Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 255
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials APPENDIX D Instrumentation OVERVIEW “The whole progress of research shows that discoveries depend on more and more powerful instruments,” said Erich Bloch, director of the National Science Foundation, in April 1985 at the tenth annual American Association for the Advancement of Science colloquium in Washington, D.C. A better illustration of his statement could not be found than the development of the scanning tunneling electron microscope, which has enabled materials scientists to “see” atoms on the surfaces of solids. We can now “see” reconstructed surfaces, defects, step edges, and even dangling atomic bonds. G. Binnig and H.Rohrer received the Nobel Prize in 1986 for their development of this new instrument. Successful development and use of scientific instruments are essential for research in materials science and engineering. Increasingly sophisticated instruments are needed as we push our understanding of materials to more microscopic levels, and as the preparation of new materials becomes more exotic. This field of science uses advanced instrumentation not only to characterize materials but also, in many cases, to prepare new materials. The cost of this equipment is rising rapidly in relation to the overall scientific budget, creating an obvious problem. The development of new instrumentation is not only costly, however; it also requires reliable long-term commitments by both the working scientists and the agencies and institutions upon which they must rely for support. The principal thesis of this appendix is that, within large areas related to materials research in the United States,
OCR for page 256
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials such commitments have been inadequate to meet present needs and opportunities. The results of inadequate U.S. commitment to instrumentation in materials research have become increasingly apparent in recent years. The current shortage of state-of-the-art instrumentation for preparing and characterizing materials is a problem that has been discussed by several national study groups and has been addressed by special instrumentation programs in several federal funding agencies. The outward manifestations of this problem are inadequate and obsolete instrumentation in universities and government laboratories, a growing dependence on foreign laboratories for the development of advanced new instrumentation, too small a number of students being trained in the use or development of sophisticated instruments, too few small U.S. companies capable of developing and manufacturing new instruments, and inabilities of larger industrial laboratories to apply modern techniques of measurement science to fabrication and processing. The committee’s survey of the current state of instrumentation in materials science and engineering has led it to the following conclusions: The shortage of modern instrumentation in U.S. laboratories is a symptom of a problem deeply embedded in the materials science and engineering community. Too low a priority has been placed on instrumentation, instrumentation development, and measurement science as a whole. The characterization of materials and the instrumentation that goes with it are still regarded as routine service functions in many materials laboratories. Scientists working on the development of new instrumentation or measurement techniques are not viewed as being part of the scientific or engineering elite. A very large fraction of the worldwide development of new instruments currently takes place outside the United States. As a result, state-of-the-art instruments often are being used in foreign laboratories to prepare or characterize materials long before they appear in the United States. The committee’s survey of the instrumentation used, for example, by surface scientists, indicates that there is approximately a 5-year delay between the time a new instrument is announced to the scientific community and the time it is transferred to research laboratories in the United States. It therefore seems possible that a significant part of U.S. materials research will fall some years behind that of our foreign competitors unless we do more to encourage the development of instrumentation in this country. The lack of research in instrument development in United States universities has created a shortage of students trained in sophisticated instrumentation. Innovative developments in instrumentation do occur in the United States, but our system often does not seem to be capable of sustaining these inno-
OCR for page 257
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials vations. A common story in the development of a new instrument is that the first paper to describe a new technique is published by a U.S. scientist, but the real development of the instrument takes place abroad. It then takes approximately 5 years before this equipment is developed by an instrument company (usually overseas) into a commercially available product. There are many economic and cultural reasons why instrumentation development does not flourish in the United States as well as it does abroad, and most of these reasons are beyond the scope of this report. But there are two problem areas where we have the opportunity to make positive changes: first, in the attitude toward instrumentation in our materials laboratories and, second, in the role played by U.S. national and federal laboratories. State-sponsored laboratories in other countries, especially in Europe and Japan, play a much more important role in instrument development than they do in the United States, and private instrument companies abroad (which are more numerous than those in the United States) work much more closely with the universities and national laboratories. The committee believes that cooperation of this kind can and should be improved in the United States. If the United States is to stay at the forefront of materials research, then major resources must be devoted to the development of instrumentation, and special attention must be paid to the transfer of innovative instrumentation technology to research laboratories, to commercial instrument manufacturers, and eventually to the industries that will use this technology. In order for this to happen, there will have to be some change in the priorities of the materials science and engineering community. Unlike other scientific communities such as high-energy physics, astrophysics, and biology, the materials community devotes only a minimal fraction of its resources to the development of new instrumentation. Less than 1 percent of the budget of the Division of Materials Research (DMR) of the National Science Foundation (NSF) is allocated for this purpose; policies in some parts of the Department of Defense (DOD) and the Department of Energy (DOE) reject responsibility for this activity. By and large, novel advances in instrumentation relevant to materials research are not generated by companies or large mission-oriented projects; they are generated by individual scientists with specific purposes in mind. It is this particular kind of inventiveness that must once again be encouraged in the United States. Programs for the development of innovative instrumentation for materials research should be encouraged and supported at U.S. universities. Funding for such programs should amount to approximately 5 percent of the support for academic materials research as a whole. Proposals from individual investigators should have high priority. Support for multi-investigator research in instrumentation also should be considered. The national and federal laboratories should begin to play a central role
OCR for page 258
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials in developing new instrumentation for materials research and should act as the vehicle for rapid transfer of this new technology to the industrial community. The National Institute of Standards and Technology (formerly the National Bureau of Standards) is expected to play a key role in instrumentation or measurement-science programs aimed at materials science and engineering. Its responsibilities include the development and testing of new instrumentation for basic research and the transfer of instrumentation technology to industry. New resources would be needed in order for this to happen. There is a real opportunity at present for U.S. policymakers to strengthen both materials research and our system of national laboratories by redirecting the missions of these laboratories. The Packard Committee, which was commissioned by the Office of Science and Technology Policy to review federal laboratories, has asked that the laboratories define their missions more clearly and make these missions different from each other. The present committee recommends that the long-term development of materials-related instrumentation become a more significant part of the new mission of one or more of these laboratories. A very good example of an already existing activity of this kind is the X-Ray Optics Center at Lawrence Berkeley Laboratory. Another example of how state-sponsored laboratories can play central roles in the development of new instrumentation is the German Max Planck institutes. The interaction between universities, government laboratories, and instrument companies must be enhanced with the objective of creating a healthy and expanding commercial instrumentation community that serves the needs of U.S. materials research. This objective might be accomplished by federal programs in which, for example, manufacturing experts from small instrument companies are encouraged to collaborate with scientists at universities or government laboratories in developing new instrumentation. PRIORITIES FOR INSTRUMENTATION AND INSTRUMENT DEVELOPMENT IN MATERIALS RESEARCH In preparing this appendix, the committee asked two different but related questions concerning the priority placed on instrumentation by materials scientists and engineers and the agencies upon which they depend for support. First, what priority is placed on the acquisition of new instrumentation relative to carrying out research with existing equipment? Second, how much emphasis is placed on the development of new instruments as part of research programs? These two questions seem to be related in the following sense: If a scientific community really believes that instrumentation is important, then it will advocate support for the development of new instrumentation in order to
OCR for page 259
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials make sure that it has ready access to the most modern equipment. The committee finds that, for reasons apparently having to do with the sociology and reward structure of this field, altogether too low a priority is placed on both the development and the acquisition of new instrumentation. How do materials scientists spend their money? Statistics from NSF, DOD, and DOE all show that the percentage of money spent for capital equipment by individual investigators at universities ranges from 8 to 14 percent. It may legitimately be argued that this small percentage does not reflect the importance of instrumentation to these scientists. The grants are usually so small that principal investigators really do not have the option of using this money to buy major pieces of equipment. When budgets are cut, priority necessarily must be assigned to people rather than hardware. In recent years, all of the major funding agencies have initiated instrumentation programs to which scientists can apply for support for new equipment. These programs have been adopted as attempts to redistribute the money so that more is spent on instrumentation. One obvious conclusion is that these programs have been made necessary by the fact that materials scientists and engineers have not been willing or able, on their own initiatives, to spend their research funds on instrumentation. Another measure of attitudes toward instrumentation may be gained by looking at the performance of the NSF-sponsored materials research laboratories (MRLs). These laboratories are block-funded by NSF at levels high enough to permit purchases of major items of equipment, and decisions about how the money is spent are made by the MRL scientists themselves. The MRLs have been operating with some stability for over 25 years, and one of their major features is supposed to be emphasis on the development of central experimental facilities. During the period from 1979 to 1982, only about 12 percent of the overall MRL budget was spent on experimental equipment. In 1982, the management of this program at NSF recognized that this percentage was too low and took steps to increase it. This was accomplished by taking money out of core support and giving it back as “equipment supplements.” The stated goal was to increase the support of instrumentation to 20 percent of the MRL budget by 1984. This goal was finally reached in 1987. Once again, one can see that the materials research community has been unwilling or unable to pay adequate attention to instrumentation and has had to depend on federal funding agencies not only for money but also for help in setting priorities. This problem is clearly related to funding, but the committee believes that it is also, to a significant extent, sociological in that there is a natural tendency in the research community to preserve skilled personnel. In regard to the question of support for—and interest in—the development of new instrumentation in the United States, the committee has observed that there is relatively little money going into this area of research. The instru-
OCR for page 260
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials mentation program in the Division of Materials Sciences at DOE explicitly excludes support for instrument development. DOD has no documented case of instrument development in materials science being funded through their new University Research Initiatives program. Only the NSF instrumentation programs show money being allocated to development. The instrumentation program in DMR has put $1 million (20 percent of the program’s budget) into instrument development—approximately 1 percent of the total DMR budget. National Science Foundation program officers have argued that, even if the funding agencies were able to supply more money for instrument development, neither the NSF nor the structure (or administration) of the scientific community in the United States would be able to provide the supporting staff necessary to sustain effective programs in this area. Instrumentation development programs do not generally produce short-term results, and they require stable, highly trained supporting staffs. Both of these characteristics are inconsistent with the way in which most academic departments operate in the United States. If it is important (as the committee believes it is) that the materials science community have an active university-based instrument development program, then the attitude of the scientific community toward instrument development must change. This change in attitude must be accompanied by changes in the hiring practices of academic departments, changes in the policies of university administrations regarding support staffs, and the initiation of instrument development programs in all of the funding agencies. The principal consequence of neglecting instrument development in U.S. universities seems clear: there will continue to be a national shortage of scientists trained in this area. This shortage of skilled personnel will have several further consequences. The time lag between the invention of new instruments and their transfer into U.S. academic or industrial laboratories will increase. The United States will lose out to foreign competition in the commercialization of instrumentation, especially by small, specialized instrument companies. More and more of the instrumentation that is purchased using federal monies will be imported. But by far the worst consequence of neglecting the development of instrumentation in U.S. universities is that, unless some other segment of our scientific community such as federal or industrial laboratories takes new initiatives in this area, our materials research community will fall behind its international competition in its experimental capabilities. In the committee’s discussions with funding agencies and with individual materials scientists, it frequently heard the opinion that instrument development is incompatible with the operation of U.S. universities. The pressures of competing for tenure and promotion make it difficult for young scientists to embark on extended and risky development programs, and the formal educational programs of graduate students are shorter than the lifetimes of
OCR for page 261
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials such projects. The committee believes that these problems are solvable. Examples of successful instrument development programs exist in state-supported academic materials laboratories in Europe and Japan as well as in a few of our own universities. Moreover, instrument development at U.S. universities has traditionally had very high priority in other fields, especially high-energy and nuclear physics, astrophysics, and biology. The committee sees no fundamental reason why this style of research cannot be accepted more broadly in materials science and engineering. If instrument development is not taking place in our universities, then the only other possibilities are industrial or government laboratories or instrumentation companies. Historically, the big industrial research laboratories like AT&, IBM, Exxon, and Xerox have supported instrument development, but this effort seems to be declining. The percentage of our equipment that comes from overseas indicates that U.S. instrument companies are not holding their own with foreign competition, particularly in the area of specialized, one-of-a-kind instrumentation. The question is whether the government laboratories are filling the gap in instrument development for materials research. EXAMPLES OF INSTRUMENT DEVELOPMENT IN MATERIALS RESEARCH: SURFACE SCIENCE Materials science and engineering is a combination of many traditional disciplines, ranging from solid-state physics to mechanical engineering; thus the instrumentation needed is quite diverse. In an attempt to understand what is happening in instrument development, the committee chose to focus primarily on one part of materials research, specifically, surface science. Surface science is a particularly crucial and active part of materials research, where, historically, there has been a large effort in instrument development. Therefore a study of the trends in this branch of materials science and engineering might give useful information about the health of the field as a whole. The danger was that by looking only at surface science one might obtain a falsely optimistic signal: surface scientists in the United States are reputed to be overly fascinated by their instrumentation. The committee made its decision to trace the history of the development of instrumentation used to characterize the surfaces of materials fully realizing that certain important instrumentation such as molecular beam epitaxy, which is used for materials preparation, would not be addressed. The findings are summarized as follows: Twenty years ago, the United States dominated instrument development in surface science. Prime examples are the Auger spectrometer and the low-energy electron diffraction (LEED) display system. In these examples the universities, industrial laboratories, and instrumentation companies were actively involved in the development.
OCR for page 262
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials In the past 20 years, the United States has almost totally lost the dominance that it once enjoyed in this field. Innovative design and unique applications of instruments still occur in the United States, but these accomplishments do not seem very often to result in commercial instruments. Typically, as is shown below, the first applications of new measurement techniques have been announced by U.S. scientists, but the real development of the instruments has taken place in foreign laboratories. Another common occurrence is that U.S. scientists in fields such as elementary-particle physics have developed experimental techniques that are applicable to materials research, and U.S. surface scientists often have been first to adapt these techniques, but that, in the long run, the U.S. scientists have been overtaken by the ability of foreign laboratories to sustain complicated and expensive instrument development programs. Brief histories of the development of several surface science instruments are given below. The committee believes that these case histories are illustrative of the general state of instrument development in the United States as summarized in the preceding paragraphs. Scanning Tunneling Microscope The scanning tunneling microscope produces an image of atomic structure at a surface by measuring the rate at which electrons tunnel quantum mechanically from the surface to a nearby ultrasharp probe. The instrument was developed at IBM Zurich by G.Binnig and H.Rohrer (Nobel Prize in 1986) and was transferred within 2 years to U.S. laboratories. The rapid transfer of this instrument into many other laboratories was a consequence of the open attitude of Binnig and Rohrer, coupled with the fact that a tunneling microscope is a relatively small and inexpensive instrument. It is important to point out that R.Young’s group at the National Bureau of Standards had developed the basic concept of the scanning tunneling microscope in 1972. Their “topografiner” had not achieved atomic resolution by the time that their program was terminated by the National Bureau of Standards, and we shall never know whether it might have done so. The fact that an instrumentation program that was pushing back the frontiers of surface science was phased out at a major, federally operated U.S. laboratory is an especially dramatic illustration of the low priority placed on instrument development in this country. Double-Alignment Ion Scattering Double-alignment ion scattering, a technique that was developed at FOM, the state-supported Institute for Atomic and Molecular Physics in The Neth-
OCR for page 263
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials erlands, uses the channeling of medium-energy charged particles and the blocking of the backscattered ions to determine atomic structures at surfaces or interfaces. The research group at FOM has published important papers reporting the use of this technique for both basic research on surface melting and more practical investigations of atomic arrangements at silicon-silicide interfaces. It took 8 years after the first paper from FOM was published before a publication based on this technique appeared from a U.S. laboratory. The second and third double-alignment ion scattering instruments in the United States have just come on-line at IBM and AT&T Bell Laboratories. The scientists at FOM have worked closely with High Voltage Engineering Europa B.V., a European company that is now selling these systems. High-Resolution Electron Loss Spectroscopy In high-resolution electron loss spectroscopy (ELS), monoenergetic electrons are inelastically scattered from a surface and subsequently energy-analyzed. The characteristic losses seen in the scattered beam measure the vibrational modes of molecules adsorbed on the surface or the phonon modes of the clean surface. The basic concepts of this experiment as well as the instrumentation were developed to study molecules in the gas phase. The first application of this technique to surfaces was presented by F.Propst and W.Piper (University of Illinois) in 1967, but the real development of the instrumentation and the procedure into a useful technique for surface science was accomplished by H.Ibach (West Germany) and S.Andersson (Sweden). One of the most demanding applications of this instrument is the measurement of surface phonon dispersion on clean or adsorbate-covered surfaces. Ibach’s group reported the first phonon dispersion curve in 1982, and it was 4 years before a phonon curve was measured in the United States (by L.Kesmodel at Indiana University). The instrument development at the Kernforschungsanlage in West Germany by Ibach’s group was transferred to Leybold-Heraeus, where a commercial ELS system was built and marketed. Until a few years ago it was the only ELS system that could be purchased. At present, two U.S. companies are marketing ELS instruments. Angle-Resolved Photoemission Using Synchrotron Radiation Angle-resolved photoemission has become the primary tool for measuring electronic properties both within the bulk and on the surfaces of solids. U.S. scientists led in the development of this technique. In 1977, an analyzer was built by scientists at the National Bureau of Standards (NBS) and the University of Pennsylvania using electron optics developed at NBS for studying gas-phase systems. In 1979 IBM introduced a new two-dimensional display system, and in 1982 a high-resolution instrument with variable momentum
OCR for page 264
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials resolution was introduced at Bell Laboratories. Yet, in spite of these U.S. developments, all commercially available angle-resolved analyzers are now made and marketed in Europe. Auger Spectroscopy Auger spectroscopy and scanning Auger spectroscopy are the standard tools used to determine chemical compositions of surfaces. The development of Auger spectroscopy by the PHI Corporation is probably the greatest success story of U.S. surface instrumentation. Undoubtedly, the success of PHI owes a great deal to the training that W.Peria gave that company’s founders when they were students at the University of Minnesota. Peria ran a surface science laboratory where every student learned about designing, building, and testing new instruments. The development of the Auger spectrometer by Palmberg was begun at Cornell, where he was a postdoctoral student, but was brought to fruition at North American Rockwell. Such developments seem increasingly difficult in the present scientific climate. Low-Energy Electron Diffraction The diffraction of electrons from surfaces was observed in 1927 by Davisson and Germer, but it was Germer working at Cornell with support from the Varian Corporation who developed the LEED display system that is used today in nearly every surface chamber. Since Germer’s work on the LEED display system in the late 1960s, most of the advances in LEED have originated in Europe. Henzler’s group at Hannover, West Germany, has developed a high-coherence, spot profile system capable of resolving surface features on a scale of 1000 angstroms. This instrument is being marketed by Leybold-Heraeus (West Germany). Mueller’s group in Erlangen, West Germany, has developed new LEED optics and a very fast data acquisition system that allow them to study diffuse low-energy scattering from surfaces as well as scattering by adsorbates that are damaged by the beam in conventional LEED systems. The most up-to-date LEED system is manufactured and sold by Omicron (West Germany). Electron Microscopy Two examples drawn from electron microscopy illustrate many of the points made above. To be useful in surface studies, an electron microscope must keep the specimen clean to ultra-high-vacuum (UHV) standards, it must include various capabilities for treating the surface within the microscope chamber, and
OCR for page 265
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials it must also be able to provide some tools for surface characterization. The first electron microscope to address these issues effectively was described in 1978 by Takayanagi and his collaborators. This group was working at the Tokyo Institute of Technology as part of a long-running program of studies of epitaxial film growth led by Goro Honjo. Since 1978, this group has published extensive studies of semiconductor and metal surfaces, attacking problems such as atomic reconstruction of surfaces, domain growth, surface phase transitions, and adsorption kinetics. Over 30 papers in this area were cited in two reviews by Takayanagi and Yagi at the 1986 International Congress on Electron Microscopy in Kyoto, Japan. In the United States, the first similar publications date from 1983 (P.Petroff and Wilson), that is, 5 years after the first Japanese work. The first commercial electron microscope with UHV capability was announced by JEOL (Japan) in 1986 and is now available for purchase by U.S. laboratories. Assuming typical instrumental start-up problems, the committee expects that increased U.S. research in this area will begin to appear in the next year or so, about 10 years after the first paper was published by the Tokyo group. The second example, scanning transmission electron microscopy (STEM), was developed in the United States by A.Crewe and his co-workers. The crucial breakthrough was the invention of an electron gun that had sufficient brightness to achieve high resolution in a scanning mode. Effective use of high-performance electron lenses based on the work of West Germans and, of necessity, UHV techniques resulted in development of a scanning electron beam system capable of focusing a current on the order of 1 nA into a spot less than 5 angstroms in diameter. Many new imaging techniques were made possible by use of inelastically scattered electrons, emitted x-rays, and various forms of the elastically scattered beam. Development of this instrument was initiated by Crewe, an accelerator physicist and then the director at Argonne National Laboratory, and was continued by Crewe and his students after he moved to the University of Chicago. Funding of the development of the instrument was initiated through the director’s discretionary fund at Argonne and continued through the National Institutes of Health and the Biological and Radiation Physics section of DOE. Again, this instrument took about 10 years to develop. It was picked up for manufacture by VG Instruments (a British company), which has sold over 20 instruments in the United States, all of which are being used for materials science. Ironically, in view of the initial funding sources, the committee knows of only one home-built instrument in the United States, at Brookhaven National Laboratory, which is being used for biological research. Current developments in this area are being funded by the DMR at NSF. This example illustrates a commitment that, in this case, has had great benefit for materials research. It also illustrates the ability of U.S. scientists to be innovative in this area if the resources are
OCR for page 266
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials available. Finally, the project was undertaken by an accelerator physicist; it seems unlikely that it could have been started by a materials scientist under today’s circumstances. Low-Energy Electron Microscope The low-energy electron microscope, which has just recently been demonstrated by Bauer and his co-workers in Clausthal, West Germany, uses a field emission source to create a very bright beam that is collimated at high energy and then decelerated before being back-diffracted from the sample. The diffracted beam is magnified and then used to form an image of the sample. Such images show great promise of revealing defect structures of surfaces. The development of this microscope took Bauer nearly 20 years. The original program began in the early 1960s while Bauer was working at the Michelson Laboratory in China Lake, California. After approximately 4 years of work, the program began to be phased out by the management because of the lack of published results. Bauer then moved himself and the uncompleted microscope to Clausthal in 1969. Progress in Clausthal was very slow for some years. Finally, in 1975, Bauer and Telieps succeeded in making this complicated microscope function. During the late 1970s, Bauer received very little support from the West German government because a referee had reported that the instrument would never work and that the project should never have been started. It was possible for Bauer to complete the microscope only because, as a chaired professor in a West German university, he had considerable support of his own. In Bauer’s own words, “The success of such a project depends one hundred percent on having the right people and the necessary support at the right time. If one of these conditions is not fulfilled, the project can last forever or fail completely. I was lucky to have fallen into the ‘last forever’ category.” Spin-Polarized Measurements Spin-polarized photoemission measures the energy distribution of the spin-dependent states, which determine the magnetic behavior of materials. This class of measurement techniques was first demonstrated in Switzerland in 1969. It was extended by using synchrotron sources and photoelectron energy analysis about 10 years later at a West German government laboratory with an excellent support staff and large financial resources. This instrumentation-intensive experiment is only now being undertaken in the United States by an NSF-supported materials research group consisting of nine institutions including government, university, and industrial laboratories. If this group
OCR for page 267
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials is successful, it will publish its first paper nearly 10 years after the first West German publication. In a related development, a qualitative improvement in techniques for producing spin-polarized electrons occurred in 1974 with the invention at the Eidgenossische Technische Hochschule in Zurich of a source using photoemission from GaAs. This source was subsequently developed there and in West Germany, and in the United States at NBS and at the Stanford Linear Accelerator (SLAC). At SLAC, it was used in the landmark measurement of parity nonconservation in high-energy inelastic electron scattering. Despite the very successful application of the GaAs-polarized electron gun in materials studies such as spin-polarized electron scattering and spin-polarized inverse photoemission studies of surface magnetism, few groups are using this device because of its lack of commercial availability. This situation is expected to change, as PHI Corporation is now licensed to manufacture such polarized electron sources. Spin polarization analyzers have traditionally been large, cumbersome devices operating at energies of 100 keV. More compact analyzers recently have been developed at Rice University, and small low-energy analyzers have been developed in West Germany and at NBS. The NBS spin analyzer has been used on a scanning electron microscope to measure the spin polarization of secondary electrons emitted from a magnetic material and thereby obtain images of magnetic microstructures. A prototype commercial instrument of this type has also been built and tested by PHI and is expected to reach the market in the near future. Field Ion Microscope and Atom Probe The field ion microscope and the atom probe, which were developed by Mueller and his collaborators at Pennsylvania State University, produce images of individual atoms at the tips of very sharp needles. They provide good examples of how instrument development can thrive at a U.S. university and how the students in such a program can transfer the technology to other laboratories. Beam Scattering The U.S. scientific community has a distinct advantage over its foreign competition in beam scattering, especially in the area of chemical physics. The use of molecular beams to study the dynamics of chemical reactions at surfaces is yielding valuable information about sticking coefficients and state-selective adsorption and desorption. More generally, it is helping physical chemists discover the pathways by which reactions take place. In contrast to the situation in chemistry, however, physicists in the United States are
OCR for page 268
Materials Science and Engineering for the 1990s: Maintaining Competitiveness in the Age of Materials far behind research groups in West Germany in the use of these techniques. Exciting new surface physics is coming from elastic and inelastic scattering experiments, for example, the soliton-like reconstruction of the (111) surface in gold, and the observation of a soft phonon on the (100) surface in tungsten that might be related to a reconstruction. At present, U.S. experimental physicists are just watching.
Representative terms from entire chapter: