The following is an edited version of NASA Astronaut Bonnie J. Dunbar's Plenary Address at the 1990 MRS Fall Meeting.

I've often been asked, “Why a materials scientist in space?” I hope my presentation tonight will provide a little more insight on “why.” I don't think it's unusual at all. I look at myself as a continuum between outer space and inner space. I've been equally fascinated looking at the stars as at atomic structures under scanning electron microscopes. So to me it's not an unusual experience. In addition, materials scientists and engineers are needed to operate in space.

Tonight I would like to do several things. I'd like to talk to you about where we've been in space, where we are, and where we hope to go. I'll emphasize the materials aspect of it, not only in how we achieved space, but how we hope to exploit the microgravity environment of space. I have with me “home movies” of a Shuttle Mission I flew on in January 1990. You'Il share with me the experience of performing research as well as retrieving the long-duration exposure facility (LDEF) with a robotic arm.

Several years ago, when my boss came to my office and said that he wanted me to manage the tribology program at Argonne National Laboratory, I was mystified. My first response was, “What is tribology?” In the years since, the cards have been somewhat reversed. The initial response of person after person when I mention my position as Tribology Section Manager of the Materials and Components Technology Division is to look blank and pose the same question to me. Having somewhat of a playful nature, I often simply try to drop the term very nonchalantly and continue speaking, taking great delight in the puzzled looks my approach engenders. When somone eventually injects the inevitable inquiry, I give a rather stilted answer, such as:

“Tribology is the science and art of interacting surfaces in relative motion with respect to one another.”

Since this succeeds in erasing only a portion of their curiosity, I eventually have pity on them and say, “It's really the rather mundane areas of friction, lubrication, and wear.” I have often pondered and discussed with colleagues whether we did ourselves any favor by coining the word in the first place, but it does serve a useful purpose in at least attempting to obtain a degree of helpful communication between people engaged in the various aspects of this broad area of endeavor.

Biomedical materials is an embryonic interdisciplinary science whose practitioners are scientists, engineers, biochemists, and clinicians who use synthetic polymers, metals, ceramics, inorganic, and natural polymers to fabricate artificial organs, medical devices, drug delivery systems, prosthetics, and packaging systems.

The intent of this special issue of the MRS Bulletin is to provide readers with insight into current biomaterials research and product development. This issue is not meant to be either conclusive or definitive, but rather a “sound bite” of the field.

For further information, please feel free to contact either the individual authors or the editors of this issue.

The topic that I am attempting to cover easily requires a whole book to do it justice. To deal with it in a short article, I will focus on a few basic principles that dominate the behavior of defects in oxides and then present some examples taken from well-studied and relatively well-understood materials.

The study of defects in oxides differs from the study of defects in metals and alloys or in semiconductors in several important ways. First, oxides generally have very high melting points so that intrinsic point defect concentrations at temperatures below ~1500°C are usually negligibly small. Second, an oxide does not always maintain the composition at which it was fabricated, as alloys do, but can exchange oxygen with the ambient atmosphere, according to the ambient partial pressure of oxygen, P(O2). Thus, the stoichiometry, and accordingly the defect concentration, can be controlled by annealing in a specific P(O2) atmosphere. Third, and most important, oxides have a high degree of ionicity so that most point defects possess an effective charge. The consequences of this fact are far reaching; some of them are as follows:

1. In a binary oxide, e.g. a metal oxide (MO), there are two sublattices. The diffusion of cations and of anions proceeds each on its own sublattice, involving different point defects. Accordingly, cation and anion diffusion rates can be very different and, in fact, usually are.

The rheology of dispersions often appears to be extraordinarily complex since these materials traverse all states of matter from simple Newtonian fluids to Hookean solids. Highly nonlinear viscoplastic or viscoelastic responses, with persistent time dependence, often frustrate attempts to address the mechanics without attempting to characterize the composition or understand the thermodynamics. The difficulty lies in the sensitivity of the stresses to the microstructure, which in turn depends on the interparticle forces and the deformation history.

However, combining rheological measurements with other probes of the structure can permit one to anticipate the response under different conditions and, sometimes, to elucidate the relevant forces. Even better, one may learn to manipulate the composition to obtain the rheology desired for a specific process or a particular product, such as thickeners for coatings. The key lies in linking the composition, interparticle forces, and rheology. This review illustrates the current knowledge of those relationships and identifies some sources, from which the bulk of the content comes for those who wish to pursue the subject further.

This brief treatment first surveys phase behavior and interparticle forces. These concepts, combined with dimensional analysis and simple ideas from statistical mechanics, motivate correlations of data for well-characterized, stable dispersions and both weakly and strongly flocculated dispersions.

The transmission electron microscope (TEM) has had a major impact on materials science in the last five decades, despite the fact that it is necessary to prepare thin samples in order to use the technique. The primary reason for this effectiveness is the ability to access both real space and diffraction data in the same instrument, and to filter in one and observe the effect in the other. This is possible because of the wave nature of electrons and the existence of effective magnetic lenses for focusing. Abbe showed that any lens has the ability to Fourier transform its input wavefield in its focal plane, and to provide a second Fourier transform in the image plane. This is schematically shown in Figure 1. A crystalline object will diffract only in certain directions, with Bragg angles (θB) depending on the inverse of the interplanar spacing. The diffraction pattern is a series of spots in the Fourier, or focal, plane of the lens. A filter placed in the focal plane serves to limit the resolution by limiting the bandwidth of the image, but it also can serve to select certain parts of the Fourier spectrum in the image. The simplest examples of this, as used in optical microscopy, are bright-field and dark-field imaging. In the former the un-scattered beam is allowed to reach the image, in the latter it is not.

At first sight it may seem self-evident that epitaxial crystals must grow with their atomic planes parallel to the lattice planes of the substrate crystal. The simplified sketch in Figure 1a shows atoms raining onto a substrate from a molecular beam, and then diffusing to step-edges in a desirably controlled growth process termed step-edge (or ledge) flow. Drawn this way, the conclusion that the planes must lie parallel seems obvious. If, however, the process is heteroepitaxial, and the new lattice (solid circles) differs in spacing or structure from the substrate template (lines), the outcome is less straight-forward. Accommodating the difference of step heights at the interface then requires local tilts amounting to a fraction of one atomic spacing in, say, 25 atoms or perhaps half a degree, depending from one point to the next on details of the neighboring interfacial structure.

In what follows I describe a qualitatively different and newly discovered behavior we term coherent tilt. The coherent tilt process causes the atomic planes of the epilayer to grow at an accurately predictable and reproducible angle with respect to the substrate planes. The tilt axis, tilt angle, and sense of rotation remain precisely fixed over an entire substrate, with the result that the process occurs coherently over the dimensions of a macroscopic sample. Large tilts over 7° have already been produced with the angle predictable from first principles to 0.1° The characteristics of coherent tilt are sketched in Figure 1b.

The Biomaterials Industry Subpanel was chartered by the National Research Council (on behalf of the National Academies of Sciences and Engineering) to address the needs and opportunities in materials science and engineering as perceived by the biomaterials industry. This report represents an initial overview and should not be considered definitive.

The Committee examined the short-term, intermediate, and long-term needs of the industry and how external factors such as regulations, lack of standards, and international competition influenced the industry. The industry is heterogeneous and was subsequently defined by the following market segments: artificial organs, biosensors, biotechnology, cardiovascular/blood products, drug delivery, equipment/devices, maxillofacial, ophthalmology, orthopedics, packaging, and wound management.

Each of these market segments then addressed the:

Role of materials in the industry,

Current materials and material needs,

Material opportunities and impact,

Industrial needs/issues,

International competition/foreign initiatives, and

Role of the U.S. government.

In spite of its importance for extrusion and molding, the rheology of polymer melts under high shear rates is not fully explained by the classical theory of Doi and Edwards. In this theory, each polymer chain is trapped in a confining “tube.” When the tube is suddenly stretched, the chain follows at first, but it soon shrinks back to its original length (“retraction process”). I propose a different view: Upon stretching, the tube becomes thinner so the chain is squeezed and tends to span a longer length of tube; there is no retraction.

Molecular tribology—the study of tribology at atomic and molecular scales—constitutes a new frontier of tribology research. In a major surge of activity, experimental methods have recently been developed to measure dynamic interfacial forces in shear. Building partly on earlier, somewhat neglected friction studies, striking new findings have been obtained. The new methods include the surface forces apparatus for measuring adhesion and static interfacial forces as a function of surface separation, new molecular tribometers for measuring friction in shear, atomic force microscopy, use of UHV tribometers, and the quartz-crystal microbalance. Theoretical calculations and molecular dynamics simulations are also emerging for friction in dry and lubricated sliding.

On the scientific side, appreciation is growing that scientific understanding of these systems, so complex and so far from equilibrium, is possible. Tribology is becoming recognized as an area with many exciting and useful surface science opportunities.

The engineering significance is that while tribology design and tribology-based applications are rooted in our economic life, too often the technologies and formulations are empirically derived. One tends to take friction, wear, and tear for granted. A scientific understanding is needed so that better design can emerge by rational extension.

This review seeks to bring out the excitement of new developments. The reader is referred to the original literature for full accounts.

The key to controlling and predicting the properties of metal matrix composites lies in understanding and controlling the interface. When properly designed, the interface between reinforcing fibers and the matrix or protective coating can act as a mechanical fuse through a controlled delamination mechanism.

Controlled delamination, in effect, results in the decoupling of fibers from early damage due to stress concentrations in the vicinity. The delamination event must precede the crack bridging and frictional pull-out mechanisms that have been so effectively demonstrated in ceramic matrix composites. The delamination event, therefore, is the necessary precondition, and so analysis of composite toughening must start with a definition of the conditions for interface debonding.

This decoupling can be expressed in terms of cohesive strength of the interface, shear strength of the interface, and fiber fracture stress. In a related but alternative manner, debonding can be expressed in terms of the intrinsic work of fracture of the interface as compared to the transverse work of fracture of the fiber.

Semiconductor devices generally contain n- and p-doped regions. Doping is accomplished by incorporating certain impurity atoms that are substitutionally dissolved on lattice sites of the semiconductor crystal. In defect terminology, dopant atoms constitute extrinsic point defects. In this sense, the whole semiconductor industry is based on controlled introduction of specific point defects. This article addresses intrinsic point defects, ones that come from the native crystal. These defects govern the diffusion processes of dopants in semiconductors. Diffusion is the most basic process associated with the introduction of dopants into semiconductors. Since silicon and gallium arsenide are the most widely used semiconductors for microelectronic and optoelectronic device applications, this article will concentrate on these two materials and comment only briefly on other semiconductors.

A main technological driving force for dealing with intrinsic point defects stems from the necessity to simulate dopant diffusion processes accurately. Intrinsic point defects also play a role in critical integrated circuit fabrication processes such as ion-implantation or surface oxidation. In these processes, as well as during crystal growth, intrinsic point defects may agglomerate and negatively impact the performance of electronic or photovoltaic devices. If properly controlled, point defects and their agglomerates may also be used to accomplish positive goals such as enhancing device performance or processing yield.

Serious studies of materials are often serious studies of defects, for control of properties of materials implies control of defects or impurities. Understanding defect phenomena is crucial, and both theoretical ideas and modeling are enhancing key areas of materials properties and processing. I shall review some of the ways theory contributes. Theory enters into all aspects of materials science, even if you don't always realize you are using it.

Even self-styled practical people, for whom theory is a luxury, use theory routinely in its first main role, as a framework for the data they lovingly collect. Elasticity theory, electromagnetic theory, and thermodynamics are normal tools for working engineers. The simplest ideas about electronic and atomic structures of solids are now so standard that one can forget their original impact, just as one forgets those within living memory who objected even to the idea of atoms. It was theory which gave clear guidelines for solids to be metals or insulators, when the real-space ideas of crystal structures based on interacting atoms were complemented by the reciprocal space notions from band theory. Such rules had been far from obvious. The behavior of amorphous solids has forced analogous theory-led upheavals in understanding.

Biomedical materials are synthetic polymers, metals, ceramics, inorganics, and natural macromolecules (biopolymers), that are manufactured or processed to be suitable for use in or as medical devices or prostheses. These materials typically come in contact with cells, proteins, tissues, organs, and organ systems. They can be implanted for long-term use, e.g., an arrtificial hip, or for temporary use, e.g., an intravenous catheter. Except in isolated cases when a material is used by itself, such as collagen injections for filling soft tissue defects, biomedical materials are used as a component in a medical device. The form of the material (perhaps a textile) how it interfaces (blood contacting, for instance), and its time of use will determine its required properties. A material's use needs to be viewed in the context of the total device and its interface with the body. One material property alone is unlikely to lead to a successful and durable device, but the failure to address a key property can lead to device failure. Until recently, medical-grade polymers, ceramics, inorganics, and metals were purified versions of commercial-grade materials. A variety of polymers, biopolymers, and inorganics is now being specifically developed for medical applications. Table I summarizes the types of biomedical materials.

In the context of this article, adhesion, friction, and wear are defined as follows: Adhesion is the resistance against separation of two bodies in a direction normal to the common interface. Friction is the resistance against motion of the contacting bodies in the direction of the common interface. Wear is the gradual change of geometry and loss of material due to the different contact processes that may occur in moving contacts.

Bowden and Tabor have shown convincingly that adhesion, friction and wear are interrelated at least to some extent. This article will review some of the current arguments that support this notion.

If surfaces are not separated by a viscous film, both friction and wear have their origin in the contact phenomena occurring at the roughness asperities of the sliding bodies.

Within one contact cycle, three primary events can be distinguished:

1. In order to obtain a (normal) load-carrying capacity, the contacting asperities must be compressed or indented (Figure 1), similar to a Brinell hardness test.

2. At the beginning of the contact cycle, the asperities interfere in the direction of sliding (Figure 2). Thus during the first half of the contact cycle, frontal loading of the asperities occurs (Figure 2b).

Particularly if the resulting deformation is plastic (as it will usually be for metals, see Reference 1), the net effect of this contributes to friction, the so-called “ploughing” effect.

Mechanical properties of interfaces between dissimilar or similar materials (e.g., grain boundaries) have become the focal point of research in several fields, including composite materials (metal, ceramic and intermetallic matrix composites), tribology, and solid state devices. This is not surprising because the interfaces between dissimilar materials are sites for mechanical stress concentrations and often nucleate the overall failure process.

Interfaces of interest in composite materials exist between fibers and their diffusion barrier coatings or between the fibers and the surrounding matrix material. In the field of tribology, interfaces exist between various types of functional (magnetic, conducting, optical, electrical), protective (thermal barrier, corrosion, wear resistant), or decorative coatings and their underlying substrates. And, finally, metal/ceramic interfaces are of interest in multilayer devices and magnetic disks and head technology. In all the above applications, mechanical properties of the interface (tensile and shear strength, toughness, etc.) often control the overall functionality of the coated part. Therefore, improving the mechanical properties of the interface for a prolonged life of the coated part is of fundamental interest. However, in ceramic and metal matrix composites, where the fiber/coating interface is used to deflect impinging cracks from the matrix, it is often desirable to impair the strength of the interface.

The structural and compositional integrity of interfaces between semiconductor multilayers can profoundly influence the optical and electronic properties of epitaxially grown heterostructures. Understanding the atomic-scale interfacial structure and chemistry is therefore essential to correctly relate electrical measurements to theoretical models and to correlate such effects with growth conditions. High-resolution electron microscopy (HREM) has played a pivotal role in this process, providing important information on interface commensurability and revealing the presence and nature of defects.

More recently, significant advances have been made in applying HREM to the difficult problem of chemical composition mapping in systems where no structural change occurs across the interfaces. The basis of such methods involves using the objective lens as a bandpass filter and tuning in on a specific range of spatial frequencies to form a chemically sensitive interference pattern. By using a suitable low-index zone axis and choosing an optimum range of specimen thickness, the patterns can indeed be extremely sensitive to the strength and periodicities of the projected potential.