Research on oxynitride glasses has expanded considerably since the first systematic compositional investigations were begun some ten years ago. This review will briefly introduce that work with an emphasis on more recent developments. Also included is some discussion of the applications of oxynitride glasses in both bulk and thin film form. The literature review is selective, however, and the recently developed phosphorus oxynitride glasses are only briefly mentioned. For more detailed information on earlier work on silicon oxynitride glasses see the reviews by Jack and Loehman. For discussions of phosphorus oxynitride glasses see Marchand, Peng and Day, and Rajaram and Day.

The development of bulk oxynitride glasses has been closely associated with research on Si3N4 and other nitrogen-containing ceramics. The initial discovery that large amounts of nitrogen can be incorporated in silicate glasses originated with analyses of glassy grain boundary phases in Si3N4 that was hot-pressed or sintered with different additives to promote liquid phase formation at the processing temperature. (Some applications of oxynitride glasses take advantage of that close connection to Si3N4. See Applications of Oxynitride Glasses below.) Later, systematic study of the bulk oxynitride glasses showed that nitrogen incorporation increases density, hardness, glass transition temperature and fracture toughness, and, to a certain extent, decreases the thermal expansion coefficient. Nitrogen additions also tend to lower glass solubility in aqueous solutions.

Ceramic precursors are the chemical constituents of ceramic materials in a form in which they can be manipulated and shaped prior to processing to form a dense body. Generally these precursors are finely ground crystalline materials such as oxides or clays. Recently, however, much effort has focused on preparing precursors by chemical rather than mechanical means. The goal is to prepare purer, more homogeneous, and more reactive precursors than mixed crystalline powders. A wide variety of chemical processes have been devised to grow ceramic precursors, including powders formed by precipitation reactions, gels formed by hydrolysis and condensation reactions, and gas phase pyrolysis, to name just a few. Due to the random nature of the growth processes, these precursors often have structures wholly unlike their mechanically derived counterparts on both an atomic and a “macromolecular” scale. Random growth and structure and the effect of these structures on the subsequent processing of ceramics is an area of growing interest.

Random structures are inherently more difficult to describe and study than are the ordered, periodic structures of crystals. With the exception of a few special cases, such as random walk polymers, little of a quantitative nature could be said about them until the development of fractal geometry and the accompanying realization that the structures which result from many random growth processes are, in fact, fractal.

There is great current interest in layered anisotropic materials. Much of the interesi relates to the possibility of an approximate decoupling of 3-dimensional (3D) behavioi into ID properties perpendicular to the layer planes and 2D properties associated with the layer planes. Since this decoupling in layered materials is enhanced by intercalation, much attention has been given to intercalation compounds in recenl years as a means to study interesting questions of low dimensional physics. Since many deliberately structured materials also are prepared in layered sequences, there are many close connections between the properties of intercalated layered materials and deliberately structured materials. This article will focus on several important connections between intercalated layered materials and multilayered superlattice structures prepared by molecular beam epitaxy (MBE) and metalorganic chemical vapor deposition (MOCVD).

Interest in intercalation compounds has led to three lively MRS symposia on this topic (Fall 1982, Fall 1984, Fall 1986) and a number of summer schools and workshops, one of which was held in July 1986 at Erice, Sicily, under the auspices of NATO and the European Physical Society. The published lecture notes1 from this summer school provide a useful introduction and reference for students and others interested in this field.

This article discusses uses of high energy ion beam scattering for materials analysis, including channeling, particle induced x-ray emission (PIXE), and nuclear reaction analysis (NRA). These additional capabilities used in conjunction with RBS equipment provide capabilities for crystalline defect studies and light element detection.

Much of the anisotropic mechanical and thermal behavior exhibited by materials can be attributed to anisotropic orientation at the molecular level. In self-supporting thin films (5–10 microns) and those (0.01–1.0 microns) deposited on solid substrates, the role of molecular orientation is even more important since it will critically determine their two-dimensional behavior and their structural integrity as well. These two aspects are extremely important because if thin organic and polymeric films are to be competitive with existing materials for such diverse applications1 as chemical sensors and integrated optics, they must be mechanically robust and, ideally, defect free. These stringent constraints dictate that sophisticated characterization techniques, which can interrogate at the molecular level, be developed or refined so as to have the sensitivity to address these critical issues.

The development of nondestructive techniques for studying thin organic films has certainly lagged behind those developed for metallic and semiconductor films. Unfortunately, many of these same techniques cannot be simply applied to organic films because they are “invasive” and often alter the structure of the system they were designed to probe. This is especially so in organic and polymer films, and this awareness within the materials science community has led to the adaptation of many photon intensive techniques to the study of thin films. A number of these will be discussed in later sections with their relative merit put in perspective.

Certainly the origins of anisotropic structure in bulk materials are manyfold, but the spatial constraints in 2-D can lead to even more complex causes of orientation. In thin films on substrates the role of the surface is important in determining the ordering and orientation of the individual molecular segments which come into intimate contact with it. The extent of this orientation and order is still somewhat controversial but there is general agreement that it probably differs depending on the nature of the substrate.

We have been interested in fabricating structures which can have dimensions on the scale of molecules (˜10 Å). By producing lithographic structures on the molecular-scale length, we obtain a new class of materials with unusual chemical properties. The materials contain pores which have been shown to restrict the mobility and adsorption of molecules. This restriction is the basis of size- and shape-selective chemistry which is used in many modern separation and catalytic processes.

Features in our materials are significantly smaller than the smallest nano-structures produced with electron beam lithography. To date 20–50 Å features have been produced with electron beam lithography which uses ionization from beam-solid interactions to drive radiation chemistry in a resist layer. These dimensions are near fundamental limits which come from beam-solid interactions and the response of the resist system. Further limitations occur when the nanostructural pattern must be transferred from the resist layer into a useful metal or semiconductor structure. Methods have not been devised to fully exploit the accuracy with which resist patterns can be defined, and the smallest useful transferred structures have dimensions of ˜80–100 Å.

To produce the smallest of all lithographic structures, we use a new method for pattern generation. Instead of using radiation chemistry to define a pattern, we use the organization of layers in deposited superlattices.

The fundamentals of the high energy ion beam technique Elastic Recoil Detection are presented. The potential of this analysis technique for the depth-resolved determination of light elements in a heavy matrix is illustrated with examples from semiconductor technology.

Concurrent energetic particle bombardment during film deposition can strongly modify the structural and chemical properties of the resulting thin film. The interest in this technique, ion-assisted deposition, comes about because it can be used to produce thin films with properties not achievable by conventional deposition. Bombardment by low energy ions occurs during almost all plasma-based thin film deposition techniques. Bombardment of a growing film, particularly by accelerated ions, can also be combined with non-plasma-based deposition techniques, such as evaporation, to simulate some of the effects observed with sputtering. The bombarding particle flux is usually controllable so that the arrival rate, energy, and species can be independently varied from the depositing flux. Thus, a basic aspect of ion-beam-based deposition techniques is the “control” often absent in plasma-based techniques. In plasmas, the voltage, current, and pressure are all interdependent. The energetic bombardment at the substrate-film interface depends on the various properties of the plasma, as does the deposition rate. It is often difficult, or even impossible, to decouple these processes. With ion-beam-based deposition techniques, the ion bombardment is essentially independent of the deposition process, and both can be more easily controlled.

The incident energetic particle contributes some of its energy or momentum to irreversibly change the dynamics of the film surface. The incident particle may also be incorporated into the growing film, changing the film's chemical nature. The changes induced by particle bombardment during deposition are often not characteristic of equilibrium thermodynamics because the incident particle's energy is often many times the local adsorption or binding energy.

In the early 1980s Prof. Bowen at Massachusetts Institute of Technology advanced the concept that sintered ceramic bodies could be improved by decreasing the defects in green bodies. In addition, his group advanced the idea that ideal green bodies should be composed of monodisperse particles packed into an ordered array. They used the hydrolysis of metal alkoxides to produce narrow size distribution sol-gel powders (i.e., amorphous). These powders were allowed to settle under the force of gravity into an ordered array which was dried and sintered. Sintering took place at much lower temperatures and at faster rates than traditional ceramic processing (e.g., broad size distributions of crystalline powders). Improvements in strength and toughness of the sintered body were never demonstrated by Bowen's group for these novel ceramics.

Ordered particles have packing faults which lead to ordered domains, as shown in Figure 1, similar to grains in polycrystalline materials. When sintered, the ordered domains shrink separately and pull away from each other as shown in Figure 2. This leads to defects in the sintered body of sizes similar to those of the ordered domains which can encompass as many as 10,000 particles. These ordered domains lead to weakness in the sintered body according to Griffith's theory.

Until recently, phosphate glasses were regarded as somewhat of a scientific curiosity—materials limited in technological application but nevertheless exhibiting interesting fundamental properties. Although the basic properties of many phosphate glasses have been studied on a laboratory scale during the past 50 years, very few of these materials proved to be of commercial interest. This lack of extensive practical application was primarily related to their poor chemical durability and a tendency to crystallize during processing.

The pace of modern technological development continues to result in an increased demand for new materials with specific characteristics, and this demand has led to renewed interest in a number of previously neglected materials—including phosphate glasses. The combined physical and chemical characteristics of phosphate glasses make them relatively unique among the families of inorganic glasses. These characteristics include: large thermal expansion coefficients, low preparation and softening temperatures, low melt viscosities, chemical compatibility with living bone, and a polymeric structure similar to that in organic polymers. Most recent research on phosphate glasses has emphasized one or more of these characteristics.

The major scientific questions pertinent to technological applications of phosphate glasses are related to utilizing the desirable characteristics of these materials while maximizing their corrosion resistance and thermal stability. In order to tailor a phosphate glass for a particular application, the corrosion mechanism of the glass must often be understood in detail, along with the relationship between the microscopic structure of the glass and its macroscopic properties.

Forty years ago there were essentially no academic departments with titles of “Materials Science” or “Materials Engineering.” There were, of course, many materials departments. They were called “Metallurgy,” “Metallurgical Engineering,” “Mining and Metallurgy,” and other permutations and combinations. There were also a small number of “Ceramic” or “Ceramic Engineering” departments. Essentially none included “polymers.” Over the years titles have evolved via a route that frequently followed “Mining and Metallurgy,” to “Metallurgical Engineering,” to “Materials Science and Metallurgical Engineering,” and finally to “Materials Science and Engineering.” The evolution was driven by recognition of the commonality of material structure-property correlations and the concomitant broadening of faculty interests to include other materials. However, the issue is not department titles but whether a single degree option in materials science and engineering best serves the needs of students.

Few proponents of materials science and engineering dispute the necessity for understanding the relationships between processing (including synthesis), structure, and properties (including performance) of materials. However, can a single BS degree in materials science and engineering provide the background in these relationships for all materials and satisfy the entire market now served by several different materials degrees?

The issue is not whether “Materials Science and Engineering” departments or some other academic grouping of individuals with common interests should or should not exist.

The larger chemical companies worldwide are in the process of major change from a commodities orientation to “downstream” products. “Value-added products,” “engineered materials,” and “effect chemicals” are terms their leaders are using as they try to explain what is happening to their boards of directors, to their stockholders, to their employees, and to their customers.

As these companies go through their reorganizations a major problem has become apparent — many professional chemists are not intellectually equipped by training and experience to be effective in research and development in the design of those new products, and the “pipeline” which is producing their “new hires” has not responded to the changing emphases needed in undergraduate curricula and graduate research programs in chemistry.

The driving force behind this new orientation is coming to be known as the science of materials or materials science. The longer descriptor, materials science and engineering, better suggests its inherently broad and interdisciplinary nature as well as bringing out that how the final product is produced affects (and may effect) its properties. Because this new discipline moves across standard science and engineering boundaries, materials science does not lend itself to neat classification. Descriptive classification is nevertheless possible:

• Materials science is solid state science, both solid state chemistry and solid state physics.

Diffusion in polymer alloys or blends can be used to extract information on the fundamentals of the dynamics of individual polymer chains in the melt and the thermodynamics of the interaction between unlike polymer species. The dynamics of individual chains are available from measurements of the tracer diffusion coefficients, D*, of the various species while the thermodynamics of interaction, represented by the Flory parameter, x, can be obtained from measurements of the mutual diffusion or interdiffusion coefficient, D. We will show that these quantities can be measured conveniently by forward recoil spectrometry (FRES), an ion beam analysis technique that can determine the concentration versus depth profile of polymers labeled with deuterium diffusing into unlabeled polymer matrices.

For high enough molecular weight of the matrix, the tracer diffusion coefficient of both species in the blend scale as D0N−2, where N is the number of monomer segments per diffusing chain; the constant D0, however, can differ by more than 104 for chemically different molecules diffusing in the same blend, suggesting that conventional concepts of chain dynamics in melts, such as monomer friction coefficients, need to be reexamined. The mutual diffusion coefficient is controlled by the faster species in the blend (the one with the larger D*N product) in agreement with what was found in metallic alloys (but in sharp disagreement with the “slow” theory of mutual diffusion which predicts that the slower species controls). Since the combinatorial (ideal) entropy of mixing of polymers is low, the thermodynamic driving force for diffusion is dominated by enthalpy and excess entropy of mixing (x) to a degree unprecedented for atomic or small molecule systems. This means that one can observe not only a thermodynamic “slowing down” of diffusion when x becomes positive as one nears the spinodal but also a large thermodynamic “speeding up” of diffusion when x is negative. Measurements of mutual diffusion turn out to be one of the most sensitive methods available for measuring x.

Electronic ceramic materials research is one of the fastest growing, most highly publicized areas of materials science. Subjects receiving considerable attention include high temperature superconductors, multilayer ceramic composites for high density microelectronics packaging, and ferroelectric electro-optic thin films. A complete review of all aspects of electronic ceramics research is beyond the scope of this article, which will focus on two general topics whose development is representative of recent contributions to the field. These two areas are synthesis and characterization of electronic ceramic films,1 and controlled use of low level dopants (1,000 ppm or less) in bulk polycrystalline ceramics, thin films, and single crystals to achieve desired properties. Perspective of the progress in ceramic film development is given by a review of single-crystal synthesis and properties.

Several examples of the impact that low level dopants and thin film synthesis have on electronic ceramics development are presented. Dopant concentrations of 1,000 ppm or less can have a dramatic effect on microstructural, optical, and electrical properties. For example, a decrease in aluminum content of 150 ppm resulted in an increase in grain size from 1 to 25 microns in otherwise identical ZnO varistors. Background aluminum concentrations for these varistors were less than 10 ppm. In another example, the photorefractive effect, the change in refractive index with optical light intensity, has been shown to be altered by orders of magnitude with ppm doping levels in ferroelectric electro-optic materials.

Several electronic ceramic devices have recently been developed due to improvements in ceramic film processing. Examples of these devices include: 1. multilayer PZT transformers, which allow fabrication of complex monolithic passive multicom-ponent networks, 2. liquid cooled multilayer ceramic substrates, with 400×800 micron liquid transfer capillaries integrated into the multilayer structure via ceramic processing techniques for high density VLSI packaging, and 3. ferroelectric electrooptic thin films that are compatible with silicon or III-V technology. For all the above applications, synthesis of electronic ceramic materials into high purity films is essential.

This article presents an overview of our current understanding of the fundamental factors underlying Secondary Ion Mass Spectrometry (SIMS). Included is a discussion of the sputtering process and possible mechanisms which produce ejected ions. Presently available instrumentation for SIMS analysis is discussed and some examples of SIMS analysis are also given.

Materials research is a long-standing tradition at Carnegie Mellon. Since its inception as Carnegie Technical Schools in 1906, the metallurgy program has flourished on the campus. Evolving from a single department involved in metals research formed in 1906, leading-edge, interdisciplinary materials research has grown considerably, with materials-related research now carried out in many departments. These include Chemical Engineering, Chemistry, Civil Engineering, Electrical and Computer Engineering (ECE), Mathematics, Mechanical Engineering, Physics, and Mellon Institute (an affiliate of the University), and, of course, Metallurgical Engineering and Materials Science (MEMS). It is beyond the scope of this article to cover every aspect of materials-related research at Carnegie Mellon. Consequently, we have decided to concentrate on materials and topics of particular interest to MRS members.

The current research pertaining to materials at Carnegie Mellon can be broadly classified by material type into three categories: metals and alloys, polymers, and electronic and magnetic materials.

The major thrust on research in metals and alloys is in MEMS. In addition, there are a number of complementary efforts in Chemical Engineering and Mechanical Engineering. For example, Prof. Sides of Chemical Engineering is evaluating electrolytic extraction of aluminum from its ores, while Professors Prinz, Sinclair, Steif, Swedlow, and Wright of Mechanical Engineering are examining the macroscopic flow behavior of metals and alloys and its relevance in manufacturing engineering. Prof. Prinz is also interested in vibratory compaction of metal powders, both from experimental and modeling points of view.

Although there are some significant exceptions, most important glass-forming systems contain elements from the sixth, or chalcogenide, column of the periodic table (oxygen, sulfur, selenium, or tellurium). The glasses which contain oxygen are typically insulators, while those which contain the heavier chalcogen elements are usually semiconductors. Even though oxygen is technically a chalcogen element, the term “chalcogenide glass” is commonly used to denote those largely covalent, semiconducting glasses which contain sulfur, selenium, or tellurium as one of the constituents.

The chalcogenide glasses are called semiconducting glasses because of their electrical properties. The electrical conductivity in these glasses depends exponentially on the temperature with an activation energy which is approximately one half of the optical gap. In this sense these glasses exhibit electrical properties similar to those in intrinsic crystalline semiconductors. The analogy is by no means perfect. The mobilities for the charge carriers in these glasses are very low (< 10 cm2/V-s) compared to crystalline semiconductors, and there are even discrepancies in determining the sign of the charge carriers from measurements of the Hall effect and the Seebeck effect.

The first detailed studies of the chalcogenide glasses were performed about 30 years ago. For many years the prototype compositions have been selenium (Se), arsenic triselenide (As2Se3) or arsenic trisulfide (As2S3), and germanium diselenide (GeSe2) or germanium disulfide (GeS2).

Ion beam processing of materials has a tradition at Oak Ridge National Laboratory that is as old as the laboratory itself. Consequently, when we began looking for a competitive way to participate in the excitement and new physics beginning to emerge from the fabrication and study of artificially structured materials, it was natural to look for a growth technique that incorporated ion beam processing. Our division, the Solid State Division, has a variety of ion implantation and ion beam analysis accelerators which are integrated with pulsed-laser sources into ultrahigh vacuum (UHV) surface analysis and processing chambers. These facilities allow us to do ion beam and laser processing of materials in UHV at temperatures from liquid helium to several hundred degrees centigrade and to study these alterations in situ by a variety of ion beam (ion scattering, ion channeling, nuclear reactions, etc.) and surface analysis (low energy electron diffraction, Auger, etc.) techniques. Since isotope separation has been done continually at ORNL for almost 45 years, the idea and advantages for altering this technique to do materials fabrication in UHV were immediately obvious. In the following article we will briefly review the history of the ion beam deposition (IBD) concept, describe our preliminary apparatus, and point out the inherent advantages of IBD for fabricating and studying artificially structured materials. Recent results obtained by IBD will be presented.