Residual stresses are common in engineering materials. They are elastic stresses that exist in the absence of external forces and are produced through the differential action of plastic flow, thermal contraction, and/or changes in volume created by phase transformations. Differential plastic flow can occur during the forming of a part. For example, the grinding of a surface will plastically extend a thin surface layer relative to the underlying material through frictional forces. This tends to throw the near-surface region into compression, which is balanced by a tensile stress through the bulk of the part. In such a situation, the requirements of force equilibrium lead to a rather high level of compressive stress in the near-surface region and a low level of tensile stress through the bulk. Differential thermal contraction often occurs during the nonuniform cooling of a large part, in the vicinity of welds, and between the matrix and reinforcement phases of a composite. Differential volume changes occur during the precipitation of second phases, i.e., the atomic volume of the precipitating phase generally differs from that of the host matrix. If it is larger, the second phase is placed in compression, with the matrix in tension, and vice versa.

The residual stress states discussed in the previous paragraph are of two basic types: macrostress and microstress. Residual macrostresses are long-range relative to the scale of the microstructure––that is, they extend continuously across a part. Residual stresses arising from forming and joining are of this type. They generally vary with position and are extensive in nature. A plate with compressive residual stresses on the flat surfaces will deflect if the compressive region on one surface is removed. Destructive stress measurement techniques such as hole drilling or strain gauging and sectioning can be used to determine residual macrostresses.

Superconductivity in potassium-doped BaBiO3 was first observed by Mattheiss et al. in 1988 and the structure of the superconducting material was determined by Cava et al. The material crystallizes in a simple perovskite structure with potassium substituted on the barium site as shown in Figure 1. Unlike the layered copper-oxide materials, this compound is simple cubic and thus isotropic in the superconducting state and shows no magnetism. Band theory calculations show that the electronic structure is simple, with conduction bands composed of Bi(sp)O(p) hybridized orbitals and, unlike the copper-oxide materials, there are no d electrons at the Fermi surface or involved in the superconductivity. Photoemission and inverse-photoemission agree well with the theoretical calculations and show metallic behavior with a low density of states at the Fermi level consisting of Bi-O sp hybrids. This material is structurally and electronically simple compared to the copper-oxide compounds and, therefore, should be much easier to investigate experimentally and understand theoretically. This article discusses recent results concerning the structural properties of the material and describes several experiments that give information on the superconducting state.

Structural ceramics are necessarily polycrystalline and their usefulness is largely determined by the interfaces between the grains. The relationship between the structure and chemistry of different interfaces and the micro-structure can be illustrated by reviewing studies of interfaces in a wide range of materials including such classical ceramics as Al2O3, the current “hightech” polyphase ceramics exemplified by ZrO2-toughened Al2O3, and the composite materials of the future. Using transmission electron microscopy is essential for a complete understanding, but limitations to its use must be recognized. Only by understanding the factors that control the behavior of these interfaces will it become possible to further extend the application of interface engineering.

Structural ceramics are a group of materials that can be used for applications requiring their strength to persist at high temperatures or in conditions that would be particularly corrosive to alternative materials, which are usually metallic. Strength and strength-related properties such as toughness depend largely on the microstructural features of the processed material.

The microstructure is defined by the morphology and size of the grains and the interfaces between these grains. If the grains are in intimate contact, then the interface is a grain boundary of the type familiar from studies of metals.

Depending upon the nature of the application, three materials currently dominate the optical substrate market today. Polycarbonate (PC), the most commonly used material, is typically used in disks with diameters up to 120 mm. For larger diameters, the inherent birefringence exhibited by PC interacts with the read laser and results in errors in the read-back signal. For 200 mm and greater, both polymethylmethacrylate (PMMA) and glass are used. Because PMMA is low in inherent birefringence, it is widely used for 200 mm video disks, an application which takes advantage of the manufacturing economies of the injection molding process. For larger diameter data storage disks, however, glass is used because it does not have the disadvantage of relatively high moisture absorption or the low heat resistance of PMMA.

This article will deal with the state-of-the-art in materials used for polymeric substrates. It will focus primarily on polycarbonate and the most significant substrate property currently confronting the field, namely birefringence. Other constraints placed by optical storage technology on the substrate material will be discussed along with new materials and research activities aimed at circumventing some of the current limitations.

The low atomic number, good chemical reistance, and low thermal expansion of carbon foams make them attractive for many specialty applications. In general, we are interested in carbon LDMMs that have a maximum cell size of 25 μm and densities less than 0.05 g/cm3. These requirements are beyond the state of the art for most commercial carbon foams; thus, the DOE National Laboratories have concentrated on producing materials that meet these demanding requirements using various phase-separation, sol-gel, and replication processes.

In all cases, the final carbon foam is derived from a polymer precursor. Most polymers (e.g., polymethyl methacrylate) simply depolymerize and revert to their monomeric form when heated to high temperatures in an inert atmosphere. Cross-linked phenol-formaldehyde and polyfurfuryl alcohol are polymers that leave substantial carbon chars (greater than 50 wt%) after pyrolysis near 1000°C. Polyacrylonitrile also leaves a large carbon char (˜40 wt%)) after partial oxidation at 200°C followed by heat treatment at 1000°C. The latter char is a partially ordered graphite while the former chars are totally amorphous.

The microstructure and properties of carbon LDMMs depend on synthetic procedure and heat treatment conditions. We have been able to make carbon LDMMs with cell sizes as small as 100 Å or with densities as low as 0.015 g/cm3. Furthermore, surface areas of 50–1000 m2/g have been obtained with microstructures ranging from ribbonlike webs to interconnected colloidal-like particles. The ability to control these various parameters presents new opportunities to design carbon LDMMs for high technology applications, examples of which include sensors, electrodes, and high temperature filters.

The rapidly growing demand for magnetic information storage systems at lower cost and higher capacity has driven the magnetic recording industry to accelerate development of thin film magnetic recording media. The ultimate recording density depends on the bit transition lengths, as shown in Figure 1 for various media, and on signal-to-noise ratio, which is proportional to the number of particles per bit. Longitudinal thin film media bit transition lengths are limited by thickness, magnetic properties, and zig-zag magnetic domain microstructures at the transitions, whereas in perpendicular media the principal limitation is the diameter of the grains. The recent development of quasi-particulate longitudinal and perpendicular thin film media indicates that areal bit densities substantially exceeding 109 bits/inch2 will be achieved in the future.

This article will review magnetic phenomena relevant to thin film recording media, describe macromagnetic and micromagnetic characterization techniques, survey the development of state-of-the-art thin film media, correlate micromagnetics with magnetic properties, media noise, and recording performance, and discuss future requirements of longitudinal and perpendicular thin film recording media.

Ferromagnetic and ferrimagnetic thin films exhibit a nonlinear, multivalued, and hysteretic relationship between the magnetic dipole moments per volume or magnetization M and the magnetizing field intensity H as shown in Figure 2 for the component of M in the direction of H.

In the early 1950s when “materials science” was beginning to take shape in the minds of educators in materials departments, discussions were heated on the subject of how (and whether) intellectually rich courses could be developed with such broad coverage. It was argued by many that materials are too complex and vary too greatly from one another in their properties and in their applications to be treated in a single course. These individuals argued that if “materials” was to be taught, then it would have to be in courses or segments of courses broken down by materials classes-metals, ceramic, polymers, semiconductors.

A full generation of faculty has passed through our ranks since those days, and the arguments regarding teaching of at least the beginning materials science subjects are now muted and perhaps moot. Few materials departments begin today with a materials-specific subject (e.g., metallurgy, ceramics) for either their own students or as a service subject for other engineering departments. Most begin with a subject in materials science or materials science and engineering that deals generically with all materials for at least a major portion of the subject. Examples are drawn from individual materials classes, and emphasis may shift to individual materials classes as the subject progresses.

The key to development of these subjects, and the intellectual foundation on which they rest, is structure and structure-property relations. We can understand, and teach, how the building blocks of materials (atoms, molecules, grains, amorphous phases, etc.) fit together to build macroscopic structures.

The investigation of structure-property correlations is a rather complex endeavor not only because interfacial Systems are intrinsically inhomogeneous, with chemical composition and physical properties differing from the surrounding bulk material, but also since three different aspects of the geometrical structure are involved — namely the macroscopic, microscopic, and atomic structures. As outlined in the Guest Editors' introduction, in addition to the choice of the materials which form the interface, five macroscopic and three microscopic degrees of freedom (DOFs) are needed to characterize a single bicrystalline interface. The importance of the atomic structure at the interface as well as the local interfacial chemistry, extrinsic (i.e., impurity segregation) or intrinsic (for example, via interfacial reactions or space-charge phenomena), greatly add to the task's complexity.

Grain boundaries (GBs) in pure metals represent ideal model Systems for investigating the strictly geometrical aspects of structure-property correlations for the following three reasons. First, the complexity due to the myriad of possible choices of materials combinations forming the interface is avoided, enabling a focus on the different roles of the three distinct geometrical aspects of the structure. Second, because GBs are bulk interfaces, dimensional interface parameters (such as the modulation wavelength in strained-layer superlattices, or the thickness of epitaxial layers) do not enter into the problem. Finally, the GB energy is thought to play a central role in various GB properties, such as impurity segregation, GB mobility and fracture, GB diffusion and cavitation, to name a few. A better understanding of the correlation between the structure and energy of GBs, therefore, promises to offer insights into more complex structure-property correlations, as well.

Many advanced technologies, such as surface science, semiconductor processing and high energy physics, call for vacuum levels of the order of 10−11 mbar and lower. These pressures can not be reached without a careful choice of materials, treatments, and evacuation means for the vacuum device involved. Non-evaporable getters (NEGs) are increasingly being recognized as an interesting and powerful solution for many vacuum problems. NEGs have been used extensively in sealed-off devices such as microwave tubes, traveling wave tubes, x-ray tubes, lamps, and infrared detector dewars, in which their main role is to assure the desired vacuum level throughout the life of the sealed device. The getter material can be considered as a chemical pump which removes the active gases in the residual atmosphere of the vacuum device by forming stable chemical compounds.

The choice of materials, treatments, and structures of nonevaporable getter materials is critical for the optimization of the sorption and diffusion processes which are the basis of the NEG pumping mechanism. The effectiveness of this pumping mechanism at very low pressures, and the cleanliness and simplicity of operation have made this pumping approach ideal, in combination with other pumping technologies, for reaching the extreme high vacuums today's advanced technologies require. This article will explain the mechanism of the gettering process, describing materials, treatments, and structures used in standard vacuum practice, and will review some of the most typical and interesting applications.

Since the discovery of high temperature superconductivity in layered copper-oxide compounds in the latter part of 1986, an enormous amount of research has been carried out on these remarkable materials. Prior to 1989, the prevailing view was that the charge carriers responsible for superconductivity in these materials were holes that move through conducting CuO2 planes. The CuO2 planes are the basic building blocks of the crystal structures of all the presently known oxides with superconducting critical temperatures Tc greater than ~30 K. Recently, new superconducting materials have been discovered in Japan and the United States in which the charge carriers involved in the superconductivity appear to be electrons, rather than holes, that reside within the conducting CuO2 planes. These findings could have important implications regarding viable theories of high temperature superconductivity as well as strategies for finding new high temperature superconductors.

The new electron-doped materials have the chemical formula Ln2-xMxCuO4-y and exhibit superconductivity with superconducting critical temperatures Tc as high as ~25 K for x ≍ 0.15 and y ≍ 0.02. Superconductivity has been discovered for M = Ce and Ln = Pr, Nd, Sm, and Eu, and for M = Th and Ln = Pr, Nd, and Sm. A related compound with the identical crystal structure, Nd2CuO4-x-y Fx, has also been found to display superconductivity withTc ≍ 25 K. Recently, it has been observed that superconductivity with Tc ≍ 25 K can even be induced in nonsuperconducting Nd2-xCexCuO4-y compounds by substituting Ga or In for Cu. Thus, it appears that the CuO2 planes can be doped with electrons, rendering the Ln2CuO4-y parent compounds metallic and superconducting, by substituting electron donor elements at sites within, as well as outside, the CuO2 planes; i.e., by substituting (1) Ce4+ or Th4+ ions for Ln3+ ions; (2) F1- ions for O2- ions; and (3) Ga3+ or In3+ ions for Cu2+ ions.

The commercial life of optical disks started in the late 1970s with prerecorded, read-only video disks. The video information was stored on the disks in the form of pits with varying lengths aligned in a spiral track, a structure which lends itself to easy mass replication. The pits were recorded in a “master” disk by exposing a positive photoresist to a focused beam of blue or UV light from a gas laser and subsequently developing the resist. Now, some 15 years later, mass-replicated, read-only video and audio disks still form the most important application of optical disk technology. Today a large majority of the corresponding masters are recorded using virtually the same technique.

In view of the relative importance of prerecorded disks, it is amazing that so little seems to have changed in the mastering processes. Part of the explanation lies in the fact that the dimensions of the relief structures on the early optical disks were defined by what a player could read using relatively cheap optics, rather than by what could be made in the mastering process. If in the beginning the analog video disk system (Video Long Play or LaserVision) was rather demanding, the digital audio disk (compact disk) (see Figure 1a), introduced later, was a much easier system from a mastering point of view. In that period a trend toward cheaper and faster mastering techniques developed, resulting in a few proposals for direct-effect and direct metal mastering. However, in the meantime very high-quality mastering became necessary for the CD-video system and for preformatted data disks. In the CD-video system, the combination of digital audio and analog video leads in some cases to pits as short as 300 nm (see Figure 1b). On preformatted data disks, wide shallow grooves may have to be combined with narrow deep header pits (Figure 1c).

The synthesis of nanometer-sized atomic clusters of metals and ceramics by means of the gas-condensation method, followed by their in situ consolidation under high-vacuum conditions, has resulted in a new class of ultrafine-grained interface materials. These nanophase materials, with average grain sizes presently ranging from about 5 to 25 nm, exhibit properties that are often rather different and improved relative to those of conventional materials. In addition, their processing characteristics appear in some cases to be greatly improved over their conventional coarser grained counterparts. The synthesis of nanophase materials by means of cluster assembly under controlled conditions should enable the design of materials, heretofore unavailable, with improved or unique properties. As such, it is likely that the combination of new capabilities to synthesize, characterize, anc) engineer the properties of materials based on the assembly of atomic clusters will significantly impact materials science and engineering in the coming years.

The assembly of matter by the consolidation of gas-condensed atomic clusters is probably as old as the universe itself, since this is thought to be the way condensed matter formed during the cooling period that followed the “big bang,” as evidenced by the structure of the earliest meteorites. The modern synthesis of ultrafine-grained materials by the in situ consolidation of nanometer-sized gas-condensed ultrafine particles or atomic clusters, however, was first suggested by Gleiter.

In an alloy, the equilibrium composition in the vicinity of an inhomogeneity, such as a surface or a grain boundary, will generally differ from the composition in the bulk. This was first recognized by Gibbs. This change in local composition can potentially affect the structure and mechanical properties of the boundary. As an example, experiments by Sass and co-workers have shown that the Burgers vector of the dislocations present in Fe-Au twist boundaries depends on the Au concentration at the boundary. This demonstrates that the structure does, in fact, depend on the local composition of the alloy at the interface.

This article presents computer simulation studies of equilibrium segregation in binary alloys at grain boundaries, and two experimental studies of interfaces using atom-probe field-ion-microscopy (APFIM). These studies address two fundamental questions: What is the composition of the boundary region as a function of the bulk composition and temperature? What is the spatial distribution of the two species in the boundary? Note, that the comparison of experiment and calculations, where possible, is crucial. Such comparisons can provide guidance in the experimental analysis, and the comparison also provides a stringent test of the reliability of the simulation methods.

This article first describes simulation and atom-probe field-ion microscope techniques and next discusses the segregation in Ni-Cu alloys to isolated edge dislocations. This is followed by a study of segregation at twist boundaries in Pt-Au alloys. Then, the segregation at an asymmetric mixed tilt-twist boundary in a Pt-Ni alloy is computed, and the results are compared with experimental atom-probe held-ion-microscopy results. Finally, the results of an atom-probe study of two-dimensional Re segregation at a twist boundary with a small tilt component in a W-Re alloy are presented.

The previous sections of this article described the synthesis, morphologies, and properties of a variety of low-density microcellular materials. This section discusses several of the analytical methods used and developed at the DOE laboratories to characterize these state-of-the-art materials.

In some LDMM applications, quantitative measurements of the material's average cell size and cell size distribution are desired. Indeed, the term “microcellular” has little meaning without such information. As seen throughout this article, however, most LDMMs do not have a readily defined cellular character. The more general problem is to quantify the spatial scale(s) of the foam. For this purpose it is necessary to define one or more “measures” of the spatial scale. The possibilities are many and include not only single numbers (e.g., cell size and cell size standard deviation, where “cell size” is meaningful) but also functional descriptions (e.g., correlation functions).

SEM provides direct images and, therefore, is the most popular technique for examining LDMM morphology. SEM, however, suffers from at least three limitations: (1) SEM examines only a very small volume of material, and thus is impractical for obtaining average morphological properties; (2) SEM requires that nonconductive LDMMs be coated, a process step that can alter the structure and introduce artifacts (particularly with delicate structures); and (3) SEM images are only two-dimensional projections of real three-dimensional structures.

Materials science is fundamentally an interdisciplinary field. For purposes of discussing undergraduate preparation for work in materials science, I think it useful to take chemistry, physics, and materials science and engineering as three more-or-less separate disciplines which combine to form the overall field of materials science. The primary reason for this particular taxonomy is pragmatic rather than philosophical. Undergraduate students choose major fields of study on the basis of disciplinary boundaries. Thus, in thinking about undergraduate preparation for work in the overall field, analysis of the present situation and/or recommendations for change must revolve around that reality.

The recent report entitled Materials Science and Engineering for the 1990s (the MS&E Study), sets forth the four elements of materials science and engineering as “structure and composition, properties, performance, and synthesis and processing.” An examination of these specific elements permits us to make useful distinctions among the three disciplines that combine to form the field of materials science. For example, while input from the point of view of physics certainly can contribute rather directly to expansion of our knowledge in the first three areas, its possible contribution to the last is, at best, indirect. To somewhat belabor the point, the research field of condensed matter physics is certainly contained within the field of materials but arguably not part of the discipline of materials science and engineering.

The MS&E Study includes a chapter entitled “Manpower and Education in Materials Science and Engineering.” Within that chapter is a section called “Undergraduate Education in Materials Science and Engineering.”