Click for next page ( 2


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 1
I Highlights, Oand Needs The focus in this volume is solely on condensed- matter physics, which is the foundation of a significant portion of the broader field of materials science, and the dividing line between the two fields is not always a sharp one. However, we are not surveying materials science nor the considerable impact of condensed- matter physics on technology. The interface between physics and technology will receive fuller treatment in another volume in this survey.

OCR for page 1

OCR for page 1
HIGHLIGHTS, OPPOR TUNI TIES, AND NEEDS 3 CONDENSED-MATTER PHYSICS AND ITS IMPORTANCE Condensed-matter physics is the fundamental science of solids and liquids, states of matter in which the constituent atoms are sufficiently close together that each atom interacts simultaneously with many neighbors. It also deals with states intermediate between solid and liquid (e.g., liquid crystals, glasses, and gels), with dense gases and plasmas, and with special quantum states (superfluids) that exist only at low temperatures. All these states constitute what are called the condensed states of matter. Condensed-matter physics is important for two reasons. The first is that it provides the quantum-mechanical foundation of the classical sciences of mechanics, hydrodynamics, thermodynamics, electronics, optics, metallurgy, and solid-state chemistry. The second is the mas- sive contributions that it provides to high technology. It has been the source of such extraordinary technological innovations as the transis- tor, superconducting magnets, solid-state lasers, and highly sensitive detectors of radiant energy. It thereby directly affects the technologies by which people communicate, compute, and use energy and has had a profound impact on nonnuclear military technology. At the fundamental level, research in condensed-matter physics is driven by the desire to understand both the manner in which the building blocks of condensed matter-electrons and nuclei, atoms and molecules combine coherently in enormous numbers (~1024/cm3) to form the world that is visible to the naked eye, and much of the world that is not, and the properties of the systems thus formed. It is in the fact that condensed-matter physics is the physics of systems with an enormous number of degrees of freedom that the intellectual challenges that it presents are found. A high degree of creativity is required to find conceptually, mathematically, and experimentally tractable ways of extracting the essential features of such systems, where exact treat- ment is an impossible task. Condensed-matter physics is intellectually stimulating also because of the discoveries of fundamentally new phenomena and states of matter, the development of new concepts, and the opening up of new subfields that have occurred continuously throughout its 60-year history. It is the field in which advances in quantum and other theories most directly confront experiment and has repeatedly served as a source or testing ground for new conceptual ways of viewing complex systems. In fact, condensed-matter physics is unique among the various subfields of physics in the frequency with which it feeds its fundamental ideas into other areas of science. Thus, advances in such

OCR for page 1
4 HIGHLIGHTS, OPPORTUNI TIES, AND NEEDS subareas of condensed-matter physics as many-body problems, critical phenomena, broken symmetry, and defects have had a major impact on nuclear physics, elementary-particle physics, astrophysics, molecular physics, and chemistry. These advances continue and over the promise of equally fundamental discoveries in the next decade. At the same time, condensed-matter physics excites interest because of the well-founded expectations for applications of discoveries in it. Of all the branches of physics, condensed matter has the greatest impact on our daily lives through the technological developments to which it gives rise. Such familiar devices as the transistor, which has led to the miniaturization of a variety of electronic appliances; the semiconductor chip, which has made possible all the myriad aspects of the computer; magnetic tapes used in recording of all kinds; plastics for everything from kitchen utensils to automobile bodies; catalytic con- verters to reduce automobile emissions; composite materials used in fan jets and modern tennis rackets; and NMR tomography are but a few of the practical consequences of research in condensed-matter physics. A whole new technology, optical communications, is being developed at this time from research in condensed-matter physics, optics, and the chemistry of optical fibers. These examples serve to illustrate the intimate connection between fundamental science and the development of basic new technology in condensed-matter physics. In both universities and industry they are carried out by people with the same research training, who use the same physics concepts and the same advanced instrumentation. Be- cause fundamental science in condensed-matter physics is so deeply involved with technological innovation, it has a strong natural bond with industry. This is the main reason why condensed-matter physics has been so successful in leading industrial innovation. Indeed, the full extent to which the consequences of research in condensed-matter physics play a role in the quality of our everyday lives, and in meeting national needs, is far greater than any such listing can indicate. In order to show this explicitly we have constructed the matrix displayed in Table 1, the first column of which lists the subareas of condensed-matter physics, and the first row the major areas of human and technological activity that are of national interest. The elements of the matrix are filled in with a solid circle, indicating a critical connection between the corresponding subarea of condensed- matter physics and the area of application; a half-filled circle, indicating an important or emerging connection; an open circle, denoting the possibility of a connection; or a blank, implying that the connection is not known. In Appendix A this matrix is repeated, but with qualitative

OCR for page 1
s 'e ~ . z v, 'e o - 'e me o a, o - 'e c) - 'e Go v - ~ - 'e A, o v Am o 'e 50 'e Id m Cal o ._ v o m `1, ~ i_ ~4 _ _. O O C) ~ Ct C) ~ i> V) o . o C Ct C) . 00 I: 't ~ .o ~ .= IS C) lo O V) . Ct Cal E 0, _ ~ so Ct V) 0 .~00~ 00 0~0~0 0 ~0~0 0~ 0~ ~0~0 G0 0 ~ ~. o u, o . - 1 CL O o ~ ~ c) c~ ao . - - `: ~ "o o 5o u, ~: c~ . - o c: . o c~ . c) e ~ - - U3 ~ c;5 ~, o - . c~ ~ v - ) ~ ~ ~ o o ~ o c~ c) . - u' "c cn o - ~ ~ ~v . - ~s - i ~s~ ~z .- 3 ~ o -v ~ - _ - o - U' Ct ._ ~S ~ Ct o ._ .^ _ ~ C) o ~ - c Ce ~ C) o _ U. _ o ~ C;~ - Ct ~ ~D ~ e~ V . c: {: ~ o .C) C~ o ~ Ct Ce o 1 _ _ ._ U) ao C: o ~ C) ,,= C~ _ o C~ Ct ~ o ~ V U. a> s~ _ O 3 ~ _ o Ct ~ .^ o ._ o _. ~ C) s: ;: C: o s:: C) o _ ~ .o ~ - o~ Ct ~ U, a~ Ce o C) _ ._ ~ V C<, o .O .S .V C<: o .O ._

OCR for page 1
6 HIGHLIGHTS, OPPORTUNITIES, AND NEEDS comments concerning the connections replacing the various circles. This table makes the point graphically that condensed-matter physics plays an indispensable role in the maintenance of the quality of our daily life and in providing for national security. DISCOVERY The 1 950s saw such achievements as the rapid development of semiconductor technology after the discovery of the transistor; the rise of many-body theory (the application of the methods of quantum field theory to large and complex systems) as a field of theoretical physics, and its crowning achievement, the solution of the 50-year-old problem of superconductivity; the heyday of magnetic resonance methods in physics; and the elucidation of the Fermi surfaces of metals. The 1960s saw the discovery of high critical fields and superconducting magnets, as well as of the Josephson erect and other electron tunneling methods and devices; the construction of the first working lasers and further giant strides in laser physics; the initial explanation of the ancient problem of the resistance minimum by Kondo and the opening up of a whole new physics of similar Fermi-surface effects in metals such as the x-ray edge; the development of pseudopotential and density functional methods, among others, that have made electronic structure calculations almost routine; and the initial development of high-energy probe methods for the study of electronic structure such as ultraviolet photoelectron spectroscopy (UPS) and x-ray photoelectron spectros- copy (XPS). From time to time, there have been those who have predicted the end of this era of discovery. Remarkably, the subject continues to produce surprises. In what follows we present a selection of some of the most interesting advances in condensed-matter physics that oc- curred in the 1970s and early 1980s. Artificially Structured Materials One area of condensed-matter physics that has progressed remark- ably in the past decade is that of artificially structured materials- materials that have been structured either during or after growth to have dimensions or properties that do not occur naturally. The most important techniques for the creation of such materials are molecular-beam epitaxy (MBE), the molecule-by-molecule deposition of material of the desired composition from a molecular beam' and

OCR for page 1
HIGHLIGHTS, OPPORTUNITIES, AND NEEDS 7 metallo-organic chemical vapor deposition (MOCVD). These are prime examples of technological breakthroughs, used primarily to make semiconductor lasers and other devices, feeding back to fundamental physics. One can fabricate artificial periodic superlattices consisting of alternating layers of different semiconductors, different metals, or semiconductors and metals, and one can also create artificial, purely two-dimensional electron gases. The latter have unique and important properties, e.g., extremely high electron mobilities, which cannot be provided by metal-oxide-semiconductor (MOS) inversion layers. The new physical phenomena to which the resulting structures have given rise include the quantized Hall effect and the fractionally quantized Hall effect. It has also been possible to grow metallic superlattices in which the electronic mean free path is appreciably longer than the period of the superlattices (the sum of the thicknesses of the two alternating metal layers). It is found that it is possible to induce new lattice structures rather easily in such superlattices. Metal/insu- lator superlattices are ideal systems for the study of dimensional effects in metals, e.g., the crossover from two- to three-dimensional superconductivity in Nb/Ge superlattices as the Ge thickness is de- creased. The Quantized Hall Eject Modern technology has made possible unique, purely two- dimensional electron gases (in the sense that only one quantum state is excited in the direction perpendicular to the plane of the gas, so that electronic motion in it is strictly confined to that plane). These systems show exciting properties and are a new laboratory for the study of fundamental physics. The most remarkable property of such systems is undoubtedly the quantized Hall eject. At low temperature and high perpendicular magnetic field, the electron states are split into so-called Landau or cyclotron energy levels. It is found that when the Fermi level is between two such levels one sees an almost perfectly flat plateau or constant value of the Hall conductance, the conductance perpendicular to the electric and magnetic fields, as well as zero parallel conductance. These plateaus are found to be quantized in units of e21h = 1/25,812.8 ohm-~. The precision of this result, at least one part in a hundred million, has led to improvement in the measurement of this fundamental constant and to a new portable resistance standard. More recently, quantization of the Hall conductance in simple fractions like 1/3, 2/5, and 2/7 of e21h has been seen, and an explanation of this

OCR for page 1
8 HIGHLIGHTS. OPPORTUNITIES, AND NEEDS effect has been proposed, and widely accepted, that involves a completely new and unexpected ordered state of matter. In this state one proposes that a new type of elementary excitation with fractional electronic charge plays a major role. Ejects of Reduced Dimensionality For many years condensed-matter theorists studied one- and two- dimensional models of solids because it was often possible to obtain exact results there where the corresponding, physical, three-dimen- sional models were intractable. The existence of such exact solutions in low-dimensional systems has prompted experimentalists to search, successfully, for physical systems whose physical properties agree well with those of one- and two-dimensional theoretical models. These include quasi-one-dimensional magnetic systems composed of chains of magnetic atoms, separated from each other by nonmagnetic atoms, and quasi-two-dimensional systems realized by layered compounds, such as graphite intercalation compounds, in which atomic layers are widely separated and weakly interacting. Other examples have arisen either out of technological discoveries or from the synthesis of interesting new materials. The inversion layers used in the quantized Hall effects are an example of reduced dimen- sionality systems important in technology, an example that has been vital to the physics of disordered systems as well. Another is the development of methods for studying adsorbed layers on surfaces that undergo phase transitions of typically two-dimensional type. A third is the discovery of methods for making freely suspended layers of a liquid crystal one or a few molecules thick. New materials showing metallic properties in only one or two dimensions have been synthesized, for instance the transition-metal dichalcogenides, which can be cleaved to produce single layers or intercalated with large molecules that separate the layers by large distances, and a number of organic one-dimensional chain metals such as polyacetylene. These various developments have encouraged ex- perimentalists and theorists to think of dimensionality as a new free parameter. Charge-Density Waves Among phenomena that are most clearly demonstrated in low- dimensionality systems are charge (or in some cases spin) density waves. A few isolated cases in which the structure of a solid was

OCR for page 1
HIGHLIGHTS, OPPORTUNITIES, AND NEEDS 9 modulated periodically had been known for decades, but it was not until modern low-dimensionality materials became available, such as the dichalcogenides and trichalcogenides of Nb and other transition metals, and some organic metals, such as polyacetylene and tetra- thiafulvalene-tetracyanoquinodimethane (TTF-TCNQj, that the phe- nomenon could be studied in general. Theory has predicted for many years that such materials should show density waves especially easily. In such materials the structure contains two periods that may be incommensurate, hence giving an overall nonperiodic structure. A particularly important possibility is the sliding of such an incommen- surate wave through the parent lattice, a new phenomenon illustrating in a clean microscopic model the age-old effects of sliding and sticking friction. The materials that display these ejects, e.g., NbSe3 and TaS3, are remarkable quantum systems with the richness of superconductiv- ity and should be excellent for studying various aspects of macroscopic quantum phenomena. Other interesting phenomena relate to defects in these waves, which have strange topological properties, fractional charge per unit area, and, in the case of polyacetylene, strange spin and charge properties. This subject continues to be actively discussed, not only because of its scientific interest but also because of its possible technical interest. Disorder It is only within the past decade that physicists have begun to focus on the problems intrinsic to disordered states of matter such as random alloys, glass, and gels. Historically they had dealt with such systems- often effectively by trying to average out the disorder in the most efficient possible way, to produce an '`effective medium." Now they have begun to look for intrinsic properties of disordered materials. The most striking of these is localization, the tendency to form quantum states that cannot move except with the help of thermal energy. Experimentally, the study of localization is much clarified by using a two-dimensional geometry, in which one often sees a unique nonclas- sical behavior of the electronic conductivity, and by technical ad- vances in microfabrication, which allow the study of effectively one-dimensional wires and of tiny loops that show strange conductivity -oscillations in a magnetic field. A second disordered material of technical importance is glass; the glass transition and the high- temperature annealing properties of glass remain almost completely mysterious, but a whole new physics has grown up around a new entity recently discovered in the low-temperature behavior of glass, the

OCR for page 1
10 HIGHLIGHTS, OPPORTUNITIES, AND NEEDS so-called tunneling centers. The structure of glass is also a great mystery; the computer may help in deciphering it, but in fact we know so little that we do not yet even believe we can program a computer to make a viable model of glass. Mixed Valence and Heavy Fermions It is not uncommon, the chemists have found, for the same chemical element to exhibit two valences in the same compound as, for example, magnetite, which contains both ferrous and ferric iron at different atomic positions. On the other hand, metals such as nickel do not necessarily have a fixed valence, as the electrons move freely through the lattice. The rare-earth metals, however, normally have a fixed valence for the inner f electrons, which can be identified because they show magnetic properties identical to those of ions in an insulating salt. It now appears that there is a large class of compounds based on the rare-earth atoms Ce, Sm. Eu, Tm, Yb, and now the actinide element U. that are intermediate between these two cases in an unusual way. Some types of measurements one-electron probes, x-ray edges show both valences simultaneously developed on the same atom. Other types of measurement, such as those of low-temperature magnetism or conductivity, show a fixed valence, sometimes interme- diate and sometimes not. It appears that electrons are quantum mechanically tunneling rather slowly in and out of the f shells, with very exotic results, such as electron bands with effective electron masses as large as 1000 times a normal electron mass, which nonethe- less exhibit superconductivity at very low temperatures. Present speculation is that these superconductors are of a totally new type, and are analogous to superfluid 3He. The valence fluctuations in other materials lead to a number of other fascinating effects: metal/insulator transitions, magnetic/nonmagnetic transitions, soft (highly compress- ible) lattices, and transitions into exotic magnetic ground states. A full explanation of these phenomena might have far-reaching consequences for our understanding of magnetism and bonding in solids. The Superfluid Phases of 3He A high point in research in condensed-matter physics of the last decade was the discovery that 3He is a superfluid (i.e., can flow without resistance through narrow channels) at temperatures below 3 mK. This is the first, and only, new superfluid to be discovered since the

OCR for page 1
HIGHLIGHTS, OPPORTUNITIES, AND NEEDS 11 superfluidity of 4He was established in 1937. The properties of 3He are very different from those of 4He because 4He obeys the quantum- mechanical laws of Bose statistics, whereas 3He obeys Fermi statistics, the same as electrons. At the same time, superfluid 3He displays a rich variety of physical properties in addition to those possessed by the previously known superfluids. This is because the interaction between pairs of helium atoms that is responsible for the superfluidity of 3He is qualitatively different from the interaction between pairs of electrons responsible for the superconductivity of all the currently known superconductors. In particular the superfluid is locally anisotropic, acting as though it was made up of molecules with internal rotational motions about a specific direction. Several major advances in condensed-matter physics were fueled primarily by new theoretical concepts. Descriptions of two of them follow. The Renormalization Group Methods These techniques are useful in dealing with physical phenomena in which there exist fluctuations that occur simultaneously over a wide range of different length, energy, or time scales. The method proceeds by stages, in which one successively discards the shortest-wavelength fluctuations until a few macroscopic degrees of freedom remain. The effects of the short-wavelength fluctuations are taken into account approximately at each stage by a renormc~lizc~tion, i.e., change in magnitude, of the interactions among the remaining long-wavelength modes. These techniques were developed initially in particle physics but came into their own in the theory of phase transitions, the branch of condensed-matter physics that deals with changes of state, such as the melting and freezing of solids and liquids and the magnetization of ferromagnets. Their use has provided a theoretical understanding of empirical relations among different properties near the phase transition or critical point of a given system and has made it possible to predict critical properties with a high degree of accuracy. These predictions have been confirmed by a wide variety of subsequent experiments. The renormalization group techniques have found applications in such diverse areas of condensed-matter physics as disordered electronic systems, impurity problems, disordered magnetic materials called spin glasses, nonlinear dynamical systems, long polymer chains, and per- colation through macroscopically inhomogeneous systems such as porous rocks.

OCR for page 1
26 HIGHLIGHTS, OPPORTUNI TIES, AND NEEDS research in condensed-matter physics more rewarding but also more expensive. A supercomputer, such as the CRAY-XMP, costs around $10 million, and at present only three universities in the United States have one. A computer of the type of a VAX 11-780 costs in the vicinity of $200,000. Even if research groups do not own their own computer, the cost of purchasing computer time can be a significant portion of the cost of present-day research. The future health of the field requires that this clearly identified need for computers be accommodated. As computing needs are diverse requiring both increased capacity and capability they cannot all be filled by the medium-sized main- frame computers found on most research campuses. There is a specific need in theoretical work for advanced computing capabilities. Some of these can often be met by adding a fast processor to a conventional computer at a cost of about $40O,OOO. Almost an order of magnitude or more additional computing power can be provided by modern supercomputers and, in some fraction of cases, the provision of time on such machines if not the machines themselves for condensed- matter research is essential. The future funding patterns must accom- modate the growing demands for computer use. To this end, We recommend that sufficient new monies be appropriated to allow the several federal agencies supporting condensed-matter research to identify a continuing fraction of the total budget to be devoted to the special computing needs of condensed-matter research. The assignment to computing of 10 percent or more of the total present budget would appear well justified. The funds so assigned could be used for the purchase of computer time or for the purchase and maintenance of dedicated equipment. In view of rapid changes in computer technology and patterns of use in the physics community, this fraction should be reconsidered after the next few years of experience. The scientific community and the federal funding agencies should work together to promote more effective use of major computer resources through networking, standardization, and the establishment of user assistance groups. FUNDING The chances of realizing the research opportunities that the coming decade offers will be significantly enhanced by an increase in the number of individuals carrying out this research, an increase in the level of support that they receive, and the provision of the increasingly more sophisticated, and the increasingly more costly, equipment that they will need in their work.

OCR for page 1
HIGHLIGHTS. OPPORTUNITIES, AND NEEDS 27 The costs of conducting a modern research program include the maintenance of equipment, the operating or running costs, and the salaries and benefits for personnel. These costs will increase more rapidly than inflation because of the increased sophistication of the required equipment and the expected increases in tuition for graduate students. The national investment required for the adequate support of basic research in condensed-matter physics by individual researchers, how- ever, is not great, even though the return on the investment is large. There is heartening evidence in the current federal budget, through its approximately 20 percent increase in support for basic science over the level for last year, that this is recognized. However, more still needs to be done to capitalize on the opportunities that exist. We estimate that implementing the preceding recommendations will require an increase in funding for research in condensed-matter physics at a steady annual expansion rate of approximately 20 percent in constant dollars for an additional 3 years. We strongly recommend that this increase take place. The special claim of condensed-matter physics for research support from federal and industrial sources lies in its record of converting deep science into benign and sophisticated industrial technology on a time scale that is often no more than 5 to 10 years. This process is still vigorously under way with such notable new scientific discoveries as the quantized Hall eject, valence fluctuations, heavy electron-mass metals, electron localization due to disorder, artificially structured materials, conducting polymers, chaotic phenomena in solids and liquids, and solitary wave phenomena in solids. If the resources become available to carry out the research necessary to exploit these new discoveries, the impact An industrial technology will be even greater than what has gone before. Support for National Facilities Some of the national facilities are comparatively new; others have been in existence for many years. Because of their importance for the nation's scientific effort, the facilities that continue to maintain a high level of scientific excellence should be adequately supported. Planning for new facilities to meet the needs of new areas of condensed-matter physics that are now developing must begin in the near future. The needs of the neutron and synchrotron facilities have been subjected to detailed scrutiny recently by several panels sponsored by the NSF and the Department of Energy (DOE). The most recent of

OCR for page 1
28 HIGHfIGHTS,OPPORTUNITIES,AND NEEDS these studies* was prepared while this report was being written. We will have occasion to refer to it in what follows. NEUTRON FACILITIES The existing high-flux reactors, the cornerstones of the U.S. neu- tron-scatter~ng program, are underfunded and understaffed. Relative to their Western European counterparts they are falling seriously behind in instrumentation. Therefore, We recommend that a concerted and coordinated effort should be undertaken to expand the effectiveness of our high-performance reactors by adding new, diversified instruments along with personnel necessary to design, build, and utilize them in the user mode. We estimate that at least ten new instruments are needed, requiring an increase in annual operating costs of $2 million to $3 million for manpower needed for their design and use. About $20 million to $30 million is required for building such instruments, to be spent over 5 to 7 years. Instrumentation plans beyond the level projected above may be warranted but should be justified by demonstrated user needs. Note that this estimate does not attempt to address the somewhat different needs of the chemistry and biology communities. A 1984 Panel on Neutron Scattering, considering the total scientific commu- nity, estimated a need for ~30 new instruments." Spallation sources provide new opportunities to expand the power of the neutron as a probe of condensed matter. The United States currently has two pulsed spallation sources. The Los Alamos Neutron Scattering Center (LANSCE) facility at the Los Alamo s National Laboratory is compromised currently by the pulse structure of the LAMPF proton beam that supplies it. This situation will be corrected by the addition of a proton storage ring (PSR) scheduled for completion in 1986. It is also restricted by the small experimental hall. The Intense Pulsed Neutron Source (IPNS) at the Argonne National Laboratory, with an active otltside-user community, an experienced stab, and an adequate experimental hall, is the highest-performance source in operation at present. * Major Facilities for Materials Research and Related Disciplines (National Academy Press, Washington, D.C., 1984). This will be referred to below as the report of the Seitz-Eastman committee. ~ Current Status of Neutron-Scattering Research and Facilities in the United States (National Academy Press, Washington, D.C., 1984).

OCR for page 1
HIGHLIGHTS, OPPORTUNITIES, AND NEEDS 29 We therefore recommend that funds be appropriated to enlarge the LANSCE instrument hall (a $15 million construction project has been proposed) and operation of IPNS be continued until the latter's ongoing activities can be accommodated by a more powerful and cost-effective LANSCE, provided this can be budgeted without jeopardizing the necessary rejuvenation of the high-performance reactors. Very recently Argonne has proposed the upgrade of IPNS by the replacement of the existing accelerator with one of new design (fixed field, alternating gradient) with a sevenfold increase in proton current. If this design is shown to be practical and cost-effective relative to LANSCE, it will be necessary to reconsider our spallation-source priorities in the light of the existing investments. There are no comprehensive plans at present concerning the status of our neutron capabilities for the 1990s. Given the uncertainties in the lifetimes of existing facilities and the time necessary for the design and construction of new facilities, it seems advisable for the neutron- scattering community to initiate discussions immediately leading to such a plan. The feasibility and desirability of both steady-state and pulsed sources should be studied. The possibility of establishing such a facility through international cooperation should also be fully ex- plored. We therefore recommend that supplemental funds be made available to interested qualified institutions to investigate various options for an advanced neutron source. These studies should be done in parallel and in consultation with a panel of outside users charged with devising a plan that will ensure that our neutron-scattering needs will be met in the l990s and beyond. SYNCHROTRON RADIATION SOURCES RECOMMENDATIONS Synchrotron radiation has had a broad impact on studies of both the structural and electronic properties of condensed matter. This is due to its unique high brightness, wide tunability, high polarization, and narrow angular divergence (and, in some instances, time structure). These properties are similar to those of laser sources, but the wave- length range of synchrotron radiation extends from that of the shortest known laser wavelength throughout the ultraviolet, soft-x-ray, and hard-x-ray regions. It is recommended that the current new generation of synchrotron facilities be completed as soon as possible since their high brightness will serve the short-term needs of the next 3 to 5 years.

OCR for page 1
30 HIGHLIGHTS, OPPORTUNITIES, AND NEEDS The main scientific emphasis of these short-term objectives should be in the following areas: (i) Current beam-line instrumentation should be refined in order to achieve higher resolution of photon monochromators in the conventional VUV (0-100 eV) and x-ray (4-15 keV) ranges. This will allow new types of studies to be made of electronic and structural phenomena in conventional solids as well as in low-dimensional sys- tems such as surfaces, polymers, and liquid crystals. (ii) Novel new instrumentation should be developed for soft x rays in the 100-4000 eV range that uses combinations of conventional diffraction-grating tech- nology with new synthetic materials such as multilayer mirrors and other x-ray optical elements. This would allow high-resolution studies of the shallow core-level spectra of all elements. In addition both extended-x-ray absorption fine-structure (EXAFS) and high-resolution near-edge studies could be performed using K or L edges of elements with an atomic number smaller than that of xenon. In order to exploit the potential of insertion devices in the x-ray region it is important that the design allow first harmonic undulator radiation at energies up to ~20 keV. A commitment should begin immediately toward the next generation of high-brightness synchrotron facilities using insertion devices. This should be a two-step approach. New undulator and wiggler devices should be constructed on existing stor- age rings so that insertion-device technology will move ahead rapidly and be ready for possible new rings. New optical devices should be developed to match insertion device sources; this should be done in parallel with the development of new sources, since higher resolution and wider tunability cannot be achieved simply by attachment of existing beam lines to new sources. As a second priority, planning should begin immediately leading to proposals for a next-generation, possibly all-insertion device machine. Ideally, this machine should be completed in the early l990s, since projected user demand will saturate then-existing facilities by that time. The design parameters, such as electron energy and physical size, should be determined by scientific considerations, but the three areas of spectroscopy, scattering, and micros- copy should be accommodated. The 6-GeV machine recommended by the Seitz-Eastman committee appears to meet these needs. The overall costs of such a next-generation synchrotron source are in the range of $160 million, and construction could take place over a period of 6-7 years. Firm decisions on when to build such a machine should be made on the basis of new scientific opportunities, user demand, and ongoing experience with the undulator and wiggler facilities discussed above.

OCR for page 1
HIGHLIGHTS, OPPORTUNITIES, AND NEEDS 31 HIGH-MAGNETIC-FIELD FACILITIES RECOMMENDATIONS Laboratories for the production of high magnetic fields (>15 T. where 1 T_ 104 Oe) and their utilization in condensed-matter research exist in France, Holland, Belgium, Japan, Poland, the Soviet Union, and the United States. The National Magnet Laboratory at the Massachusetts Institute of Technology is the only major user facility for high-field research in the United States. A wide variety of steady- field magnets exist there and are categorized by their peak fields, bore sizes, and homogeneity. The largest field currently available there is 29 T. in a 3.3-cm-bore hybrid configuration. Magnetic fields above 30 T are economically feasible only in pulsed operation. Nondestructive, repetitive pulsed fields in the range 40 T c H c 60 T are now available in Holland, Japan, and the Soviet Union. A 75-T configuration will soon be operating in Osaka, Japan. A high-magnetic-field facility has just been completed at the Institute for Solid State Physics (ISSP) in Tokyo, Japan, at a cost of about $10 million. It can produce a variety of nondestructive pulsed fields (c50 T); it can produce fields of 50-100 T by plasma compression that may be nondestructive; and it can produce a 100-500 T implosion-generated field that is totally destructive of the sample. The ISSP group has been generating fields of 100 T for several years, which have been used in studies of cyclotron resonance and various other phenomena in semi- conductors. No comparable facilities are available in the United States, although much of the seminal technology was developed in this country. The availability of high magnetic fields has yielded such experimen- tal results as the discovery of the fractionally quantized Hall effect. More generally, high-field magnets expand the phase diagram of a solid by adding a new variable, the magnetic field, to the usual variables, pressure and temperature, thereby increasing our knowledge of prop- erties of solids under extreme conditions. For these reasons, and the paucity of high-field magnets in the United States, We recommend that new money should be made available to enable greater emphasis to be placed on the generation of pulsed high magnetic fields at the National Magnet Laboratory and/or at a new site elsewhere in the United States. The cost of duplicating the high-magnetic-field facility in Osaka is estimated to be $1 million to $2 million. ELECTRON-MICROSCOPE FACILITIES RECOMMENDATIONS The country's electron-microscope facilities provide a reservoir of talent and expertise necessary to generate the innovative instrumenta

OCR for page 1
32 HIGHLIGHTS, OPPORTUNITIES, AND NEEDS lion crucial for promoting the growth of the power and subtlety of electron-microscopic investigations in the coming decade. There ap- pear to be four major areas in which advanced instrumental initiatives could have a major impact on the development of the field during this period: (1) development of ultrahigh-vacuum sample environments for surface studies; (2) development of efficient instrumental accessories for microanalytical techniques such as electron energy loss spectros- copy; (3) development of low-temperature specimen stages and spec- imen preparation techniques necessary for systematically attacking questions about the structures of large biological molecules and of many others that are of interest to condensed-matter physics; and (4) development of computerized data collection and analysis. It is esti- mated that the cost of the major capital equipment required for im- plementing these instrumental initiatives would average $1 million for each, spread out over a period of 2 years, for a total of $4 million. The increase in the operating budgets of the institutions participating in these initiatives is estimated to be $4 million, to be achieved over a period of 3-4 years. Our recommendation in this area is as follows: Advanced instrumentation initiatives in the four areas of electron microscopy cited above should be established in response to competitive proposals from interested institutions. If necessary, the federal funding agencies should stimulate the submission of such proposals. GENERAL RECOMMENDATIONS CONCERNING NATIONAL FACILITIES There are two broad categories of users of national facilities. Committed users are those whose research programs are built nearly exclusively around the use of these facilities and include the scientific staff of the facilities. By contrast, occasional users have research programs based on other techniques, usually at their home laborato- ries, but whose research is increased in scope by the power of these other specialized techniques. The long-term vitality and future growth of national facilities depend crucially on a broad base of these occasional users who have neither the time nor the financial resources to become expert in these techniques but who furnish nonetheless a wealth of novel materials and ideas for experiments. In order to aid the integration of these occasional users into the activities of the facilities, We recommend that special funding be set aside for the purpose of accommo- dating occasional users at the national facilities. This money would help finance travel and living expenses, particularly for university users, and

OCR for page 1
HIGHLIGHTS, OPPORTUNITIES, AND NEEDS 33 provide an increase in the in-house support staff. This program should be formulated by the individual facilities in consultation with university and industrial collaborators and funded on the basis of separate proposals from these facilities. We estimate that a significant trial program would require $4 million to $6 million per year over a 3-4 year period. Recently established national facilities (e.g., those dedicated to research employing synchrotron radiation or high-resolution electron microscopes) have been developed as user facilities or as DOE Centers for Collaborative Research. The independent peer review of the experiments approved to be done improves the quality and nature of the research at these facilities. At the same time, the ability to respond to rapidly emerging scientific opportunities and the timely development of new experimental techniques requires that a certain fraction (per- haps 30 percent) of the available time be allocated at the discretion of the in-house staff. Therefore, We recommend that in the future it is desirable that national facilities should operate in the user mode in which the majority of experimental time is allocated by independent peer review. There are at least two modes in which this peer review may operate: review of experiment-by-experiment proposals by occasional users and peer review of proposals for participating research teams (PRTs) that undertake to construct, maintain, and carry out research programs using instruments on a shared basis with non-PRT members. Finally, it is our strongly held view that the needs of the individual researcher, which have been outlined above in the section on Support for Individual Researchers, are so great at this time that the highest priority for the use of new monies for the support of condensed-matter physics is in meeting those needs and for the upgrading of the existing national facilities that is necessary for the achievement of their full potential. When this has been accomplished, the construction of the new national facilities should begin. University-Industry-Government Relations One of the primary strengths of condensed-matter physics is that forefront research of the highest quality is carried out at industrial laboratories as well as at universities and government laboratories. This is due to the fact that condensed-matter physics is closest to applications in technology of all the subfields of physics. It argues for a strong coupling between universities and national laboratories, where

OCR for page 1
34 HIGHLIGHTS, OPPORTUNI TIES, AND NEEDS most of the basic research in condensed-matter physics is done? and industry, where the results of that research, as well as of the research done in-house, is transferred into technology. industry also benefits greatly from the pool of condensed-matter physicists produced each year by this country's universities and from those trained in postdoc- toral programs at national laboratories. For their part, universities have received support from industry in the form of grants of equipment, funding for research projects, and support for graduate students. How- ever, if the strongest possible coupling between universities, govern- ment laboratories, and industry is to be achieved, the support of university and laboratory research by industry should go well beyond the mere provision offunds and equipment: research cooperation is also required. At the same time, continuing efforts should be made to in- crease the research cooperation between the national laboratories and university scientists, since special facilities exist at the national laboratories that are not available elsewhere. The realization of such cooperation will require the coordinated efforts of universities and industry, and of the federal government as well. The following recom- mendations outline our views of the roles of each of these partners in this process. 1. What government should do: Establish policies, including tax incentives, to stimulate fundamental research in industry. Provide support for students engaged in cooperative university- industry research. Encourage and facilitate the flow of scientists between federal labora- tories and universities for cooperative research programs. Maximize access by outside users to the special facilities available only at the federal laboratories. 2. What industry should do: Increase the amount of in-house research even beyond the levels directly supported by the policies suggested in point 1 above, i.e., through the use of corporate funds. Establish and fund programs that enable industrial scientists to take sabbatical leaves in universities and at national laboratories. Receive university faculty and laboratory researchers in industrial laboratories for sabbatical leaves and summers. Provide direct support of faculty and departmental research grants (e.g., the IBM programs). Provide direct support of graduate and postdoctoral fellowships (e.g., the IBM fellowship program).

OCR for page 1
HIGHLIGHTS. OPPORTUNITIES. A ND NEEDS 35 Formulate cooperative research projects with graduate students (e.g., the MIT-AT&T Bell Laboratories program). Provide instrumentation for special facilities at national laboratories. 3. What universities should do: Implement cooperative research and support programs with indus- try, as MIT has done in materials processing. Adopt a limited form of the "Japanese model" in which applied physics research in high-technology areas, such as semiconducting lasers, photonics, and electronics is supported by industrial firms directly involved in the manufacture of materials, devices, compo- nents, and systems employing these technologies. Cooperate in the graduate training of industrial employees engaged in applied research. Arrange for sabbatical leave for federal laboratory researchers in university departments. This support can take the form of direct research contracts; the gift or loan of equipment, devices, and components; and the support of graduate students.

OCR for page 1