Part 3
CURRENT TOPICS IN MATERIALS RESEARCH



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



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 319
Advancing Materials Research Part 3 CURRENT TOPICS IN MATERIALS RESEARCH

OCR for page 319
Advancing Materials Research This page in the original is blank.

OCR for page 319
Advancing Materials Research Ensuring Contributions to Materials Science from Small-, Intermediate-, and Large-Scale Science Introduction HERBERT H.JOHNSON Materials is not to be thought of as a single discipline, but rather as a broad and vital field of knowledge and techniques that constitute an essential foundation stone of modern technological societies. In that respect, materials resembles other sprawling fields such as energy, communications, and medical science, each of which encompasses several disciplines and is characterized by intellectual ferment and enormous impact on society. The several cultures of materials research are a distinguishing feature of the field, a primary source of its intellectual richness and organizational diversity. In contrast to many disciplines the materials field in its present form is relatively new. The materials community has evolved rapidly from separate disciplinary bases in the past quarter century. This process of integration has brought a welcome, but still partial, coherence to the field. It is unlikely, however, that the materials community will ever coalesce into a single discipline. The intellectual and factual breadth of the field is simply too great to be confined within the boundaries of a single disciplinary structure. It is inevitable, then, that the materials field will on occasion appear disorganized, even turbulent, when compared with more tightly focused and hierarchical fields such as high-energy physics. Materials also differs from high-energy physics and astronomy, again to use them as examples, in the scale of instrumentation required for experi-

OCR for page 319
Advancing Materials Research mental research. Advances in fundamental problems in high-energy physics and astronomy require complicated and expensive instruments such as accelerators, storage rings, and telescopes (optical, radio, and orbiting). It is intrinsic to these fields that many experiments require large teams of researchers and a scale of coordinated effort that is absent in most other areas, including materials. Frontier materials research is, in fact, at present carried out in several modes. Small group research is prominent throughout the materials spectrum in universities as well as in industrial and government laboratories, and small group research continues as a vital source of forefront discoveries. In recent years interdisciplinary research directed toward specific goals, as pioneered by the Materials Research Laboratories (MRL) program, has become increasingly important, as complex materials problems have required the coordinated talents of several investigators. The MRL program has demonstrated the impressive results that can ensue when interdisciplinary groups work toward specific goals with the support of well-developed central laboratory facilities. Finally, a small but growing number of materials investigators are working at large machines, especially synchrotron radiation facilities, obtaining invaluable results that could not be obtained in any other way. This is small group research carried out in a big-science facility and context. These multiple research modes have arisen because of the increasing complexity of many frontier research problems in materials. Progress often requires the use of several techniques and the associated instrumentation. Interdisciplinary groups become an effective organizational strategy for tackling multifaceted problems. The development of centralized laboratory facilities is essential to minimize equipment costs and to maximize the use of expensive equipment, which should not and cannot be duplicated in every investigator’s laboratory. Each research mode makes a distinctive contribution to the overall strength of the materials field. Instrumentation will remain a major problem for the universities, not only for research, but for graduate education. The proper training of graduate students requires instrumentation that does not lag in quality and sophistication too far behind the equipment used in industrial and government laboratories. This is essential if new graduates are not to founder in their early professional careers. The cost of the necessary equipment continues to rise rapidly, placing a growing burden on university research groups. Unless present trends can be reversed, the number of universities with comprehensive and high-quality materials research programs will surely decrease in the years ahead. The conditions for funding of materials research have become increasingly tight and complicated in recent years. There has been a clear trend toward larger grants on more sharply focused topics, at the perceived cost of support to small university groups built around a single professor and his or her

OCR for page 319
Advancing Materials Research graduate students. Agency program managers appear to be under increasing pressure to turn over their programs in shorter time periods. They sometimes assume an active role in local program decisions, apparently again under pressure to produce specified results over a predetermined period. This perceived micromanagement of research has put the university system of small group research under additional strain. The time scale in which funding agencies expect significant research results is now equal to or less than the time required for a student to carry out a graduate thesis. This situation has made it much more difficult for faculty members to fund and manage their individual research groups. As a consequence, the university small research group appears to many to be an endangered species! Problems Facing Small-Science Research in Materials WILLIAM D.NIX The quality of materials research depends directly on the quality of the people doing it, whether it is done on a small, intermediate, or large scale. Thus, it is most important, and clearly in the national interest, to attract the brightest and the best to the field. The small-science research group is the basic unit around which graduate education programs are built, and from that perspective it is essential to the entire materials research and development area. Small-scale research groups typically have close contact with students who are not yet involved in research, so these groups carry the primary responsibility for recruiting for the field. The best candidates sometimes look for ways to be unique and to stand out. They are often idealistic and yet want to do something outstanding that will bear their name. Graduate education through the small-science research group route gives them the opportunity not only to develop their research capability, but also, and of equal importance, to develop intellectually and to prepare themselves for leadership in the field. For this reason alone, small group research is of central importance to the entire field. A major problem facing small research groups is the escalating need for instrumentation and associated support. The need for modern research instrumentation has been much discussed, is now widely recognized, and is being addressed through various instrumentation programs. Nevertheless, formidable problems remain, especially in the smaller universities. Some universities with substantial past accomplishments can no longer compete in top-rank materials research because of inadequate facilities.

OCR for page 319
Advancing Materials Research An equally formidable, even more expensive, problem is the need for vastly improved laboratory space and facilities to house future materials research programs. This problem is endemic across the science and engineering fields. Many universities are forced to put modern research programs into space that was constructed many years ago, usually for undergraduate instruction. The need for greatly expanded and improved facilities and the inability to generate the necessary funds through conventional sources have led some universities to approach the Congress directly for specific appropriations. The concomitant end run around the peer review system has generated a storm of controversy, which shows no sign of abating. It has also surely damaged the financial health of the programs approved through the peer review system. The universities are not well structured to handle the new instrumentation that is essential for advanced research in all fields of engineering and the physical sciences. Funds are generally not available for new or upgraded laboratory space, for service contracts, or for permanent staff to maintain and operate the increasingly complex new equipment. As a consequence, equipment is often operated at neither optimum specifications nor maximum efficiency. Of course, it is the formal responsibility of the universities to provide funds for these purposes, but they have been slow to realize that modern graduate research programs require new administrative and support structures and sources of funds. The problem is not handled well, even at major institutions. The Materials Research Laboratories program and the Materials Research Group (MRG) program, both administered by the National Science Foundation, have been a great help in this connection at the universities where these programs exist, but they provide only a small fraction of the help that is needed. It is sometimes suggested that principal investigators at universities should voluntarily include support personnel in their individual research budgets or apportion their research funds to take care of these needs. However, the system contains strong forces that make this suggestion impractical. Research funds for individual principal investigators are limited. Department heads and deans often expect faculty members to generate as much of their salary as possible from contract funds and also to support as many graduate students as possible. The keen competition for funds causes principal investigators to reserve their research funds for only those things that contribute directly to the scientific output of a given project. It is almost invariably counterproductive to individual programs to allocate funds to general support services. A generally acceptable solution to this problem is not yet evident. It may eventually be necessary to require major research universities to allocate a

OCR for page 319
Advancing Materials Research reasonable fraction of their funds to research support as a condition for receiving external support. Strong forces are operating to move university researchers away from the small-science mode and toward a team concept of research. These forces include (1) the need for instrumentation, (2) the necessity for sharing instrumentation, and (3) the increasing complexity of many advanced materials research problems. In addition, the funding agencies appear to be under steady pressure to justify their programs in terms of short answers to application-oriented problems. This trend has positive features, but it surely has a negative effect on the intellectual development of graduate students. The team concept does prepare students for some forms of industrial research, and it allows them to be associated with high-visibility projects. However, team research also very much restricts the opportunity for intellectual growth during thesis research, as the opportunities for exploratory and original research are usually limited. The planning and goal setting associated with team projects can on occasion reduce a graduate student’s role to that of a cog in a large machine. Prospective employers invariably ask about the originality shown by graduate students in their thesis research. They rarely ask about students’ ability to fit into a team, except in the context of their ability to get along with people. Originality is best developed and displayed in an unstructured environment. Students must have the opportunity to explore their own ideas and, on occasion, to fail. All evidence suggests that employers of graduate students are interested in people who have been encouraged to think independently and creatively and who are prepared for independent work. The MRLs and MRGs provide in their interdisciplinary thrust programs a satisfactory compromise between small-scale and team research. Often it is possible to develop a major thrust in a chosen area by clustering groups that operate in a small-science mode. The success of such groups depends on the personalities and interactive chemistries of the people involved. It is a satisfactory experience when it works well but a disaster when done poorly. The most successful collaborations are those that arise spontaneously. Continuity of support is becoming an increasingly serious problem for university researchers who work in the small-science mode. The research is conducted primarily by graduate students who take between 4 and 5 years to complete their studies, including the thesis. The time scale for this process has not changed significantly in 40 years and is not likely to change in the foreseeable future. Yet, the availability of grants or contracts that extend beyond 1 or 2 years is rare in today’s fast-paced world. It is not uncommon to see graduate students shifted from one project to another several times in the course of their studies. This is inefficient at best, and in some cases even destructive to the student involved. Small-scale research thrives on stable

OCR for page 319
Advancing Materials Research support that extends over the thesis lifetimes of several students. Most university researchers believe strongly that they have been most productive (as judged by significant papers published or doctoral degrees granted per dollar) in research programs for which support was provided over an extended period of time. It is often suggested in informal discussions that the development of a new idea in materials science takes a minimum of two graduate-student lifetimes. The first student explores the idea or effect, and the second brings it to fruition and develops the application. However, because the second part of the process depends on the success of the first, some projects would be expected to extend over several student lifetimes. In spite of the need for stable support, many funding agencies are not able to provide support over an extended period. This may be because of limited total funds, or perhaps because of a perceived need for rapid turnover in the subject matter in an agency program. In any event, their attention span is all too often much shorter than the characteristic time constant for small-science research. In some cases this means that the most pressing problems of the agencies are not addressed by the most gifted and productive university research groups. Academic materials research is supported almost wholly by the federal government; industry has not been a stable source of long-term funding. This may change as a result of rapidly growing interest in university-industry interactions. However, current university research is directed primarily to basic problems that are of interest to the federal government. This may occasionally lead to neglect of areas that are important to national economic strength. For example, the materials community has played a relatively minor role in the area of microelectronic materials. Magnetic materials is another area that has been neglected by the academic community. The increasing industrial interest in academic materials research may in time lead to a more balanced national materials program. To the university practitioners of small-scale science, it appears that support for small-scale science is being continually eroded in favor of big science. The reasons for this are both political and sociological. First, it must be acknowledged that many exciting problems in science require large facilities for their solution. However, it is also true that major projects and big science come naturally to the attention of policymakers in the top ranks of government, especially when they are presented by a persuasive and prestigious group of scientists. Furthermore, the big-science communities are considerably more cohesive, essentially because their research progress depends critically upon the development and operation of large facilities. Hence, there is a strong internal driving force that leads big-science communities to develop a tightly focused set of priorities and to present a united front in the never-ending quest for funds.

OCR for page 319
Advancing Materials Research In contrast, small-science communities such as materials are inherently more decentralized, for the availability of large facilities is not the primary determinant of research success. In materials there are many areas where exciting research progress is possible; some require extensive instrumentation and some do not. Consequently, materials programs appear throughout the budgets of the agencies, but only rarely at a level that attracts the attention of top policymakers. Furthermore, there is no single widely acknowledged organization that can speak for the materials field and convey an authoritative sense of its prospects, accomplishments, and needs. Indeed, researchers in small-science communities are more commonly critical of their colleagues than supportive. This is a problem that the materials community must address. Basic Research Supported by Mission Agencies MILDRED S.DRESSELHAUS A problem that affects all of the scientific communities, including materials, is the question of how to maximize the effectiveness of the basic research programs supported by the mission-oriented agencies. Independent and persuasive studies indicate that the cost of research has been increasing consistently by about 65 percent more than the Consumer Price Index, independent of what the Consumer Price Index is doing at any instant in time. When that fact is considered in relation to the budgeting trends in federal agencies, the only conclusion that can be reached is that there will shortly be a decline in the number of people who will have the privilege of pushing the frontiers of materials science forward. The materials research community for the first 25 years of the Materials Research Laboratory program has operated on the premise that the federal establishment would continue to provide support on a more-or-less one-way basis. There is of course a different approach, one in which the research community takes the initiative and provides a much more comprehensive rationale for supporting basic research. The following suggestion has less to do with small science, intermediate science, or big science, individually, than it does with the entire research community and the way in which it should relate to the larger technological enterprise. The suggestion is to place funding of basic science more on a basis of mutual benefit. The core idea of the proposal comes from an experience that most researchers have had at one time or another—consulting for private industry if they are university faculty members or interacting with university

OCR for page 319
Advancing Materials Research faculty members if they work in industry. Similar relationships hold for staff members of the federal research laboratories. The proposal is to encourage senior investigators, selected from among the basic research grantees, to visit appropriate groups in the mission agency laboratories for a few days each year to share the experience and expertise gained from years of research in the field. The senior investigators participating in the proposed program would normally be university professors. Many would have significant experience as consultants to private industry; their interaction with the R&D groups in the mission agency would be similar to that of consultants. Participation in this program would of course be voluntary, although in the aggregate it might be expected that about 40 percent of the qualified investigators would participate after receiving research funds from the mission agency for an extended time, perhaps 5 years. Young investigators with less than 10 years of professional experience would normally not be expected to participate. The program might be especially attractive to “elder statesmen” of science, or people who have gone far enough in their careers that they can afford to spend a week or more per year in this kind of activity. The proposed program has essentially three objectives. The first is to enhance the cost-effectiveness of all programs—the university programs and the programs at the government laboratories, whether they be DOD, DOE, or other federally funded laboratories. There would be a clear gain if this program would enhance the cost-effectiveness of the R&D activities of the laboratories where most of the expenditures of the mission agencies are directed. In this way the basic research programs would gain leverage, and there would be a stronger justification for the expenditures necessary to maintain an effective basic research program in each agency. An expanded justification is desirable, as the cost of research continues to increase, while rapid scientific and technological breakthroughs continue to expand the opportunities for exciting basic research. The second objective is to develop much stronger bridges of communication between the basic research community and the mission agencies. The benefits of the proposed program would flow in both directions. The results of basic research would be brought in a timely and effective way to the development efforts. At the same time, contact with applied programs often leads to a recognition of new and exciting areas of basic research that are ripe for exploitation. An important additional benefit is that research scientists would be much more aware of the activities in the mission laboratories. This knowledge is important and useful in providing advice to students about the scientific challenges and opportunities that careers in the mission agencies can provide. The third objective is to broaden understanding and appreciation of the role of basic research, and in this way to accomplish two things: the first is

OCR for page 319
Advancing Materials Research simply to increase the total amount of resources going into basic research by making it more cost-effective; the second is to buttress the role of basic research so that it can provide even more effective contributions to the technological strength of the nation. To implement this proposal a pilot program with a small number of participants should be established to evaluate the concept and to learn from early experience. If that evaluation shows that the program would be viable on a national scale and of mutual benefit to enough members of the research and development community, then the program should be enlarged and extended to all who wish to participate. The Two Domains of Materials Science ALBERT M.CLOGSTON Materials science is a highly interdisciplinary field consisting of diverse specialties, including physical metallurgy, solid-state physics, solid-state chemistry, ceramic science, polymer science, materials preparation, and materials analysis. Other individuals would no doubt construct somewhat different lists, depending on their perspective, but that is an indicator of the richness and diversity of the field. However, these specialties tend to divide into two separate domains, the microscopic and the macroscopic. The microscopic view is concerned mainly with atoms and molecules and the electromagnetic forces that bind them. There is a strong emphasis on such topics as electronic structure, lattice vibrations, and the many interactions of radiation and particles with condensed matter. The macroscopic point of view focuses on the properties of matter in bulk, with typical topics such as microstructure, phase transitions, continuum behavior, and mechanical properties. These two ways of thinking about materials tend to be vertically integrated with respect to measurements performed, instrumentation used, phenomena studied, and the technologies to which they lead. It is also true that few researchers cross the boundary between these two domains, although those who do often make strong contributions. In the microscopic domain, which includes solid-state physics, the materials and phenomena studied, and the kinds of instrumentation and measurements required, tend to be associated with what are often described as high-technology industries and materials. With some exceptions these materials are used for their electronic, magnetic, or optical properties. In contrast, research at the microstructure or continuum level leads to technologies

OCR for page 319
Advancing Materials Research that use high-performance materials, developed primarily for their mechanical properties, often under a wide variety of rigorous operating conditions. There are tremendous opportunities to advance the science of materials by horizontally integrating studies of the phenomena that are of interest in the microscopic and macroscopic domains. The integration that has occurred over the past quarter century is impressive, but the full potential of the field has not yet been realized. For example, physical metallurgy and solid-state physics have much to say to each other about such topics as interactions at surfaces, fracture, dislocation physics, and electronic materials. Many other examples could be cited. Both physical metallurgy and solid-state physics would derive vast benefits from closer interaction with solid-state chemistry. As the previous discussion indicates, there is a close connection between materials science and basic materials technology. This tight coupling is one of the striking characteristics of materials science, and certainly one of its greatest strengths. It is the reason why materials science has been the source of major contributions to other sciences and, perhaps even more importantly, to industrial innovation, and why it has such potential for future contributions. The strong coupling of materials science and technology leads to a second major point, which is the critical role played by basic technology as a link between research and development. This somewhat unconventional view of the research and development process is nevertheless the view of research and development held, at least implicitly, by most of the major industrial laboratories, and also in a formal way by the Department of Defense and the Department of Energy. Basic research as defined by those agencies, for example, can be read to include not only the increase of basic knowledge, but also the increase and enlargement of the technology base for exploratory and advanced development. Basic technology should be recognized as an important research activity, and as the critical link between research and development. This leads to the proper place for materials research in the overall research and development process. Basic materials science and basic materials technology should both be regarded as research activities in the research and development process. They couple to basic science and basic technologies coming from other sources to make possible the exploratory and advanced development of systems of all kinds, including systems for communications, energy, national security, and transportation. It is important that basic technology be recognized as a legitimate research activity. It is carried out by the same kinds of people who do basic research for new knowledge. They use the same kinds of instrumentation and the same research methodologies. They are the people who, in industry, do basic research one day and basic technology the next.

OCR for page 319
Advancing Materials Research New Demands on Materials Science PRAVEEN CHAUDHARI Materials science, drawn from studies at the scale of atoms to macroscopic bodies, encompasses much of what we know about the physical world. To cite two examples: the laws of thermodynamics have proved useful not only in designing engines but also in understanding chemical reactions, and quantum mechanics is essential to understanding many scientific phenomena as well as the operation of the silicon transistor. Materials science is characterized further by the role of empiricism in the practical use of knowledge. It is sometimes believed that if perfect understanding were available, then and only then could a perfect device, or mechanism, or structure be built. However, as those who are knowledgeable about industry know, technology is often at the same level of advancement as science, and occasionally is ahead of it. Thus, scientific understanding and the building of new devices may go hand in hand, with a substantial assist from empiricism. The interdisciplinary nature of materials science gives rise to the broad scope of its activities and to its importance. This is also true of other interdisciplinary fields such as medical science and computer science. There are also differences between these fields that must be recognized. In medical science the issues of purpose are well recognized by society. For instance, no one would dispute that to find a cure for cancer is a worthy goal. There is broad and intense interest in knowing how the brain or the human body functions. There is also a sense of immediacy in the medical sciences: a cure for cancer or AIDS is an urgent need. Computer science differs from medical and materials sciences. It stands in relation to its future much as materials science did before the laws of thermodynamics were discovered. The laws for computer science are still being discovered. It is a nascent, exciting science that will evolve with all of the complexity that is found in materials science. Materials science is sufficiently complex that to one unfamiliar with the field it appears diffuse and aimless. There are no specific goals and no sense of urgency. Materials researchers need to articulate their role in society. We at the Research Division in IBM have attempted to do this. In so doing we have found it useful to divide scientific work into two categories, called area science and general science. In area science, scientists and technologists jointly study a particular technology and extract the key technical issues for today and for the future. Those key technical issues are then examined to extract what is called essential, or generic, science—the knowledge that is needed to develop or evolve tech-

OCR for page 319
Advancing Materials Research nology. Thus, there are two key elements in the process: first, to identify the technical issues and, second, to identify the generic science. Using this approach, we have found that continuing progress in electronic devices—from data storage to the central processing unit of a computer— depends crucially upon materials and processing sciences. By processing of materials we mean, for example, adding or removing atoms where and when desired. There are many ways to add atoms, including crystal growth, chemical vapor deposition, vacuum deposition, molecular beam epitaxy, sputtering, and electroplating. There are also many processes by which atoms can be removed. Let us use an etching process as an example of how generic science issues are developed in a given area of science. In the electronics industry, reactive ion etching, an emerging process that is attracting much attention, illustrates the complex demands placed on materials science by advanced technology. Reactive ion etching consists of applying a voltage across charged species in a plasma to accelerate ions, which hit the surface of a substrate. By shielding various areas of the substrate with a “resist,” the substrate can be etched in a directional fashion. Structures can then be constructed by selective deposition of materials into the cavities formed by the original etching treatment. The density of the plasma used in reactive ion etching lies between the density of matter in intergalactic space and that in nuclear fusion. The chemical and physical properties of the plasma of interest in reactive ion etching are not well known. Moreover, the radicals that exist in these plasmas are not well identified. Until recently, techniques for identifying the chemical species both spatially and temporally were not available. After the radicals have been identified, the next problem is to investigate the mechanism by which they interact with the substrate. Why is a particular material etched more efficiently than another? Why do polymers behave differently from metals? Why does p-type silicon behave differently from n-type silicon? The etching reaction occurs not only on the surface of the substrate but also beneath the surface. In fact, the atoms penetrate below the surface. They can be found tens to hundreds of angstroms deep, depending on how the process is carried out. It is important to understand this process in detail, for not only is it desirable to have very clean substrates on which to deposit a substance in a controlled way, but it is also important to be able to produce damage-free regions near the surface of a semiconductor material. One can ask the following question: If an atom or molecule hits a surface, how does it lose its energy? This question leads to many more detailed questions. What are the modes of energy transfer that apply here? Is there chemisorption or physisorption? How do atoms diffuse near a surface when

OCR for page 319
Advancing Materials Research a charge is present? Such questions transform a mundane, practical process into a series of questions of fundamental scientific interest. To go a step further, there are many processes other than reactive ion etching that require understanding a great deal about surfaces and about particle interactions with them. Such understanding is important not only to the computer and electronics industries but also to processes ranging from electroplating, to catalysis, to the evolution of hydrogen in the universe from the atomic to the molecular state. The study of complex phenomena and processes in industrial technology suggests two important points. The first is that within a given area of science there must be a spectrum of activities that proceed from science to technology. These activities should be evaluated on the basis of their value to society, not on the basis of some arbitrary criterion by which “basic science” is deemed more acceptable than “applied science.” Moreover, distinctions between big science and small science are irrelevant when studying a problem as complex and important as reactive ion etching. Both kinds of science are frequently needed in modern industrial research. In the case of reactive ion etching, many of the modern techniques of materials research are necessary. These include Rutherford backscattering, ion scattering, synchrotron radiation, various surface spectroscopies, nuclear resonance, and transmission electron microscopy. An important point that cannot be taken for granted or emphasized enough is that the research enterprise of the nation requires an infrastructure that nurtures general science, or science that cannot be identified at present with any particular area of application. This provides the freedom to move freely in a spectrum of specific activity according to the merit of the question being pursued. In materials science three recent developments illustrate the importance of such freedom. The first is the scanning tunneling microscope, which evolved from a desire to improve understanding of the uniformity of dielectrics. When it was shown, however, that atomic resolution could be achieved, the research was redirected into much broader areas of atomic and electronic structure of surfaces. The second example is the quantum Hall effect, which is leading to a better understanding of the behavior of electrons in matter, especially in lower-dimensional systems with various degrees of disorder. The third is the discovery of quasicrystals, which may or may not represent a new structural state of matter but must surely be studied and understood.

OCR for page 319
Advancing Materials Research This page in the original is blank.