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 31
Page 31 1 Electronic, Optical, and Magnetic Materials and Phenomena: The Science of Modern Technology Important and unexpected discoveries have been made in all areas of condensed-matter and materials physics in the decade since the Brinkman report.1 Although these scientific discoveries are impressive, perhaps equally impressive are technological advances during the same decade, advances made possible by our ever-increasing understanding of the basic physics of materials along with our increasing ability to tailor cost-effectively the composition and structure of materials. Today's technological revolution would be impossible without the continuing increase in our scientific understanding of materials, phenomena, and the processing and synthesis required for high-volume, low-cost manufacturing. The technological impact of such advances is perhaps best illustrated in the areas of condensed-matter and materials physics discussed in this chapter, which will examine selected examples of electronic, magnetic, and optical materials and phenomena that are key to the convergence of computing, communication, and consumer electronics. Technology based on electronic, optical, and magnetic materials is driving the information age through revolutions in computing and communications. With the miniaturization made possible by the invention of the transistor and the integrated circuit, enormous computing and communication capabilities are becoming readily available worldwide. These technological capabilities, which enabled the information age, are fundamentally changing how we live, interact, and transact business. Semiconductors provide an excellent demonstration of the strong 1National Research Council [W.F. Brinkman, study chair], Physics Through the 1990s, National Academy Press, Washington, D.C. (1986).
OCR for page 32
Page 32 interplay between and interdependence of science and technology. Perhaps in no other area are advances in technology more closely linked to advances in understanding. This chapter is not intended to be comprehensive; rather, it seeks to illustrate the pivotal role of condensed-matter and materials research in providing the understanding required to develop enabling technologies. At the same time, the development of these new technologies has greatly expanded the tools and capabilities available to scientists and engineers in all areas of research and development, ranging from basic research in physics and materials to other areas of physics and to such diverse fields as medicine and biotechnology. The examples discussed also make evident the importance of long-term, sustained research in realizing the benefits to society of improved scientific understanding of materials (see Figure 1.1). Figure 1.1 Incorporation of major scientific and technological advances into new products can take decades and often follows unpredictable paths. Illustrated here are some selected technologies supported by the foundations of electronic, photonic, and magnetic phenomena and materials. These technologies have enabled breakthroughs in virtually every sector of the economy. The two-way interplay between foundations and technology is a major driving force in this field. The most recent fundamental advances and technological discoveries have yet to realize their potential.
OCR for page 33
Page 33 Although technological advances today are most often associated with the information age or communications and the computing revolution, impressive advances continue to be made across a broad spectrum of technologies and scientific disciplines (see Box 1.1). For example, progress in condensed-matter and materials physics has led to advances in biology, medicine, and biotechnology. New tissue diagnostics based on diffusing light probes use understanding borrowed directly from the physics of carrier transport in mesoscopic random materials. The development of new optical microscopies, such as two-photon confocal, optical coherence, and near-field optical microscopy, together with the widespread use of optical tweezers, have started a revolution in the observation and manipulation of submicrometer-sized objects in cell biology, in new forms of spectroscopic endoscopy, and in gene sequencing techniques. The emergence of high-power solid-state lasers and solid-state detectors and the widespread use of fiber optics make new optical approaches for diagnostics, dentistry, and surgery increasingly easy. A new form of magnetic resonance imaging enabled by semiconductor laser pumping of spin-polarized xenon gas has allowed the three-dimensional mapping of lung function. The generation of femtosecond pulses of light by the use of new solid-state lasers has begun another revolution in our understanding of the subpicosecond dynamics of biological molecules on the important frontiers of molecular signal processing and protein folding. Although not covered in detail here, such advances in the use of optics in medicine and biology are discussed in detail in another National Research Council report.2 In addition, semiconductor and other solid-state lasers or enhanced solid-state detector arrays, offshoots from condensed-matter physics, are enabling major advances in the fields of atomic and molecular physics, physical chemistry, high-energy physics, and astrophysics. New optical materials and phenomena are also responsible for a number of advances in the technologies associated with printing, copying, video and data display, and lighting. In the realm of magnetic materials, the loss of cobalt in the 1980s because of political unrest in Zaire prompted an intense research effort to find cobalt-free bulk magnetic materials. This led to major advances in creating magnetic structures from neodymium and iron, which had superior properties and lower cost compared with cobalt alloys for electric motors and similar applications requiring magnets with high permanent magnetization. These new magnets, which are achieved through complex alloys and even more complex processing sequences, are vastly expanding the industrial use of bulk magnetic materials. Advances in magnetic materials and their applications are not limited to bulk materials with high permanent magnetization and magnetic materials used in information storage. Improvements in soft bulk magnetic materials play an important role in transformers used in the electric power distribution industry. In- 2National Research Council [C.V. Shank, study chair], Harnessing Light: Optical Science and Engineering for the 21st Century, National Academy Press, Washington, D.C. (1998).
OCR for page 34
Page 34 BOX 1.1 The Science of Information-Age Technology The predominant semiconductor technology today is the silicon-based integrated circuit. The silicon integrated circuit is the engine that drives the information revolution. For the past 30 years, this technology has been dominated by Moore's Law: that the density of transistors on a silicon integrated circuit doubles about every 18 months.* Moore's Law articulates the increased functionality-per-unit cost that is the origin of the information revolution. Today's computing and communications capability would not be possible without the phenomenal 25 to 30 percent per year exponential growth in capability per unit cost since the introduction of the integrated circuit in about 1960. That sustained rate of progress has resulted in high-density memories with 64 million bits on a chip and complex, high-performance logic chips with more than 9 million transistors on a chip. This trend is projected to continue for the next several years (see Figures 1.1.1 and 1.1.2). If the silicon integrated circuit is the engine that powers the computing and communications revolution, optical fibers are the highways for the information age. Although fiber optics is a relatively recent entrant into the high-technology arena, its impact is enormous and growing. Fiber is now the preferred technology for transmission of information over long distances. There are already approximately 30 million km of fiber installed in the United States and an estimated 100 million km worldwide. In part because of the faster than exponential growth of connections to the Internet, optical fiber is being installed worldwide at an accelerated rate of Figure 1.1.1 Computing power versus time in microprocessors. (Courtesy of Intel.) *Moore's Law, first articulated by Gordon Moore of Intel Corporation, is not a statement of physics. It is a statement that the industry will perform the R&D necessary and supply capital investment at the rate required to achieve this doubling rate. (Box continued on next page)
OCR for page 35
Page 35 (Box continued from previous page) Figure 1.1.2 History of semiconductor technology. more than 20 million km per yearmore than 2,000 km per hour, or around Mach 2. In addition, the rate of information transmission down a single fiber is increasing exponentially by a factor of 100 every decade. Transmission at 2.5 terabits per second has been demonstrated in the research laboratory, and the time lag between laboratory demonstration and commercial system deployment is about 5 years. The analog of Moore's Law for fiber transmission capacity, which serves as a technology roadmap for lightwave systems, is shown in Figure 1.1.3. Figure 1.1.4 summarizes the history of optical communications technology. Compound semiconductor diode lasers provide the laser photons that transport information along the optical information highways. Semiconductor diode lasers are also at the heart of optical storage and compact disc technology. In addition to their use in very high-performance microelectronics applications, compound semiconductors have proven to be an extremely fertile field for advancing our understanding of fundamental physical phenomena. Exploiting decades of basic research, we are now beginning to be able to understand and control all aspects of compound semiconductor structures, from mechanical through electronic to optical, and to grow devices and structures with atomic layer control, in a few specific materials systems. This capability allows the manufacture of high-performance, high-reliability, compound semiconductor diode lasers that can be modulated at gigahertz frequencies to send information over the fiber-optic networks. High-speed semiconductor-based detectors receive and decode this information. These same materials provide the billions of light-emitting diodes sold annually for displays, free-space or short-range high-speed communication, and numerous other applications. In addition, very high-speed, low-power compound semiconductor electronics play a major role in wireless communication, especially for portable units and satellite systems. Another key enabler of the information revolution is low-cost, low-power, high-density information storage that keeps pace with the exponential growth of corn- (Box continued on next page)
OCR for page 36
Page 36 BOX 1.1 Continued Figure 1.1.3 Exponential growth in data transmission rate in fibers. (Courtesy of Bell Laboratories, Lucent Technologies.) puting and communication capability. Both magnetic and optical storage are in wide use. Recently, the highest performance magnetic storage/readout devices have begun to rely on giant magnetoresistance (GMR), a phenomenon that was discovered by building on more than a century of research in magnetic materials. Although Lord Kelvin discovered magnetoresistance in 1856, it was not until the early 1990s that commercial products using this technology were introduced (see Figure 1.1.5). In the past decade, our understanding of condensed matter and materials converged with advances in our ability to deposit materials with atomic-level control to produce the GMR heads that were introduced in workstations in late 1997. It is hoped that with additional research and development, spin valve and colossal magnetoresistance (CMR) technology may be understood and applied to workstations of the future. This increased understanding, provided in part (Box continued on next page)
OCR for page 37
Page 37 (Box continued from previous page) Figure 1.1.4 History of optical communications technology. Figure 1.1.5 History of magnetoresistance. (Courtesy of IBM Research.) (Box continued on next page)
OCR for page 38
Page 38 BOX 1.1 Continued by our increased computational ability arising from the increasing power of silicon integrated circuits, coupled with atomic-level control of materials, led to exponential growth in the storage density of magnetic materials analogous to Moore's Law for transistor density in silicon integrated circuits (see Figure 1.1.6). Figure 1.1.6 ''Moore's Law'' for magnetic storage: logarithm of storage density versus time. (Courtesy of IBM Research.) creases in the magnetic permeability and decreases in the losses by incremental improvements in amorphous metglass leads to decreases in the losses suffered in transmission and distribution. Magnetoelectronics, an emerging area based on advances in the understanding of the properties and processing of magnetic materials, shows promise for future applications. Despite the numerous recent discoveries and technological advances in the understanding and use of magnetic materials, our fundamental understanding of magnetism remains remarkably incomplete. Some of the basic questions and important challenges in magnetism facing the scientific community are discussed in this chapter. Electronic Materials And Phenomena Materials and Physics That Drive Today's Technology Silicon Technology As noted in the introduction, semiconductor technology is the key enabler of the information age. The science of materials as a specific discipline is a relatively
OCR for page 39
Page 39 modern development. The physics and materials science of semiconductors is an even more recent development. Metals and ceramics were commercially important materials when the transistor was demonstrated about 50 years ago. Despite the fact that the science of semiconductors is relatively new compared to that of metals and ceramics, the commercial importance of semiconductors is now comparable to that of metals and ceramics. Advances in semiconductor technology are driving the rapid growth of business sectors involved with computing, communications, consumer electronics, and software, and are enabling emerging fields such as biotechnology. Today's transistor performance and the incredible advances of integrated circuits in silicon technology are the result of more than 50 years of dedicated research in electronic materials. The understanding achieved from this focused research has enabled high-volume manufacturing of circuits with ever-increasing complexity and performance. In addition to driving computing and communications, the steady decrease in cost-per-function has created literally hundreds of applications for silicon integrated circuits. Semiconductors are ubiquitous. Microprocessors are used almost everywhere todayfrom household appliances, to banking and smart cards, to automotive and aircraft control systems, automatic fuel injection engines, and cockpit instrumentationand will be in even more applications tomorrow. The same sustained rate of progress that permitted the widespread application of semiconductors has created a global semiconductor industry with 1997 revenues of about $150 billion, supported by a materials and equipment infrastructure of about $60 billion. Semiconductor technology is also the heart of the $1 trillion global electronics industry and vital in many other areas of the approximately $33 trillion global economy. The increasing functionality of integrated circuits, which comes as a by-product of scaling to smaller feature sizes, has been achieved by comparable increases in their complexity and that of the attendant manufacturing process. Today's leading-edge microprocessors are manufactured with minimum feature sizes of 250 nm and require six levels of metallization to connect the transistors and circuit components. A beneficial by-product of the steady decrease in feature size is higher speed devices and circuits. Based on technology projections that form the basis of the National Technology Roadmap for Semiconductors,3 the semiconductor industry expects to manufacture integrated circuits with feature sizes of 180 nm in 1999 and 150 nm by 2001. If the scaling trend continues as indicated by Moore's Law, which the industry has followed since its inception, integrated circuits with minimum feature sizes of 50 nm will be manufactured in high volumes within 15 years (see Box 1.2). Continuing to advance this technology requires that the industry invest in expensive new manufacturing facilities and an ever-increasing scientific understanding and control of semiconductor 3Semiconductor Industry Association, National Technology Roadmap for Semiconductors, SEMATECH, Austin, Tex. (1997).
OCR for page 40
Page 40 BOX 1.2 A Brief History of Ion Implantation In the early 1940s, the basic machines that were later adapted for ion implantation in the semiconductor business were used at Oak Ridge National Laboratory for uranium isotope separation. This was a critical part of the Manhattan Project. Ion beams were first used as part of semiconductor-device processing at Bell Laboratories in 1952. Bell filed a comprehensive patent in 1954 covering the use of ion implantation for doping semiconductors, but it was not until 1966 that implantation was actually used to manufacture commercial semiconductor devices. Hughes Research Laboratory used the technique to form junctions in the manufacturing of diodes. In 1970 Texas Instruments began using ion implantation in integrated circuit manufacturing to set threshold voltages. Concurrent with these developments in processing, several companies attempted to enter the implant-tool manufacturing business with only moderate success, most successful among them being Accelerators Incorporated. In 1971, however, a new company, Extrion, was formed to build commercial implanters specifically designed for integrated circuit manufacturing. Extrion soon became the primary supplier of implant tools. This led to the development of a whole new industry in America. Today, ion implantation is used in several steps of the integrated circuit manufacturing process to control the concentration and depth distribution of dopants. Ion implantation tool manufacturing, an almost exclusively U.S. industry, has grown to a more than $1 billion per year business. Three U.S. companies (Applied Materials, Eaton, and Varian) supply virtually all the commercial ion-implant systems worldwide. materials and manufacturing processes. Conversely, our rate of understanding has been greatly enhanced by the technology created by the rapid advances in semiconductor-related technologies. Many daunting scientific and engineering problems must be overcome in order to continue at the Moore's Law rate of progress for the next 15 or even 10 years. For instance, the number of wires needed to connect the transistors grows as a power of the number transistors. As transistor dimensions are shrunk, computer chip manufacturers pack an ever-increasing number of them into their devices. The complexity of wiring the transistors in these devices may eventually reach the limits of known materials. Moreover, the cost of manufacturing increasingly layered and complex wiring structures may limit the performance of these systems. Even if solutions to the interconnect problem can be identified, continued scaling of silicon technology will ultimately encounter fundamental limits. For example, metal-oxide semiconductor transistors can be built today with gate lengths of 30 nm (only about 150 atoms long) that display high-quality device characteristics. Manufacturing complex circuits that rely on devices with these feature sizes will require several hundred processing steps with atomic-level control. However, the performance of complex integrated circuits with tens of millions of transistors may be degraded because of nonuniform operating
OCR for page 41
Page 41 characteristics. In time, continued decreases in device dimensions may result in the information being carried by an ever-decreasing number of charge carriers; ultimately, simple statistical fluctuations will limit the uniformity of device characteristics as the number of charges used to convey information decreases. To delay this limitation as long as possible, research is under way on new materials with a high dielectric constant for memory applications or to limit leakage from tunneling currents. As understanding of synthesis and processing increase, ferroelectric materials are being introduced for nonvolatile memory applications. Even with these advances, as feature sizes continue to decrease, integrated circuits based on field-effect transistors will eventually encounter fundamental limits, such as interconnect delays caused by the ever-increasing number of interconnects, heat generation, or quantum limits of transistors that are too small to confine the electrons in the channels. Today's approach to the design and manufacture of integrated circuits will no longer be extendible to smaller feature sizes and higher densities. The fundamental limits of the current technology and our addiction to exponentially increasing computational power offer exciting scientific and engineering challenges in the search for the materials and device structures of the future. Compound Semiconductor Technology Compound semiconductors, which consist of more than one element, offer intrinsically higher speed and lower noise compared with silicon. These advantages have been exploited to develop very high frequency electronic devices and circuits for microwave and wireless communication applications. The worldwide electronics market for compound semiconductors is estimated to be growing at about 40 percent per year and is expected to be about $1 billion in 2000. In addition to the high-speed microwave applications for which they have long been the materials of choice, discrete components are widely used in the low-noise receivers of telephone handsets. Compound semiconductors such as GaAs, AlGaAs, InGaAs, SiGe, GaN, and GaAlN are key to the development of next-generation wireless telephones, which will use higher frequency microwaves to transmit more information and allow more channels. GaN transistors, for example, have a high breakdown voltage and great robustness, although extensive research and development is required before the material can be understood and fabricated in a well-controlled fashion. Advancing the limits of semiconductor materials technology is essential for increasing the speed of transistors and advancing our ability to modulate light-emitting diodes and semiconductor lasers for high-speed optical information transmission. Because compound semiconductors are composed of more than one element, they offer a vastly increased range of materials from which to create structures with desired electronic properties. The technology of modern compound semiconductor device fabrication is predicated on the ability to produce extended planar layers of uniform composi-
OCR for page 82
Page 82 Figure 1.11 Micrograph of prototype 256-kb nonvolatile magnetic random access memory chip. (Courtesy of Honeywell.) The Physics of Magnetism It is evident from the above discussion that there is a strong technology pull for research in magnetism. The last 10 years have seen numerous exciting discoveries, a number of which have already had a direct impact on technology. Others are providing us with a better understanding of the world around us and/or are helping lay the groundwork for future technology. It is amazing that, although much is known about magnetism and while effects drive a $100 billion per year industry, our basic understanding of magnetism, even in a material such as iron, is incomplete. As an example, Figure 1.12 shows the results of a simula-
OCR for page 83
Page 83 tion of a two-dimensional array of magnetic moments approximating a 10 nm × 500 nm × 1000 nm sheet of permalloy. One might anticipate that for a system of this size, the moments would all work in unison, switching together under the influence of an applied field; however, as shown in Figure 1.12, such is not the case. The behavior of this rather elementary system exhibits a great deal of structure and complexity. There is neither experimental nor theoretical consensus on the detailed behavior of such a system. Nor is it understood how small such a system must be to ensure true, single-domain behavior in which all of the moments always remain parallel to each other. The fundamental limit on stability of magnetic domains is an important area of basic investigation in magnetism. The advancing march of magnetic technology makes investigation of these limits inevitable, but probing these limits raises some of the most challenging questions for condensed-matter physics and materials science, such as, What is the smallest size magnetic element stable against external perturbations such as temperature fluctuations? and, Given that quantum mechanics sets bounds on the lifetime of any magnetic state, how do such bounds ultimately establish limits on the size of the smallest possible magnetic entities useful for technological applications? Molecular magnets such as the ferric wheel shown in Box 1.13 are the subject of a wide variety of physical analyses aimed at shedding light on these and other challenging questions about magnetism in small structures. Traditional Figure 1.12 Micromagnetic simulation of the switching behavior of a 10 nm × 500 nm × 1000 nm permalloy dot. The arrows indicate the direction of the magnetization (m) at the point in the hysteresis cycle indicated on the curve in the lower left. (Courtesy of IBM Research.)
OCR for page 84
Page 84 BOX 1.13 Nanomagnets Fundamental questions about nanomagnets could not be addressed were it not for the remarkable advances made by materials scientists and chemists in the synthesis of exquisitely controlled nanometer-scale magnetic structures. Many "traditional" synthesis techniques have been applied successfully in this area: sub-100 nm scale permalloy particles can be made with lithographic techniques, and growth assisted by scanning-tunneling microscopes has proven that it is possible to fabricate pure iron particles with a variety of sizes and aspect ratios. The most exciting new techniques involve various "self-assembly" strategies that are emerging as extensions of chemical synthesis techniques. The relatively low-tech techniques of colloid growth have been adapted to make cobalt particles 8 nm in diameter with close to atomic control. Actual atom-by-atom control in the magnetic domain is now achieved by metallo-organic molecular synthesis; magnetic molecules containing exactly 10 iron atoms (the "ferric wheel,'' see Figure 1.13.1), or exactly 12 manganese ions, for example, are now routinely available. Figure 1.13.1 Structure of the Fe10 "ferric wheel" cluster, where the large solid circles represent the iron atoms and the empty circles are, in order of decreasing size, chlorine, oxygen, and carbon. The 10 Fe3+ ions, each with a magnetic moment corresponding to the same angular momentum or spin, are bound together into a perfectly regular ring. High magnetic field experiments have shown that the Fe3+ ions exhibit antiferromagnetic behavior; neighboring spins prefer to be antiparallel. The spin structure of the molecule passes through a rich sequence of phase transitions resembling those in bulk layered antiferromagnets. These experiments open the prospect of precisely controlling the structure, interactions, and dynamics of nanomagnets. [Reprinted with permission from D. Gatteschi, A. Caneschi, L. Pardi, and R. Sessoli, "Large clusters of metal ions: The transition from molecular to bulk magnets," Science 265, 1056 (1994). Copyright © 1994 American Association for the Advancement of Science.]
OCR for page 85
Page 85 characterization gives the basic magnetic parameters of the particle: ionic moment (i.e., the local state of spin), exchange interaction (the coupling strength between the spins), and anisotropy (the height of the energy barrier in the double-well potential separating the "up" state from the "down" state). Such characterizations demonstrate that the ferric wheel behaves very much as an atomic-scale analog of a layered antiferromagnet. Less traditional characterizations are required to understand the ultimate stability of such molecular magnets, which is determined by quantum tunneling. One particularly interesting example is the observation of what might be called "quantized hysteresis" in the Mn(12) molecule: at low temperatures, this structure shows a propensity to switch from up to down at a sequence of regularly spaced magnetic fields. Some evidence suggests that these magnetic fields coincide with resonances between quantum levels of the up and down wells, resulting in enhanced tunneling. Although some details of the tunneling mechanism remain to be understood, this is a particularly simple example of the stability (up versus down) of the moment of a molecular magnet being ultimately limited by a purely quantum effect. Of more profound significance than the observation of quantum tunneling would be the observation of "quantum coherence" in these nanomagnets. The phenomenon is closely analogous to the microscopic quantum coherence (MQC) effect sought in small SQUIDs for yearsthe creation of a quantum state in a controlled superposition of up and down states. The regular advance of the phase of this superposition would result in the sinusoidal oscillation of the magnetic domain between the up and down states. Observation of such coherence oscillations would foreshadow a significant change in the role that quantum mechanics might play in the dynamics of magnetic domains. Although we might view magnetic tunneling as a nuisance, destroying the stability of a bit stored in the magnetic domain, coherence, if controllable, could be the resource needed to realize the basic element of storage and processing in quantum computing. Signs of magnetic quantum coherence have in fact been observed in a naturally occurring magnetic nanoparticle. In the past 10 years rapid progress has been made in the characterization and understanding of magnetic multilayers, exchange coupling, and spin-dependent transport through magnetic materials and interfaces. Results from an experiment representative of this exciting, ground-breaking work is shown in Figure 1.13. This experiment measured the oscillatory exchange coupling between iron layers separated by a chromium spacer of varying thickness. The chromium wedge was grown epitaxially on the nearly perfect surface of an iron whisker crystal whose magnetization is split into two opposite domains along the  direction. A thin iron film was deposited on top of the chromium, and its magnetization was measured using scanning electron microscopy with polarization analysis (SEMPA). The SEMPA image, drawn on the wedge schematic, clearly shows that the exchange coupling reverses direction with almost every single monolayer change in chromium thickness. The oscillatory coupling period, which arises
OCR for page 86
Page 86 from nesting features in the Fermi surface of chromium, is actually slightly incommensurate with the chromium lattice, producing the phase slips observed at chromium layers 24 and 44. The applications focus provided by GMR has helped to stimulate and invigorate the search for new magnetic heterostructures and nanostructures and new magnetoresistive materials. GMR materials consist mainly of nanometer thicknesses of interleaved metallic layers that are alternately ferromagnetic and non-ferromagnetic. (Analogous nano-dispersed two-phase composites can also exhibit GMR properties.) Key to strong in-plane GMR are spin-dependent scattering at layer interfaces and an electron mean free path (approximately 10 nm) greater than sublayer thickness. The relative resistance change can be greater for current flow perpendicular to the layer planes. Resistance changes as large as 100 percent relative to the low-resistance state have recently been reported. Work on magnetic multilayers is also stimulating new thinking concerning novel devices that can be made by integrating magnetic materials with standard semiconductor technology. An example of this is shown in Figure 1.14 where a GMR (Co/Cu) multilayer serves as the base of an n-silicon metal base transistor. Biased in the common base configuration, this device exhibited a 215 percent change in collector current in a magnetic field of 500 G at 77 K with typical GMR characteristics. The in-plane GMR of the multilayer was only 3 percent. Although by no means a practical transistor, this structure allows the study of spin-dependent scattering of hot electrons in magnetic multilayers. More practical spin transistors may be forthcoming, particularly if ways to achieve 100 percent spin-polarized injection can be devised. Figure 1.13 Oscillatory exchange coupling in Fe/Cr/Fe. [Physical Review Letters 67, 140 (1991).]
OCR for page 87
Page 87 Spin-polarized tunneling experiments in magnetic thin-film planar junctions are helping to elucidate novel magnetic properties as well as demonstrate features of considerable device potential. The first successes in spin-polarized tunneling between two ferromagnets through an insulating tunnel barrier occurred only very recently, even though spin-polarized tunneling was predicted more than 20 years ago. A simplified structure of this type of junction is shown in Figure 1.15. The magnetization of the ferromagnetic base electrode is pinned in the direction indicated. The magnetization of the ferromagnetic counter electrode is shown as aligned with that of the base electrode but can be reversed by application of a modest magnetic field. A proper magnetic design yields a hysteretic response curve symmetric about H = 0. Resistance changes of greater than 30 percent have been demonstrated. Such junctions are of considerable interest as potential storage elements in one approach to magnetic RAMs. Slightly different configurations give devices with nonhysteretic characteristics but with similar magnetoresistance. These devices are attractive candidates for sensor applications and may provide a follow-on to GMR sensors for hard-disk drives. The enormous surge in the synthesis and study of high-Tc perovskite materials spawned a concerted effort to explore the magnetic properties of similar materials. Some of these materials exhibit what has been termed colossal magnetoresistance (CMR). The magnetoresistance of doped manganite structures such as La1-xSrxMnO3 changes by a factor of 2 or 3, although not at temperatures and magnetic fields suitable for practical device applications. These systems share much in common with high-Tc cuprate superconductors, from which dozens of new crystal structures have emerged. Replacing copper with manganese, for example, could generate a platform of new crystal chemical systems, some of which will undoubtedly exhibit promising CMR properties. Figure 1.14 Schematic cross section of a prototype spin-valve transistor. [Physical Review Letters 74, 5260 (1995).]
OCR for page 88
Page 88 Figure 1.15 Magnetic tunnel junction structure. (Courtesy of IBM Research.) Steady advances in improving the characteristics for technological applications has been realized in the more traditional bulk materials. For example, in the case of high-permeability soft materials, we have learned how to compensate for the deleterious magnetic anisotropy from one magnetic element by introducing a second element having an anisotropy of the opposite sign. Another way to cancel anisotropic effects, and thereby increase permeability, is to rapidly quench a magnetic ribbon, as with magnetic metglass, thus making its structure amorphous. The preferred axes of the atomic moments are now random and therefore the atomic contributions to anisotropy energy tend to cancel, making it easy to remagnetize. In the case of permanent magnets, the magnetic anisotropy, and therefore the coercivity of a ferromagnet, decrease steeply as its temperature approaches the Curie critical point, where these parameters necessarily vanish. We have learned how to increase the Curie temperature, and thereby the coercivity, of permanent magnet materials of the RexFey type by introducing interstitial N or C or, to lesser degrees, Ti, V, W, Mo, or Si. The complexity as well as the importance of processing details in the synthesis of high-performance magnetic materials is well demonstrated in the case of NdFeB discussed previously. Measurement techniques are vital in the research of magnetic materials and phenomena. Experimental advances that have contributed to breakthroughs in the last decade include scanning-tunneling microscopy (STM), magnetic force microscopy (MFM), magneto-optic Kerr imaging, and scanning electron microscopy with polarization analysis (SEMPA). STM has been critical to understanding how subtle differences in physical structure can make profound differences in magnetic structure or properties. Characterization techniques based at major facilities are equally important. For CMR, as for high-Tc research, neutron dif-
OCR for page 89
Page 89 fraction is needed to obtain the position of the oxygen atoms within the unit cell as a function of temperature, field, pressure, and doping. Electron microscopy is needed to understand growth inclusions that form two-dimensional stacking faults. Synchrotron sources enable advanced spectroscopies to identify the +3 and +4 valence states of Mn and their ratio. Diffuse x-ray scattering and quasi-elastic neutron scattering are used to investigate the presence and dynamics of polaronic distortions. A number of other techniques based on such effects as spin-polarized photoemission, magnetic circular dichroism, and second harmonic generation are becoming increasingly prevalent, while many others are in the initial stages of demonstration. Major Outstanding Materials and Physics Questions and Issues in Magnetism Many outstanding scientific questions remain in the field of magnetism. Answers to a number of these questions will have an important technological impact and are necessary to continue the momentum and growth of the magnetics industry. With a few notable exceptions, we lack detailed understanding of the magnetic properties of nanostructured magnetic elements and arrays of such elements. Examples include the following: • The nature of domain structure and its influence on switching behavior; • The dynamics and switching times in such elements or systems; • The influence of temperature, both in the context of stability against thermally induced switching and in the context of structural change at elevated temperature; • The nature of the interaction of spin-polarized currents with such elements, both reversible and irreversible; and • The role and the impact of quantum coherence and macroscopic quantum tunneling in the smallest of such structures consisting of a cluster of atomic spins. We need to understand the impact of the issues above and related issues regarding technologies such as magnetic recording and the synthesis of new materials with improved properties such as higher BH-products. Because of the resurgence of the science and applications of magnetism, it is important that we reestablish the teaching of magnetism as a priority in our universities as a whole rather than at only a few institutions that presently teach it. Much remains to be learned concerning the nature of spin-polarized transport. Questions need to be answered about the role of structure and the relationship of surface and interface structure to magneto-transport, the scattering mechanisms at interfaces in GMR, and the physics of the temporal and spatial decay of nonequilibrium magnetism. Also required is a detailed understanding of the mechanism of spin injection, either directly or through tunneling barriers from
OCR for page 90
Page 90 magnetic metals into metals and semiconductors. Answers to these types of questions are needed to engineer better GMR, CMR, and magnetic tunnel junction (MTJ) materials and devices. In a particularly useful spin transistor with true on and off states separated by many orders of magnitude in conductance will require very nearly 100 percent spin-polarized current injection from one region of the device into another. Advanced synthesis and processing techniques need to be developed to produce novel material with a high potential for scientific and technological impact. Layered structures, nanostructured three-, two-, and one-dimensional materials, and materials with higher BH-products are all attractive areas for further exploration. As in other areas of condensed-matter and materials physics, a systematic approachtechnically and organizationallyis needed to explore the vast phase space of magnetic materials. As applications develop, methods to bridge the gap between fabrication techniques that serve to produce initial demonstrations and more controlled and reliable techniques that can be migrated to development and manufacturing will be needed. Several difficult challenges remain in the measurement area. We need to understand how to magnetically probe individual electron and nuclear spins directly. Questions to be addressed include, What is the ultimate spatial resolution of magnetic measurement techniques? And, Can we fabricate a spin-polarized STM? Advances in measurement technology related to such questions will have a profound impact on our ability to understand the nature of magnetism in nanostructures. Finally, in the technology arena, continued focus on scaling the density for magnetic storage is needed. This will involve a strong, systematic, ongoing program in both media and detectors that will draw heavily on several of the condensed-matter and materials physics areas mentioned above. In addition, there is enormous opportunity in the arena of magnetoelectronics for new magnetic effects and devices that may set the stage for magnetism to play a key role in future microelectronic chip technology. A key element will be integration of complex magnetic materials with mainstream semiconductor technology. Future Directions and Research Priorities Numerous outstanding scientific and technological research needs have been identified in electronic, photonic, and magnetic materials and phenomena. If those needs are met, it is anticipated that these technology areas will continue to follow their historical exponential growth in capability per unit cost for the next few years. Silicon integrated circuits are expected to continue to follow Moore's Law at least until the limits of optical lithography are reached; transmission bandwidth of optical fibers is expected to grow exponentially with advances in optical technology and the development of soliton propagation; and storage density in magnetic media is expected to continue to grow exponentially with the
OCR for page 91
Page 91 maturation of GMR and development of CMR and MTJs in the not too distant future. Although these changes will have major impact on computing and communications over the next few years, it is clear that extensive research will be required to produce new concepts, as will new approaches to reduce research concepts to practice, if these industries are to maintain their historical growth rate over the long term. Continued research is needed to advance the fundamental understanding of materials and phenomena in all areas. For example, despite the extensive technological application and impact of magnetic materials and, despite more than a century of research in magnetic materials and phenomena, we lack a first-principles understanding of magnetism. By comparison, the technology underlying optical communication is very young. The past few years has seen enormous scientific and technological advances in optical structures, devices, and systems. New concepts such as photonic lattices, which are expected to have significant technological impact, are emerging. We have every reason to believe that this field will continue to advance rapidly with commensurate impact on communications and computing. As device and feature sizes continue to shrink in integrated circuits, scaling will encounter fundamental physical limits. The feature sizes at which these limits will be encountered and their implications are not understood. Extensive research is needed to develop interconnect technologies that go beyond normal metal and dielectrics in the relatively near term. Longer term, technologies are needed to replace today's silicon field-effect transistors. One approach that bears investigation is quantum state switching and logic as devices and structures move further into the quantum mechanical regime. A major future direction is nanostructures and artificially structured materials, which was a general theme in all three areas. In all cases, artificially structured materials with properties not available in nature revealed unexpected new scientific phenomena and led to important technological applications. As sizes continue to decrease, new synthesis and processing technologies will be required. A particularly promising area is that of self-assembled materials. We need to expand the research into self-assembled materials to address such questions as how to control self-assembled materials to create the desired one-, two-, and three-dimensional structures. As our scientific understanding increases and synthesis and processing technologies of organic materials systems mature, these materials are expected to increase in importance for optoelectronic and, perhaps electronic, applications. Many of the recent technological advances are the result of strong interdisciplinary efforts as research results from complementary fields are harvested at the interface between the fields. This is expected to be the case for organic materials; increased interdisciplinary effortsfor example, between condensed-matter and materials physics, chemistry, and biologyoffer the promise of equally impressive advances in biotechnology.
OCR for page 92
Page 92 In conclusion, the committee identifies a few major outstanding scientific and technological questions and research and development priorities. Major Outstanding Scientific and Technological Questions • What technology will replace normal metals and dielectrics for interconnect as speed continues to increase? • What is beyond today's FET-based silicon technology? • Can we create an all-optical communications/computing network? • Can we understand magnetism on the meso/nano scales needed to continue to advance technology? • Can we fabricate devices with 100 percent spin-polarized current injection? Priorities • Develop advanced synthesis and processing techniques, including those for nanostructures and self-assembled one-, two-, and three-dimensional structures. • Pursue quantum state logic. • Exploit physics and materials science for low-cost manufacturing. • Pursue the physics and chemistry of organic and other complex materials for optical, electrical, and magnetic applications. • Develop techniques to magnetically detect individual electron and nuclear spins with atomic-scale resolution. • Increase partnerships and cross-education/communications between industry, university, and government laboratories.
Representative terms from entire chapter: