLDER INTERCONNECTS: Steven M. Heinrich*, J.A. Swanson*, and P.S. Lee**; *Rockwell Automation, Allen-Bradley Co., Milwaukee, WI 53233; **Department of Civil and Environmental Engineering, Marquette University, Milwaukee, WI 53233
A semi-analytical model is developed and implemented to analyze the deformation of solder columns in column-grid-array (CGA) assemblies. Each solder column is modeled as a prismatic beam of circular cross-section, subjected to end shearing deflections caused by thermal mismatch between the module and the circuit board. The solder is idealized as an elastic-perfectly plastic material whose yielding is governed by the Von Mises criterion. Since the columns are relatively short, transverse shear deformation has been incorporated into the beam model. The results generated with the model indicate the following: (a) yielding is governed by bending for slenderness ratios (height-to-diameter) of h/d
1/
3; (b) the nonlinear stiffness relationship for a sheared column, presented in dimensionless form, reduces to a single curve which is valid for arbitrary values of slenderness ratio (1/3) and material parameters; (c) the dimensionless relationship between maximum shear strain (in the Tresca sense) and the relative end deflection depends on Poisson's ratio but is independent of the other material parameters and the slenderness ratio. The nonlinear stiffness results presented in the paper may be used to create more efficient finite element models of entire assemblies by replacing each column with a single nonlinear spring element. When used in conjunction with an appropriate Coffin-Manson relationship, the maximum shear strain curves presented herein may be utilized to estimate column fatigue life.
4:15 pm INVITED
EFFECT OF ASSEMBLY STIFFNESS AND TEMPERATURE CYCLE PARAMETERS ON THERMOMECHANICAL RESPONSE OF SOLDER JOINTS: Ahmer Syed, Delco Electronics Corporation, M/S T100-26, 700 East Firman Drive, Kokomo, IN 46904
Grain boundary sliding and matrix creep are recognized as the main contributors to creep behavior of eutectic Sn/Pb solder joints in thermomechanical loading. The relative contribution and dominance of these mechanisms in the total response is governed by the external (temperature, assembly stiffness, CTE mismatch) as well as the internal shape (shape, size and thickness of the solder joint) factors. This paper examines the effect of external factors on the grain boundary and matrix creep response of solder joints. Experimental as well as numerical techniques are employed to compare which mechanism dominates under several different assembly stiffness and temperature profile conditions. A fatigue life prediction model which accounts for different mechanisms is used to determine the impact of grain boundary sliding and matrix creep on solder joint fatigue life.
4:40 pm INVITED
BEHAVIOR OF SOLDER JOINTS UNDER COMPLEX DISPLACEMENT LOADING: Matthew G. Bevan, Manfred Wutting, The John Hopkins University, Applied Physics Laboratory, John Hopkins Road, Laurel, MD 20723
In many applications, solder is exposed to a complex fatigue spectrum composed of thermal effects, shock and vibration. In the laboratory; however, solder is usually tested using simple wave forms generated by mechanical testers or by thermal cycling. In order to apply laboratory results to predict the real world fatigue, several crucial steps are necessary. One of these steps is understanding how to combine the effects of simple displacement (or stress) waves to predict the effects of complex waves. One method of combining waves is to use a linear damage accumulation model employing rainflow analysis. Briefly, rainflow analysis establishes an envelope to strain (or stress) range of the waveform, and then measures the reversals within the range. Damage from the envelope is added to the damage from the individual reversals within the envelope to calculate the damage of the whole cycle. The linear damage accumulation model is not valid for all loading conditions. Damage is accumulated in a linear fashion, assuming there is no wave-to-wave interaction. For Mode I (tensile) loading, this assumption is known to be false. Six mechanisms for-wave-to-wave interaction are known. The wave-to-wave interaction can significantly accelerate or delay fatigue crack propagation. In contrast, Mode III (tearing) loading has shown no wave-to-wave interaction. Very little research has been done to investigate damage accumulation in solder. The paper looks at single solder joints in shear. First the fatigue behavior of the solder joints is characterized using sine waves under displacement control. Then by looking at sums of waves with a constant displacement envelope, the damage from minor waves may be separated from damage from the major, envelope wave. The results show that the linear damage accumulation model (using rainflow analysis) is accurate as long as the minor waves remain in the elastic range. Once the minor waves begin to generate plastic deformation, the damage accumulates more rapidly than predicted.
5:05 pm INVITED
THERMOMECHNICAL FATIGUE TESTING OF SOLDER ALLOYS: Mark A. Palmer, P.E. Redmond, R.W. Messler, Jr., Materials Science and Engineering Department, Rensselaer Polytechnic Institute, Troy, NY 12180
Thermomechanical fatigue (TMF) is one of the most common sources of failure in solder joints. TMF occurs due to thermal cycling, as a cyclic stress is induced on the solder joint due to thermal expansion mismatch. Due to the complicated nature of thermomechanical fatigue, directly testing a material's resistance to TMF is not straight forward. A novel test apparatus has been developed which allows the direct measurement of stress due to thermal cycling. The TMF behavior of three alloys, eutectic Sn-Bi, Sn-Pb, and Sn-Ag as examined by this apparatus will be presented. The data will be compared with that generated by other test methods.
5:25 pm INVITED
EVALUATION OF INTERMETALLIC PHASE FORMATION AND CONCURRENT DISSOLUTION OF INTERMETALLIC DURING REFLOW SOLDERING: M. Schaefer, W. Laub, R.A. Fournelle and J. Liang, Materials Science Program, Marquette University, 1515 W. Wisconsin Ave., Milwaukee, WI 53201-1881; Allen-Bradley Company, 1201 S. Second St., Milwaukee, WI, 53204
Room: 340C
Session Chairs: J.P. Sullivan, J.A. Floro, Sandia National Labs, Albuquerque, NM 87185
2:00 pm INVITED
EVOLUTION OF GRAIN STRUCTURE IN THIN FILM REACTIONS: K. Barmak, Department of Materials Science and Engineering, Lehigh University, Bethlehem, PA 18015
The granular nature of polycrystalline thin films is playing an increasingly important role in their performance as the dimensions of the grains and those structures in which the thin films are used become comparable. Consequently, greater attention is being paid to the factors that affect the grain structure of thin films and its evolution. This paper will address the evolution of grain structure during the reaction of polycrystalline thin films. Experimental evidence from calorimetry, x-ray diffraction and transmission electron microscopy studies of a number of thin film systems will be reviewed and the role of reactant phase microstructure in these reactions will be highlighted. Theoretical models that combine nucleation and growth processes in the formation of the product phase will be presented and the impact of heterogeneous boundary nucleation on the evolution of grain structure in thin film reactions will be discussed.
2:40 pm
THE MICROSTRUCTURAL DEVELOPMENT OF THIN FILM COPPER GOLD ORDERED INTERMETALLIC COMPOUNDS: Jonathan Gorrell, Paul Holloway, Dept. of Material Science and Engineering, University of Florida, Gainesville, FL 32611-6400; Hal Jerman, EG&G IC Sensors, 1701 McCarthy Blvd., Milpitas, CA 95035
Recent developments in microelectromechanical systems (MEMS) have created a need for stronger metal thin films that are resistant to stress relaxation. Intermetallic compounds are noted for their strength and resistant to creep, but these properties have rarely been studied in thin films of intermetallics. We have sputter deposited layered structures of copper and gold so that the intermetallic compounds Cu3Au, CuAu and CuAu3 would form. The samples were initial heated to 475°C to allow intermixing of the copper and gold layers and the evolution of the intrinsic and extrinsic stresses were characterized during the process. The samples were then annealed at temperatures where the intermetallic phases would form. The intermetallic phases were tested for strength and stress relaxation using a Tencor Flexus 2320. The microstructure and composition were examined with TEM, EDX, Electron Microprobe, and X-Ray diffraction. This data allows us to relate strength and resistance to stress relaxation to microstructure and the heat of formation of the various copper gold intermetallic compounds.
3:00 pm
CHARACTERIZATION OF GRAIN BOUNDARIES IN Al INTERCONNECTS BY ORIENTATION IMAGING MICROSCOPY: C. Wu, C.L. Bauer, B.L. Adams, W.W. Mullins, Department of Materials Science and Engineering, Carnegie Mellon University, Pittsburgh, PA
Characterization of microstructure (grain orientation, grain-boundary inclination, microtexture, etc.) in thin-film interconnects is especially important because grain size usually approximates film thickness, thereby enhancing short-circuit diffusion induced by electric fields (electromigration), mechanical stress (stress voiding), temperature gradients (thermomigration), and capillarity (grain-boundary grooving). In this research, grain boundaries in polycrystalline Al interconnects have been characterized by orientation imaging microscopy and analyzed in terms of the interconnectivity of triple junctions. In general, results indicate that a variety of thermodynamic and kinetic properties can be extracted by rapid acquisition and processing of large data sets and correlated with the (five) degrees of crystallographic freedom of individual grain boundaries. Research supported, in part, by the National Science Foundation under Grant DMR-9319896.
3:20 pm
REAL-TIME MEASUREMENTS OF MICROSTRUCTURAL EVOLUTION IN Ag THIN FILMS ON SiO2: Eric Chason, Jerry Floro, Sandia National Laboratories, Albuquerque, NM 87185-1415; Steven C. Seel, Carl Thompson, Massachusetts Institute of Technology, Cambridge, MA
Understanding and controlling the microstructure in thin metal films used for IC interconnects is essential for maintaining high reliability. Although the microstructure of Al films has been heavily studied, the switch to Cu for interconnects in the future will lead to different microstructural evolution than in current technology. In order to study the kinetics of grain growth and texture evolution, we have developed an in situ X-ray system that can measure the concurrent evolution of (111) and (100) textured crystallites during annealing. We present results from the annealing of Ag films, which have similar elastic properties to Cu. This work was supported by the U.S. Department of Energy under contract DE-AC04.
3:40 pm BREAK
4:00 pm INVITED
STRESS RELAXATION AND THERMAL EVOLUTION OF FILM PROPERTIES IN AMORPHOUS CARBON: J.P. Sullivan, Sandia National Labs, Albuquerque, NM 87185
Large stress relaxation is observed in amorphous carbon films deposited by pulsed-laser deposition which are subsequently thermally annealed. In the as-deposited state, the films exhibit very high compressive stress, > 6 GPa, which has been thought to be either necessary or unavoidable in order to form a high percentage of 4-fold coordinated (diamond-like) carbon bonds and which can also hamper important electronic applications of these films. Stress measurements performed in situ and ex situ following thermal annealing up to 600°C indicate the stress may be reduced nearly two orders of magnitude. The stress relaxation is not dominated by interfacial relaxation nor is it accompanied by large scale changes within the film (e.g. graphitization) as indicated by in situ and ex situ electrical measurements, ex situ X-ray reflectivity, Raman spectroscopy, and experiments using different substrates. The bonding structure and mechanisms of thin film evolution in these unique amorphous films will be discussed. Work at Sandia was supported by an LDRD through the U.S. DOEunder contract no. DE-AC04-94AL85000 and by a CRADA.
4:40 pm
MICROSTRUCTURE AND TEXTURE DEVELOPMENT IN CUBIC BORON NITRIDE THIN FILMS: D.L. Medlin, P.B. Mirkarimi, G.F. Cardinale, K.F. McCarty, Sandia National Laboratories, Livermore CA 94551
Cubic boron nitride (cBN) is an sp3-bonded material with many properties and applications that are similar to diamond. Although cBN can be synthesized in bulk form at high temperature and pressure, synthesis in thin film form requires the simultaneous bombardment of the growing film with a high flux of energetic ions. cBN films grow with a unique, layered microstructure in which sp2-bonded graphitic boron nitride initially forms near the substrate interface, and nucleation and growth of the sp3-bonded cubic phase occurs further up in the film. Both the graphitic and cubic layers exhibit strong preferential crystallographic orientations: the graphitic layer possesses a strong in-plane [0002] orientation, whereas the cBN possesses an in-plane [111] orientation. This preferential orientation is consistent with an alignment between the cBN {111} planes and the basal planes of the layer of highly oriented graphitic boron nitride that forms in the initial stages of film growth. This relationship provides insight into the mechanisms controlling the initial nucleation of cBN and subsequent microstructural evolution of the films. This work is supported by the U. S. Department of Energy under contract DE-AC04-94AL85000 and in part by OBES-DMS.
5:00 pm
CHARACTERIZATION OF SrRuO3 THIN FILMS: MICROSTRUCTURE/PROPERTY RELATIONSHIP: F. Chu, Q.X. Jia, C. Adams, T.E. Mitchell, Materials Science and Technology Division, Mail Stop K 765, Los Alamos National Laboratory, Los Alamos, NM 87545; Q. Zhu, Physics Department, Brookhaven National Laboratory, Upton, NY 11973
Metallic oxide SrRuO3 thin films have been grown using pulsed laser deposition on LaAlO3 substrates at different substrate temperatures. The surface morphology and microstructural properties of the SrRuO3/LaAlO3 system have been studied using high resolution synchrotron x-ray diffraction, conventional x-ray diffraction, field emission scanning electron microscopy, and transmission electron microscopy. Electrical properties of SrRuO3 thin films with different microstructures have been measured. It is found that films deposited at 250°C are amorphous, and show semiconductor-like temperature dependence of electrical conductivity. Films deposited at 425°C are crystalline with very fine grain size (100~200Å), and show both metallic and semiconductor-like temperature dependence of the electrical conductivity in different temperature regions. Synchrotron x-ray diffraction and transmission electron microscopy unambiguously indicate that epitaxial [001] and [110] growth of the orthorhombic films takes place for deposition temperature above 650°C, where the [001] texture is dominant. Films deposited at 775°C show a resistivity of 280 mW-cm at room temperature. Microstructures of epitaxially grown films and possibilities for improving the thin film growth are discussed, based on geometric considerations for both film and substrate. The optimized deposition conditions to grow SrRuO3 thin films on LaAlO3 substrates have been found. Possible engineering applications of SrRuO3 thin films with different microstructures are discussed.
Room: 330E
Session Chairpersons: Chain T. Liu, Oak Ridge National Laboratory, Metals & Ceramics Div., PO Box 2008, Oak Ridge, TN 37831-6115; G.T. (Rusty) Gray, III, Los Alamos National Laboratory, Los Alamos, NM 87545
1:55 pm OPENING REMARKS
2:00 pm INVITED
THE ROLE OF GRAIN SIZE AND SELECTED MICROSTRUCTURAL PARAMETERS IN STRENGTHENING FULLY-LAMELLAR TiAl ALLOYS: D.M. Dimiduk, Wright Laboratories Materials Directorate, WL/MLLM, WPAFB, Dayton, OH 45433; P.M. Hazzledine, T.A. Parthasarathy, UES, Inc., 4401 Dayton-Xenia Road, Dayton, OH 45432; S. Seshagiri, SYSTRAN, Inc., 4126 Linden Ave., Dayton, OH 45432
More than five years ago wrought processing was first used to produce fully-lamellar microstructures in TiAl alloys having grain sizes less than ~400mm. These alloys exhibit an improvement in overall balance of properties, especially at high temperatures. More recently, such microstructural forms led to exceptional yield strengths (>1000Mpa at low temperatures), while maintaining attractive high-temperature properties. The improvements appeared to be attained principally through an unusually high apparent sensitivity of strength on grain size. Studies reported an apparent value for the slope of the "Hall-Petch plot" approaching 5 Mpa
for fully lamellar gamma alloys, while that for single phase or duplex microstructures is near unity. The present investigations examine the slope of the Hall-Petch plot for fully lamellar microstructures, paying particular attention to the lamellar microstructural variables. Results show that 
2-lamella thickness and spacing, and 
-lamella thickness can vary over more than two orders-of-magnitude. These in turn influence the value of 
o in the Hall Petch plot, and often change concomitantly with grain size in processing. Nonetheless, the corrected value of the Hall-Petch constant exceeds a magnitude of 2 Mpa; and it is strain dependent. The investigations also examine dislocation activity, flow behavior, glide barriers and slip multiplicity for polysynthetically twinned or PST crystals (single-grain analogue of fully lamellar material), then map this behavior into an explanation of the yield behavior of high strength fully lamellar gamma alloys.
2:30 pm INVITED
BULK, DEFECT, AND INTERFACIAL PROPERTIES OF TiAl and Ti3Al: M.H. Yoo, C.L. Fu, Metals and Ceramics Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831
Available theoretical and experimental data on the elastic constants, point defect properties, shear fault energies, and ideal cleavage energies of TiAl and Ti3Al are reviewed, including the recently calculated / and 2/ interfacial energies. Enhanced slip by ordinary dislocations along pseudo-twin and rotational / and the 2/ interfaces is a contributing factor to the strong plastic anisotropy of a fully lamellar microstructure. According to the calculated interfacial fracture energies, cleavage cracking is to occur on 2/ boundaries and the least likely on true-twin boundaries. The roles of misfit dislocations, kinetics of dislocation-interface interactions, and hydrogen embrittlement in deformation and fracture of two-phase TiAl alloys are discussed. This research was sponsored by the Division of Materials Sciences, U.S. Department of Energy, under contract number DE-AC05- 96OR22464 with Oak Ridge National Laboratory, managed by Lockheed Martin Energy Research Corporation.
3:00 pm
INTERNAL STRESSES IN LAMELLAR TiAl: P.M. Hazzledine, UES, Inc., 4401 Dayton-Xenia Road, Dayton, OH 45432; M.A. Grinfeld, Department of Mechanical and Aerospace Engineering, Rutgers University, P.O. Box 909, Piscataway, NJ 08855; D.M. Dimiduk, Wright Laboratories Materials Directorate, WL/MLLM, WPAFB, Dayton, OH 45433
Internal stresses necessarily develop in multilayered coherent materials with mismatched lattice parameters as the individual layers adjust to their neighbors. In Ti-Al the layers consist of 2 plates and six orientation variants of gamma plates which occur in an apparently random sequence. The internal stresses in any one layer are affected by the orientations of all its neighbors. We have calculated, in closed form, the stresses present in one layer of a stack of gamma layers in which the sequence of variants is in any order and of any length and in which the thicknesses of each layer may take any value. In addition we have calculated the stresses in a stack of alternating 2 and layers of arbitrary thicknesses. The results are applicable directly to an idealized form of PST TiAl in which the domain size is large and the lamellae thickness is small. The relaxation of internal stresses which occurs in thick lamellae and by the formation of domains is discussed. The consequences of both these relaxations to the plastic strength of PST TiAl are examined.
3:20 pm
RELATIONSHIPS BETWEEN MICROSTRUCTURE/COMPOSITION AND MECHANICAL PROPERTIES OF GAMMA TITANIUM ALUMINIDES: W.O. Soboyejo, Y. Ni, C. Mercer, Dept. of Materials Science and Engineering, Ohio State University, 2041 College Road, Columbus, OH 43210; A.B.O. Soboyejo. Dept. of Aerospace Engineering, Applied Mechanics and Aviation, Ohio State University, 155 W. Woodruff Ave., Columbus, OH 43210; R. Armstrong, Dept. of Mechanical Engineering, University of Maryland, College Park, MD 20742
Quantitative relationships will be presented for the prediction of basic mechanical properties (yield/ultimate tensile strength, fracture toughness and plastic elongation to failure) at room- and elevated- temperature. These include Hall-Petch relationships between basic mechanical properties and the average equiaxed grain size/lamellar packet size. Multiple linear regression expressions are also presented for the prediction of the effects of grain/packet size, lamellar volume fraction and composition. The observed grain size dependence of mechanical properties is shown to be consistent with Hall-Petch exponents predicted for dislocation/dislocation and dislocation/grain boundary interactions in duplex 2 + alloys. The implications of the empirical relationships are assessed for the engineering of balanced mechanical properties in gamma titanium aluminides.
3:40 pm BREAK
3:50 pm INVITED
TEMPERATURE DEPENDENT DEFORMATION IN GAMMA TITANIUM ALUMINIDES: S.H. Whang, Z.M. Wang, Q. Feng, C. Wei, Polytechnic University, Six Metrotech Center, Brooklyn, NY 11201
Deformation of gamma titanium aluminides may be best described by the lack of sufficient ductility at RT as well as intermediate temperatures, and the anomalous yielding at high temperatures. Such deformation behavior is mainly attributed to unusual behavior of two independent slip systems: superdislocation slip and ordinary dislocation slip, both of which simultaneously operate over the entire temperature range in the polycrystalline alloys. For this reason, the study on the role of individual slip system is required to carry out deformation experiments with single crystals of a -TiAl alloy. In -Ti-56Al, the yield stress curves of superdislocation slip show positive temperature dependence over two distinct temperature regimes while those for ordinary dislocation slip exhibit a single temperature range. These distinct temperature regimes can be explained with the characteristic features of the dislocation structures which were obtained from the postmortem examinations of the deformed specimens conducted by TEM. The thermodynamic aspects of both cross-slip configurations of superdislocations will be presented based on anisotropy strain energy calculations. Finally, the CRSS values depend on deformation orientation in superdislocation slip, in consistent with the cross-slip models. The details of the orientation dependence will be discussed.
4:20 pm
PLASTIC INSTABILITIES IN 
-TiAl AND THEIR RELATION TO WORK HARDENING AND TOUGHNESS: François Louchet, Groupe Physique du Metal, LTPCM- UMR.CNRS/INPG/UJF, B.P. 75, Domaine Universitaire, 38402 - St Martin d'H'eres, France
Flow stress anomaly in -TiAl is associated, as in many other intermetallics, to a strain rate sensitivity (SRS) close to zero, to plastic instabilities, and to a high work hardening rate (WHR). In situ TEM observations also show a jerky motion of groups of cusped ordinary screw dislocations (1), in agreement with macroscopic instabilities. The present paper investigates these instabilities in the framework of the Local Pinning Unzipping (LPU) model (2), in which the flow stress anomaly and the low value of the SRS are predicted as a consequence of a balance between dislocation exhaustion and multiplication rates. This balance is applied here to groups of N dislocations, and results in a non linear equation in N, from which the minimum stress required to deform the material at a given strain rate can be determined. The corresponding average value of N is obtained as a function of temperature. The consequences of the temperature dependence of N on both WHR and toughness are discussed. Possible consequences on the particular dramatic brittleness of PST materials loaded perpendicular to lamellar boundaries (3), in which slip in -TiAl takes place across lamellae, can also be contemplated as a function of these instabilities, and particularly in terms of the variations with temperature of the number N of dislocations in pile-ups.
(1). S. Farenc and A. Couret, Mat. Res. Soc. Symp., 1993, vol.288, p.465.
(2). F. Louchet et B. Viguier, Phil. Mag. A , 1995, vol.71, p.1313.
(3). M. Yamagushi, H. Inui, K. Kishida, M. Matsumoro and Y. Shirai, Mat. Res. Soc. Symp., 1995, vol.364, p.163.
4:30 pm
MORPHOLOGICAL ASPECTS OF 1/2<110] UNIT DISLOCATIONS IN AND (
2 +) TiAl ALLOYS: Sriram Seshagiri, SYSTRAN Corporation, 4126, Linden Ave., Dayton, OH 45432; Dennis M. Dimiduk, Materials Directorate, WL/MLLM, WPAFB, OH 45433; Vijay K. Vasudevan, Dept. of Materials Sci. & Eng., Univ. of Cincinnati, Cincinnati, OH 45221
The b=1/2<110] unit dislocations are a dominant part of the deformation substructure in both single phase () and two-phase (