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 14
Opportunities in High Magnetic Field Science 2 Scientific Challenges and Opportunities with Higher Fields Magnetic fields are powerful tools for studying the properties of matter because they couple directly to the electronic charge and magnetic moments of the protons, neutrons, and electrons of which matter is made up. The properties of most materials are only weakly dependent on the strengths of the magnetic fields to which they are exposed, and for these substances, magnetic fields can be used analytically to determine fundamental properties such as their characteristic electronic energy scales and the band structures of metals and insulators, the placement of atoms in molecules, or even the internal structure and dynamics of living creatures. On the other hand, in some materials the magnetic field couples strongly and dramatically influences their properties: for example, in quantum Hall devices, magnetic materials, and superconductors. For these substances, magnetic field strength is as important a thermodynamic parameter as temperature or pressure. Included in this category are many materials important for the production, control, and measurement of high magnetic fields such as high transition temperature (Tc) superconductors. As the committee argues elsewhere, improved understanding of these superconducting materials, which will derive in part from experiments done using state-of-the-art high-field magnets, will lead to the construction of better magnets. Research using high-field magnets has been remarkably fruitful in the past.1 1 For additional historical context, see National Research Council, High-Magnetic-Field Research and Facilities, Washington, D.C., National Academy Press, 1979; National Science Foundation, Final
OCR for page 15
Opportunities in High Magnetic Field Science (See Appendix A for a list of Nobel prizes awarded for research that used or significantly affected the development of high magnetic fields.) There is every reason to believe that it will continue to be so, especially if the field strengths of the magnets available to the scientific community continue to increase. In this connection, it is important to note that charged particles move in circular orbits in a magnetic field, the radius of which shrinks as the magnetic field strength increases. Similarly, the smallest size resolved by magnetic moment or spin probes shrinks with increasing field strength. Thus the need to study and characterize ever smaller objects, both those that exist in nature and those fabricated artificially, will not be satisfied unless magnets are fabricated that deliver fields of ever increasing strength and instrumentation is developed that supports their effective use. Paralleling the distinction made above, this chapter is divided into three sections. It begins with a discussion of high magnetic field research in condensed-matter and materials physics that emphasizes new phenomena that are likely to be revealed and known phenomena that would be better understood if higher fields were available. The chapter continues with a discussion of the impact of high-field magnets on the disciplines of biology, chemistry, biochemistry, and physiology as a result of their use in instruments that exploit nuclear magnetic resonance (NMR). In particular, the committee highlights the impact high magnetic fields have had, and continue to have, on the study of the solution structures of biological macromolecules by NMR, on solid-state NMR of biological and inorganic materials, and on electron paramagnetic resonance (EPR) of metal centers in proteins and catalysts. The committee discusses the impact high magnetic fields have had on two forms of magnetic resonance spectroscopy that have developed since the Richardson report—namely, magnetic resonance imaging (MRI) and ion cyclotron resonance (ICR) mass spectroscopy. In all these areas, magnets that operate at higher fields than those available today would yield large scientific dividends. CONDENSED-MATTER AND MATERIALS PHYSICS High-field research in materials science is intrinsically multidisciplinary, merging ideas from physics, chemistry, biology, and engineering, and integrating both theory and experiment. It is pursued predominantly by condensed-matter physicists, the largest subfield within physics today. Materials science, the dominant activity at the world’s high magnetic field laboratories, utilizes techniques as diverse as thermal and electrical transport, thermodynamic characterization, magnetization, optical spectroscopy, and magnetic resonance. Many classes of materials are Report of NSF Panel on Large Magnetic Fields, Arlington, Va., National Science Foundation, 1988 (also known as the Richardson report).
OCR for page 16
Opportunities in High Magnetic Field Science investigated, and measurements are done over a wide range of temperatures, pressures, and magnetic fields. In the 1920s, when high magnetic fields first became available in Europe, they were used initially to investigate simple metals and, later, semiconductors. This work resulted in the first experimental determinations of how individual electrons behave in solids and was extremely influential in the development of the theory of solids between 1930 and 1950. Among the many successes of this synergistic enterprise must be counted the first microscopic explanations of how electrical and thermal transport occur in metals and insulators, and why certain metals become magnetic. This work led ultimately to the development of the science that enabled invention of the first solid-state electronic device, the semiconductor transistor.2 It would be hard to overstate the impact of these developments on the economies of the industrialized nations in the second half of the 20th century. Electronic correlations are at the intellectual heart of modern condensed-matter physics. Interactions within populations of electrons lead to emergent collective properties that transcend those of individual electrons, such as superconductivity, magnetic order, and even the formation of the electronic gaps that distinguish metals from insulators. These properties reflect a balance of interactions among the electrons in a population and are strongly affected by differences in dimensionality, crystal symmetry, the spin of constituent atoms, and chemical bonding. Research on correlated-electron systems deals with issues ranging from the most fundamental (e.g., determination of the mechanism responsible for high-transition-temperature superconductivity in copper oxide layered compounds) to the most highly applied (e.g., learning how to control the microstructure of materials so that high-Tc superconductors with the highest possible critical fields can be produced for superconducting magnet construction). Historically, this field has been constantly refreshed and reinvigorated by the discovery of new materials, such as copper oxide superconductors, heavy fermion magnets and superconductors, organic conductors, and nanoscopic materials such as fullerenes. It has also benefited tremendously from the availability of low-dimensional semiconductor structures of improved quality and purity. As was the case for the noninteracting electron science of the early 1900s, experimental discoveries in this field have had a significant impact on the development of theoretical understanding, which in turn has led to fruitful suggestions about new directions to pursue in materials development. In the past three decades, eight Nobel prizes have been awarded for work in this field (2003, 1998, 1996, 1987, 1985, 1977, 1972, 2 Indeed, it was a combination of cyclotron resonance and the Hall and de Haas–van Alphen effects that helped characterize the electron transport properties of solids that enabled these inventions.
OCR for page 17
Opportunities in High Magnetic Field Science 1970), most recently in 2003 to A. Abrikosov, V. Ginzburg, and A. Leggett for their work on superconductivity and superfluidity. High magnetic field research in advanced semiconductor structures in particular led to the discovery of the integer and fractional quantum Hall effects, resulting in the physics Nobel prizes awarded to K. von Klitzing in 1985 and to R. Laughlin, H. Stormer, and D. Tsui in 1998. In addition to its intellectual importance, research in correlated-electron systems has already led to numerous technological advances, such as improvements in the sensitivity of the magnetic read heads used for information storage, which depend on the giant magnetoresistance of hybrid magnetic/metallic systems, and the improvements in communication that have resulted from the superior signal-to-noise ratios and interference rejection of high-Tc superconductor filters. The economic promise of this research area is enormous. Improvement in the properties of permanent magnets would impact both the efficiency of electric motors and the density and reliability of magnetic storage media. The quest to understand materials that become superconducting at high temperatures and to discover new materials that superconduct at even higher temperatures has already had important practical results. Improvements in magnetic field sensors and in key electronic components have resulted, as well as the development of high-field inserts for superconducting magnets, which will soon be used for research but may also have bioimaging applications. While currently only at the demonstration stage, superconducting power cables could have a huge economic and environmental impact by reducing power losses in electric transmission networks. Finally, electronic correlations induced by the collapse of metallic screening and finite size effects become increasingly important as the size of electronic components decreases. The trend toward miniaturization has naturally led to an increased interest in nanoscale devices that have novel electronic properties because the devices combine superconducting and magnetic components with more conventional semiconducting components.3 In every case, progress will be linked to the discovery of materials with improved collective properties. Understanding how electronic correlations are manifested in the macroscopic behavior of correlated-electron materials is key to the rational design of future generations of advanced materials. This task will require the skills of both experimentalists and theorists from a broad range of disciplines and will need the most advanced tools and techniques. The next section outlines the most important classes of correlated-electron materials and highlights the role high-field measure- 3 Parallel advances in the speed and miniaturization of electronics have allowed greater exploitation of high fields by enabling experiments in the compact, transient environments offered by pulsed-field magnets.
OCR for page 18
Opportunities in High Magnetic Field Science ments play in developing our understanding of them at both the fundamental and the technological level. Superconductors, heavy fermion compounds, and organic molecular metals are classes of complex materials in which magnetic, electronic, and structural properties are strongly related. When temperature, pressure, and doping are varied, the existence of multiple phases is often revealed. Paramagnetic, long-range, magnetically ordered, and superconducting phases are seen, which sometimes coexist. In high-Tc superconductors, the reference scale is the transition temperature Tc. In heavy fermion systems, it is the single-impurity Kondo temperature that competes with intersite magnetic couplings, while in organic conductors it is the coupling between chains or planes that often governs other properties. High-Temperature Superconductivity The discovery of high-temperature superconductivity in La2-xBaxCuO4 ceramics by J. Bednorz and K. Muller in 1986 inaugurated a new era in solid-state physics. Within the next 6 years, the family of high-temperature superconductors had expanded to include Y-, Bi-, Tl-, and Hg-based systems with maximum Tc ranging from 90 to 130 K, respectively, and more than 10,000 scientific papers had been published. Thus in the last decade of the 20th century, high-temperature superconductivity emerged as a major area in physics. Experiments done at high magnetic fields have contributed much to the characterization and elucidation of high-temperature superconductivity and indeed have revealed many of its more remarkable features. Why is high-temperature superconductivity so important, or, more precisely, why do so many condensed-matter physicists choose to work on this subject? Both fundamental and practical considerations come into play. On the fundamental side, the essence of the challenge is to solve the strong correlation problem. What happens when the electrons in a metal can no longer be described using the noninteracting electron paradigm, L.D. Landau’s theory of Fermi liquids? How do electron-electron interactions change a half-filled band, which Landau’s Fermi liquid theory indicates should make an excellent metal, into an insulating antiferromagnet? Nevertheless, the great theoretical challenges presented by the strong correlation problem do not in themselves explain the high level of international activity in this area. There is an additional ingredient—namely, the prospect that we might someday be able to make superconductors that work at room temperature and above. The idea of practical, room-temperature superconductors, with their distinctly quantum mechanical properties, such as the Meissner and Josephson effects, is tantalizing. What we now know about the mechanism of superconductivity in
OCR for page 19
Opportunities in High Magnetic Field Science the materials being investigated, which necessarily arises from Coulomb interactions and quantum statistics, suggests that transition temperatures of several hundred kelvin might be possible. It is this combination of fundamental theoretical importance and exciting practical potential that drives the field. Avenues of Research All high-temperature superconductors share a key feature that appears to be responsible for their high-temperature superconductivity: the presence of planes containing Cu and O atoms separated by bridging materials that act as charge reservoirs for those planes. These materials become superconducting at temperatures significantly higher than those of the previously known highest-Tc compounds, which are now called low-temperature superconductors: Nb compounds, for instance, have a maximum Tc around 23 K. High-Tc materials also have extraordinarily high upper critical magnetic fields (Hc2)—for example, 170 T for the widely studied YBCO and maybe 500 T for bismuth- and thallium-based compounds. On the one hand, the high critical fields of these materials make them attractive as conductors for use in high-field magnets, but on the other, their high critical fields are a serious barrier to their full characterization. The critical fields of many of these materials are so high that their normal (nonsuperconducting) states cannot be studied using even the most powerful magnets available today.4 Superconductivity in conventional materials is explained by a theory proposed by J. Bardeen, L. Cooper, and R. Schrieffer (BCS theory) and is understood to result from an interaction between electrons and phonons that causes an effective attraction between the electrons, allowing them to pair up. When a conventional superconductor becomes superconducting, the transition to this new, paired state causes a reduction in the potential energy of its charge carriers and a slight increase in their kinetic energy. The net amount of energy released is defined as the condensation energy. In many materials, the electron pairs that result have a fully symmetric internal symmetry, which is a natural consequence of phonon-mediated pairing. In the cuprate high transition temperature superconductors, condensed pairs have a different symmetry, which is indicative of an entirely different pairing mechanism.5 Thus models for superconductivity in these materials propose a different “glue” for binding carriers together. 4 That is, studies of the normal state at relatively low temperature; even the HTS materials available today with the highest Tc are not superconducting at room temperature. 5 The working fluid of superconductors consists of pairs of electrons (or pairs of the holes left behind in a crystal when an electron moves somewhere else). These Cooper pairs form a coherent state with specific symmetry properties.
OCR for page 20
Opportunities in High Magnetic Field Science There are several reasons for the extraordinary focus on high-temperature superconducting (HTS) materials in recent years: their intrinsic scientific interest; the cross-disciplinary nature of the field, which reaches across boundaries that often divide materials scientists and chemists from experimental and theoretical physicists; the potential applications of materials that superconduct at temperatures above the boiling point of liquid nitrogen (77 K); and, finally, the possibility of finding a superconductor that has a critical temperature above room temperature. Applications for HTS materials include filters for cellular phone systems; superconducting transmission lines, generators, motors, transformers, and fault current limiters; higher field MRI instruments and NMR spectrometers; microwave systems; and (of course) magnetically levitated transportation systems. The scientific challenge posed by HTS materials is more fundamental than simply understanding why they superconduct. The oxide high-Tc superconductors are a family of materials in which even the properties of the normal state are not as well understood as they are for metals like aluminum, lead, or niobium. The identities of the carriers of charge and spin—that is, the HTS equivalents of the electrons and holes in metals, semiconductors, and low-temperature superconductors—are still being debated. Thus one of the key challenges posed by these materials is understanding the physics of their normal states, either at temperatures above Tc or at fields high enough to quench their superconducting states. Since the low-temperature/high-field regime is inaccessible for many of these compounds because current magnets do not deliver fields high enough, most measurements of the HTS normal state have been done above Tc. This approach is often unsatisfactory because thermal energies are so large at those temperatures that the details of the physical phenomena of interest are obscured by thermal fluctuations.6 The competing phases of magnetism and superconductivity that exist in HTS materials are illustrated in the phase diagram provided in Figure 2.1. As the carrier doping (number of holes) is increased in these materials (usually by raising the oxygen content), they are transformed from an antiferromagnetic insulator into a metallic superconductor that has a Tc that is also dependent on carrier concentration and peaks at a level termed “optimal doping.” While this behavior has been understood for some time now, more recently a pseudo-gap regime has been added to the phase diagram. It is believed that in this range of doping and 6 Note that transition temperature is not the only driving parameter. One might think that research in the low-temperature/high-field limit might best be done with HTS materials that have low transition temperatures. This is not necessarily so, because these materials are often less amenable to analysis using techniques such as photoemission and optical spectroscopy. Finally, sample purities for the different families of cuprate HTS compounds can vary widely.
OCR for page 21
Opportunities in High Magnetic Field Science FIGURE 2.1 Generic phase diagram for CuO2 (cuprate) superconductors, showing temperature versus doping concentration (the latter variable maps onto the order parameter for the material). The properties of the cuprates vary with temperature (y axis) and the doping per unit cell of CuO2 (x axis). Theorists are unable to explain why the superconducting transition temperature (thick black line) is so high in the cuprates. However, if they could understand the behavior of the cuprates in the pseudogap region (blue), they might be able to explain high-temperature superconductivity. temperature (above Tc), the carriers are paired but the pairs do not yet form a superconducting state. As described in greater detail below, research using high magnetic fields has been critical to uncovering the secrets of high-temperature superconductivity. It is expected that high magnetic fields will continue to be essential as understanding of this phenomenon grows and potential applications are realized.
OCR for page 22
Opportunities in High Magnetic Field Science Although the primary goal of HTS research is to understand the origin of the superconducting state in materials with the highest transition temperatures, much can be learned from the study of HTS materials with lower transition temperatures, because there is every reason to believe that the underlying physics is the same in all of them. The most promising material of this kind is YBa2Cu3O7-x (x ≈ 0.5, Tc ~60 K). It has approximately the same transition width, and the same level of intrinsic disorder as optimally doped YBa2Cu3O7-x (x ≈ 0.05, Tc ~92 K). The lower-Tc form has a lower carrier concentration, which should make it an easier material in which to observe quantum oscillations and cyclotron resonance. These phenomena can be used to characterize the shape of the Fermi surface in the normal state and to yield the effective masses of the carriers at the Fermi energy. Experiments with YBa2Cu3O6.5 in the normal state have additional advantages; the temperature can be stabilized using liquid nitrogen (~77 K). It is fundamentally important to understand the behavior of the upper critical field, Hc2, of HTS materials as a function of temperature, which is why determination of the Hc2 of HTS materials at low temperatures has become an important issue in HTS research. So far most studies have been limited to fields less than 20 T, which are well below the zero-temperature values of Hc2 (~100 T). In addition, high magnetic fields may prove invaluable in revealing the nature of the pseudo gap in high-temperature superconductivity, a potential key to understanding the microscopic mechanism of the HTS state. These experiments will require very high magnetic fields—around 100 T—and at lower temperatures, even higher fields may be required. High magnetic fields (>35-40 T) are a requirement for studying the H-T (magnetic field and temperature) phase diagram of the recently discovered two-gap superconductor MgB2. (For more discussion of MgB2, please see the section “Emerging Superconducting Materials” in Chapter 3.) The superconducting energy gap is essentially the energy needed to break the Cooper pairs apart: It also determines the thermodynamic properties of the material and is directly related to the superconducting transition temperature. Most superconductors have just one energy gap, but experiments suggest that magnesium diboride (MgB2) has two. The gaps correspond to transition temperatures of 15 K and 45 K and combine to give an overall transition temperature of 39 K. A spectacular increase in the upper critical field of MgB2 was achieved recently by selective alloying of s and p bands with nonmagnetic impurities. Hc2(0) values have increased tenfold: from 3 to 5 T for single crystals to 35 T for H∥c and to 50 T for H∥ab for dirty MgB2 samples. These advances offer an exciting opportunity for studying the novel physics of two-gap superconductivity, nonequilibrium interband phase textures, and vortex dynamics and pinning at high magnetic fields greater than 50 T. Because such studies require sweeps of magnetic field strength, extended averaging times for sensitive measurements, and carefully controlled conditions so that samples can be compared, steady-state fields are generally required.
OCR for page 23
Opportunities in High Magnetic Field Science Opportunities in Vortex Physics Work on the mechanisms of superconductivity, particularly high-temperature superconductivity, has revealed a rich new scientific landscape. One of the important areas opened up by these activities, vortex physics, is described here in some detail because of its intellectual vitality. There are two classes of superconducting materials: Type I and Type II. Type I superconductors correspond loosely to the pure element low-temperature superconducting (LTS) materials Hg, Sn, and Pb, in which superconductivity was first discovered. External magnetic fields are fully excluded from the bulk of Type I superconductors by surface currents flowing within the London penetration depth, and the critical fields at which superconductivity is lost (quenched) are very low, less than 0.1 T, which is why Type I superconductors have few commercial applications. Type II materials have much higher critical fields because they enter a mixed state at a lower critical field Hc1 of ~0.01 T (see Figure 2.2). Above Hc1, Abrikosov vortices form, which consist of flux tubes containing a quantum of flux 0 = 2 × 10–15 Wb. Each field filament is encircled by a supercurrent flowing to a depth equal to the London penetration depth. The centers of these vortices can be viewed as normal cores where the superconducting order parameter is suppressed. When a current is driven through the material, the flux lines experience a Lorentz force that tends to push them perpendicular to the current. If they so move, the process is dissipative and introduces resistance. This phenomenon is well understood, but exactly how magnetic fields penetrate into superconductors, how the flux lines move, and how they interact with defects in the material are not well known. The flux lines can also repel each other, so flux flow is a complex, many-body effect. We are now coming to understand that flux flow is like other dissipative effects such as earthquakes and avalanches, and the physics of vortex motion has many parallels to other areas of physics. This quantized flux-tube state was first conceived of by A. Abrikosov to describe the situation that occurs when the superconducting coherence length ξ is much shorter than the penetration depth λ. He found that the vortex state was both quantized and characterized by a lattice structure, the existence of which was subsequently confirmed by neutron diffraction, magnetic decoration, and magneto-optical and transmission electron microscopy experiments. A particular curiosity of Type II superconductors is that the interface energy between their normal and superconducting regions is negative, making fine-scale subdivision of the vortex state energetically favorable. The vortex density depends on magnetic field as (0/B)0.5. Bulk superconductivity is destroyed when the normal cores of the flux tubes in a material overlap, which occurs at a field Hc2 of . Hc2 values can be remarkably high. For LTS materials such as Nb-Ti or Nb3Sn, the values are
OCR for page 24
Opportunities in High Magnetic Field Science FIGURE 2.2 Phase diagram for penetration of the magnetic field into a Type II superconductor. The green mixed-state region is where Abrikosov vortices are formed and vortex physics comes into play. about 15 T and 30 T, respectively, but for cuprate superconductors with high Tc, Hc2 can exceed 100 T. LTS and HTS Type II superconductors differ in the degree of interaction among the vortices with each other and with defects in the material; HTS materials offer a much richer spectrum of physics. In the essentially isotropic LTS (Type II) metallic superconductors, vortices are line objects with significant line tension and are thus effectively pinned by even dilute microstructural defect arrays, provided
OCR for page 58
Opportunities in High Magnetic Field Science higher field magnets for solution NMR has been an incremental process. NMR spectrometers operating at 1H NMR frequencies of 270 MHz, 360 MHz, 400 MHz, 500 MHz, 600 MHz, 750 MHz, 800 MHz, and 900 MHz have been developed in sequence, with each step requiring 2-4 years to accomplish (see Figures 2.6 and 2.7). Most of these steps took place in industry, but with an increasing degree of academic research collaboration. For example, one of the most recent 900-MHz spectrometers to come on line was developed almost entirely at NHMFL. It is likely that instruments operating at still higher fields will one day be developed in a similarly incremental way and that this will gradually increase the complexity of the systems that can be characterized successfully by NMR. It is important to realize, however, that although each step in the process has been small, the cumulative effect has been large. The 800- and 900-MHz NMR spectrometers available today are vastly superior to the 360-MHz spectrometers that came on the market 20 years ago in almost every respect. If NMR spectrometers with stable, homogeneous fields of 30 T or higher (1.3 GHz or higher) were to become available in the future, the impact on both biological and nonbiological users would be considerable. The increased sensitivity and resolution of such an instrument would probably allow using solution NMR to determine the structure of proteins two to five times larger than the proteins whose structures can be determined today. The tendency of macromolecules to align in magnetic fields, which is detectable but quite small in today’s NMR spectrometers, would become more significant (scaling as the square of the field strength), enabling new approaches to structure determination through anisotropic nuclear spin interactions. Biological solid-state NMR measurements would be qualitatively transformed, principally because 1H NMR signals would become well enough resolved to begin to be as useful and informative as they already are in solution NMR. Progress would be made in solving the general problem of membrane protein structures. Recent studies of relatively small model proteins in microcrystalline form have demonstrated the feasibility of full structure determination by solid-state NMR. The improved spectral resolution at higher fields will facilitate the extension of these results to larger membrane proteins. Solid-state NMR spectroscopy of inorganic materials would improve dramatically as second-order quadrupole effects become small spectral perturbations rather than the dominant feature in NMR spectra of quadrupolar nuclei. As mentioned earlier, certain types of NMR have less stringent requirements for magnet homogeneity and stability than solution NMR. In particular, many biological and nonbiological solid-state NMR measurements can be performed in fields with about 1 ppm homogeneity over 0.1 cm3 (rather than 1 ppb over 1 cm3). Furthermore, magnet instabilities are manageable if the field drift can be calibrated or monitored during experiments. Provided drift rates are not too great, it
OCR for page 59
Opportunities in High Magnetic Field Science FIGURE 2.6 The magnet of the 900-MHz NMR spectrometer in Pacific Northwest National Laboratory’s Environmental Molecular Sciences Laboratory. This picture illustrates some of the issues associated with high-field magnets. This 21-T magnet was manufactured by Oxford and has a 63-mm room-temperature bore and a stored energy of 27 MJ; it is sited inside a cylindrically shielded enclosure 24 ft in diameter and extends 15 ft above and below the main laboratory floor (the magnet itself is 8 ft in diameter and 21 ft tall). The center of the magnetic field is located at exactly floor level to preserve the desired symmetry of the magnet and shield. The person standing next to the magnet is just over 6 ft tall. Photo courtesy of William R. Wiley Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory.
OCR for page 60
Opportunities in High Magnetic Field Science FIGURE 2.7 The growth of the field strength of NMR spectrometers in common use. The types of superconducting wire used for the highest-field NMR magnets in each era are indicated at the top of the figure. Before 1992, magnets operated at the temperature of liquid helium (about 4 K). Since then, the highest-field magnets have required pumped-helium cooling systems that provide even lower temperatures (about 2 K). Current technology (Nb-Ti, Nb3Sn, and (Nb,Ta)3Sn) may enable 1-GHz operation, but new high-field conductors based on MgB2 or HTS materials will have to be developed for higher frequencies. (The committee notes that a 600-MHz NMR magnet was introduced in 1978 at Carnegie Mellon University; the magnet was constructed of Nb3Sn tape and was nonpersistent, setting it apart from the more heavily commercialized NMR magnets. As a proof-of-principle, this design may have important value for the future.) Figure courtesy of Bruker Biospin, Inc. is possible to correct data for magnet drift either in real time or after the fact. Thus, a 30- to 50-T magnet for NMR, with 1 ppm homogeneity over 0.1 cm3 and 1 ppm/min drift specifications, would be valuable even if it could not be used for high-resolution NMR of proteins in solution. For any such a magnet to achieve its full potential, however, ancillary equipment would have to be developed. In particular, NMR probe and cold-probe circuitry that permits double- and triple-resonance experiments at 1H NMR frequencies of 1.3 GHz and above will have to be developed and incorporated into the probes, including those with high-speed, magic-angle spinning capabilities. In addition, capabilities for both low-temperature and high-temperature measurements must be part of the package as well as radio-frequency power amplifiers that
OCR for page 61
Opportunities in High Magnetic Field Science produce ~500-W pulses at the NMR frequencies of hydrogen and other relevant nuclei. Finally, the spectrometer console, which controls the execution of pulse sequences and the acquisition of signals, must have the full capabilities of a modern NMR instrument. Strategic Considerations for Higher Field NMR The committee believes that the construction of 1.3-GHz NMR spectrometers is a scientifically justifiable objective. Its endorsement is strongly influenced by the fact that in the past, NMR spectroscopy has shown an astonishing capacity to reinvent itself in productive but unanticipated ways in response to improvements in instrumentation. However, the committee realizes that this enterprise will present major challenges for the agencies that support research dependent on high-field NMR spectroscopy, the people who manufacture these instruments, and the user community. The highest-field NMR spectrometers available today, which operate at 900 MHz, cost approximately $5 million each, and the cost of a 1.3-GHz spectrometer, assuming that it can be built, will certainly be much higher, perhaps as much as $20 million. Thus, unlike the number of 600-MHz NMR spectrometers available in the United States today, for example, the number of 1.3-GHz NMR spectrometers will never be large, and demand for access to them is certain to exceed supply. Two conclusions follow: The technical challenges that surround the construction of the magnets required for these spectrometers are so great and the potential market for them is so small that manufacturers are unlikely to undertake their construction without at least some direct public support. Thus, in addition to playing their traditional role as the main purchaser of the product, federal agencies interested in high-field NMR research will have to shoulder some of the risk up front or these instruments will not be built. Although the NMR data obtained with a 1.3-GHz spectrometer is expected to be of substantially better quality than that obtained with current spectrometers, which should allow new problems to be addressed, the throughput of these instruments might only be twice that of current 900-MHz instruments. Thus the NMR community will need to address the challenge of providing fair access to these spectrometers for all interested parties while at the same time ensuring that only experiments that absolutely require the highest available fields are performed on them. Given that the NMR community is used to running experiments on locally controlled instruments rather than sharing centralized facilities, some social engineering may be required to ensure a sensible outcome.
OCR for page 62
Opportunities in High Magnetic Field Science Ion Cyclotron Resonance The phenomenon of ICR can be used to determine atomic and molecular masses. The technique is called ion cyclotron resonance mass spectroscopy (ICRMS). In an ICRMS instrument, modest numbers of ionized molecules are introduced into an ion trap in the middle of a homogeneous magnetic field. The trajectories of such ions are circular, viewed along the direction of the field, and the frequencies at which they orbit are determined by their mass-to-charge (m/z) ratios, and are proportional to magnetic field strength. The radii of ion orbits at the time of injection are small, and since the ionized molecules in the injected sample are randomly distributed around those orbits, viewed from the outside, the circulation of these ions has no net electrical effect. Ions can be forced to enter orbits of higher radius by the application of radio frequency pulses of the appropriate frequency, duration, and polarity, and the coherent orbital motions of the ion clouds that result induce oscillating voltages in detection circuitry that surrounds the ion traps, from which m/z values can be determined. (The similarities between ICRMS and NMR experiments are many and obvious.) Because the lifetimes of the coherent cyclotron motion of such clouds, which are limited by collisions and by magnetic field inhomogeneities, can be quite long, m/z values can be determined with remarkable precision (~1 ppm) by ICRMS. The molecular weights of the components of mixtures containing thousands of chemically distinct ions can be determined simultaneously by ICRMS. The mass accuracy can be less than that of a single electron, so that chemical compounds with the same nominal molecular weight but different elemental compositions can be distinguished by ICRMS. ICRMS is rapidly growing in importance. One application is in the field of proteomics, where the goal is to determine the full array of proteins expressed within a given cell type or tissue (and the posttranslational modifications of these proteins)—for example, as a function of exposure to hormones, drugs, or other bioactive compounds; as a function of developmental stage; or as a function of disease state. These measurements can be done with ICRMS by extracting proteins from cells or tissue, fragmenting them into shorter peptide segments, and then determining the masses of all fragments. The sensitivity of ICRMS is remarkably high: About 100 ions of the same species can generate a detectable signal. Thus, it is entirely compatible with the usual scales of molecular biology experiments. Because of the sensitivity and resolution of ICRMS spectra, ICRMS methods may be employed in the future to establish biomarkers for disease states in humans. ICRMS is also a tool of unparalleled power for determining the composition of complex mixtures such as crude oil. ICRMS benefits from high magnetic field. Commercial ICRMS spectrometers currently operate at fields up to 12 T, and instruments up to 14.5 T are under
OCR for page 63
Opportunities in High Magnetic Field Science construction. Provided that field homogeneity is not limiting, resolving power increases linearly with field, and upper mass limit and dynamic range improve quadratically. Homogeneity requirements are not as stringent as in high-resolution NMR because the orbital motions of ions tend to average out field inhomogeneities. Thus, magnets with about 10 ppm homogeneity and about 10 ppm short-term stability are suitable for ICRMS. High-field resistive magnets could conceivably be used. ICRMS is still in its infancy, and it has benefited mightily from the many techniques that have been developed in recent years for preparing molecular ions for mass spectrometric analysis. New applications are being discovered every day, and it is likely that before long, every research university will feel it must have ICRMS spectrometers on its campus. As it happens, ICRMS is one of the big success stories at NHMFL, whose Center for Interdisciplinary Magnetic Resonance in Tallahassee, Florida, has been an important locus for the development of ICRMS ever since it was founded. Electron Paramagnetic Resonance The phenomenon of EPR was discovered in bulk matter by E.K. Zavoisky in 1944, before the initial observations of NMR. EPR and NMR spectroscopy are similar in many respects; both examine the response of a sample exposed to radiofrequency radiation in a magnetic field. In EPR, the species detected are unpaired electrons. In NMR, it is nuclei that have nonzero spins. The technical differences between the two kinds of spectroscopy derive mainly from the difference in gyromagnetic ratio between electrons and nuclei. In a 1-T field, for example, the resonant frequency of a simple organic free radical is about 28 GHz, but the resonant frequency for protons is only 43 MHz. The wavelength of 28 GHz radiation is around 1 cm, which implies that a 1-T EPR spectrometer must be equipped with microwave electronics, a constraint that limits what can be done. Microwave devices tend to be relatively narrow-band, so EPR spectra are usually collected by setting the electronics of the spectrometer to a single frequency and then sweeping the field strength of its magnet over a wide range so that the different paramagnetic species in the sample can be brought into resonance one at a time. EPR spectrometers available commercially today operate at 95 or 130 GHz and have 3.5- to 5-T magnets. As is the case for NMR and ICRMS, in principle, the performance of EPR spectrometers improves with increasing magnetic field, but there are limitations. If an EPR spectrometer were built around a 21-T magnet, i.e., a state-of-the-art NMR magnet, the resonant frequencies for simple organic free radicals would rise to about 588 GHz and wavelengths fall to about 500 µm. For some of the metal
OCR for page 64
Opportunities in High Magnetic Field Science centers (in metalloproteins and surface-catalytic metals) interesting to chemists and materials scientists, the frequencies might be even higher, about 6 THz, and the wavelengths even shorter, about 50 µm. Radiation having these characteristics lies in an awkward part of the electromagnetic spectrum, between microwaves and the far infrared; it is hard to work with. The development of an integrated spectrometer operating at 21 T would present major challenges, but the rewards could be commensurate. New radiation sources, phase shifters, resonators, digital switches for pulse generation, power amplifiers, and preamplifiers would all have to be devised. However, if they were, it would become possible to do EPR experiments equivalent to the multidimensional, double-resonance experiments now routine in NMR at all field strengths and in EPR only at low field (9 GHz). The impact would be transformative. The spin Hamiltonian explored by NMR spectroscopy is simple to deal with theoretically because each of its terms is three to six orders of magnitude smaller than the dominating Zeeman interaction. It is different for EPR spectroscopy, because many of the internal interactions that must be taken into account are comparable to, if not larger than, the electron Zeeman interaction. Thus, electron spin Hamiltonians often depend on a dozen or more parameters that cannot easily be determined independently. At high field, EPR spectroscopy will approach the limit where these parameters can be measured accurately from first-order spectra. In addition, many interesting metal-centered species important in catalysis and materials have EPR transitions that can be measured only at high fields because of the large internal fields inherent in the metal ion or the material containing them. Specialized, one-of-a-kind continuous-wave EPR spectrometers have been used in Europe, Israel, and NHMFL to demonstrate that metal centers of the kind just described can indeed be observed at high field in catalysts and optoelectronic and magnetic materials, demonstrating the potential of high-field EPR. It is time to develop the full range of pulsed EPR capabilities for high field so that the measurements now made routinely on simple organic free radicals and some metal centers at conventional EPR fields can be done on systems that require high field. Development of pulsed EPR spectrometers that operate at 15 T and beyond is already under way in the university and government labs of Germany, the Netherlands, and Italy. The United States should not allow itself to fall behind in this area. Importance of Ancillary Technological Development As has been pointed out repeatedly above, technological and methodological developments have led to constant advancement of the capability of magnet-based measurement devices and a concomitant increase in their impact on many areas of science. The development of higher field magnets has played a central role in these
OCR for page 65
Opportunities in High Magnetic Field Science advances to be sure, but the importance of developments in other areas of technology should not be forgotten. In fact, without these advances in ancillary technologies, the full potential of higher-field magnets would not have been realized, and, given the expense of high-field magnets, investments in technologies needed to optimize their utility makes good economic—as well as scientific—sense. While federal funding for the application of existing technology and methods to specific scientific problems has generally been good, federal funding for the development of novel technology and methodology has been inadequate. NMR and MRI instrument manufacturers have done a good job of advancing the ancillary technology relevant to these techniques when relatively large commercial markets for their products justified their doing so. For example, the recent development of cryogenically cooled probes for high-field solution NMR, which enhance sensitivity severalfold, was carried out entirely by instrument manufacturers, based on their expectation about the market for them. However, there are many other areas where technological advances are sorely needed but where the commercial market is not large enough to attract the attention of instrument manufacturers. Several examples follow. Biological solid-state NMR measurements on integral membrane proteins, which, in principle, could be used to determine the structures of receptor-bound hormones, neurotransmitters, and other biological signaling molecules, are severely limited by the availability of appropriate samples. It is often very difficult to make such materials in quantities sufficient for the instrumentation now available. This problem would be greatly alleviated by the development of magic-angle spinning probes for solid-state NMR that operate at 10-30 K. Further sensitivity enhancements might be obtained by techniques such as dynamic nuclear polarization (DNP) and optical pumping. Recent DNP experiments, in which paramagnetic compounds are added to frozen solutions to permit cross-relaxation between electron spins and nuclear spins during microwave irradiation of the electron spins, suggest that signal enhancements of two orders of magnitude may be achievable quite generally with further developments in microwave technology, solid-state NMR probe design, and paramagnetic reagents. In the area of MRI, the advent of whole-body imaging systems operating at 8 T and above creates new problems in detector coil design, largely because the wavelengths of the radio-frequency radiation detected are comparable to anatomical dimensions, making the amplitudes of the radio-frequency fields within the body highly nonuniform. Optimal coils for high-field MRI will probably not be developed unless a significant research program is undertaken by groups outside the commercial sector, and this is unlikely to
OCR for page 66
Opportunities in High Magnetic Field Science happen without federal support. Similarly, MRI microscopy at the highest available fields, which could be used for noninvasive imaging at the cell or organelle level within model organisms, would benefit significantly from the development of improved microcoils and field gradient coils. Hybrid magnets with fields greater than 40 T may not have the homogeneity and stability required for biological NMR, but they do have considerable potential for NMR studies of inorganic materials and phenomena in condensed-matter physics. Realization of this potential will require the design and construction of appropriate NMR probes with high- and low-temperature capabilities and the development of field stabilization methods to extract the maximum possible spectral resolution and sensitivity. Finally, perhaps more than any other class of experimental techniques now in common use, NMR and MRI have both benefited from methodological advances that are purely concept-based, as opposed to equipment-based. Physical, chemical, and biological scientists have benefited enormously from the design of radio-frequency pulse sequences that excite and manipulate nuclear magnetic moments in new ways. Developments in pulse sequence methodology have usually resulted from new insights into the way nuclear magnetic moments evolve in external and internal magnetic fields and from new ways to describe these evolutions mathematically. Because higher fields significantly change the relative strengths of the interactions that determine how nuclear magnetic moments evolve in response to radio-frequency pulse sequences, improvements in pulse sequences will be required if the field is to take full advantage of high-field magnet development. Projects aimed at improving NMR and MRI capabilities, from which large numbers of investigators are certain to benefit, are surely as worthy of support as the specific projects based on current NMR and MRI techniques. A strong program requires a healthy balance between research aimed at improving current capabilities and research that exploits current capabilities. OTHER SCIENTIFIC USES OF HIGH-FIELD MAGNETS High-field magnets are critical components of instruments used in several research areas that so far have been mentioned only in passing, most notably high-energy and nuclear physics, plasma science, and fusion energy research. Advances in high-energy physics have long been strongly coupled to developments in magnet technology, because increases in magnet performance have been required for the construction of ever more power particle accelerators and particle detectors.
OCR for page 67
Opportunities in High Magnetic Field Science The U.S. high-energy physics community established an early dominance in superconducting accelerator technology by constructing the Tevatron at Fermi National Accelerator Laboratory (FNAL), which is still the world’s highest energy accelerator. Its 4-mile circumference ring includes 1,000 Nb-Ti superconducting magnets supplying a field of 4.4 T. It began operation in 1983 and has been upgraded over time to reach an energy of 1 TeV. An even larger accelerator is the Large Hadron Collider (LHC) at CERN, designed to collide proton beams at 14 TeV. Still under construction, it will use a dual-aperture beamline 27 km in circumference, which includes 5,000 Nb-Ti superconducting magnets. These magnets will operate at 1.9 K with a peak field of 8-9 T. The expected start-up date is 2007. Because Nb3Sn is brittle, its use requires special magnet construction techniques that are more costly than those for the tough and ductile Nb-Ti. Thus, unless the higher-field performance is required, Nb-Ti is generally employed, as it was for the LHC, where it was found easier to design with Nb-Ti than Nb3Sn. A prototype 16-T dipole magnet for the next generation of accelerator dipole magnets (for proton machines) was demonstrated in 2003 at LBNL; these magnets will require Nb3Sn, MgB2, or an HTS conductor. High-energy physics magnets are predominately dipoles, quadrupoles, and higher order configurations, but although their field and force distributions are very different from the solenoids used for most other purposes, the stresses and other engineering problems confronted in their construction are similar. The Tevatron and the LHC achieve their goals using huge quantities of Nb-Ti magnets, but it is unlikely that Nb-Ti magnets will suffice for the large circular accelerators that might succeed the LHC. The field strengths attainable by Nb-Ti magnets will simply not be enough. For this reason, the U.S. high-energy physics community has established a research program to develop Nb3Sn-based dipoles and quadrupoles. LBNL has achieved a 16-T peak field in a small dipole prototype coil using an advanced, very high critical current density Nb3Sn wire. Other laboratory partners in the DOE high-energy magnet technology effort include Fermilab and Brookhaven National Laboratory. Fields up to 20 T may be achievable with further development, but this will require a focused program devoted to the commercial production of large quantities of very high current density Nb3Sn wire. Because the different magnet user communities have been isolated from one another, developments in accelerator magnet technology have not had a broad impact outside the field. Recently, however, researchers at FNAL and LBNL have begun to form partnerships with fusion science magnet designers and superconducting materials experts outside the traditional laboratory centers. Research into the development of fusion as a future energy source has been going on around the world for years. Fusion devices are operating in Europe,
OCR for page 68
Opportunities in High Magnetic Field Science Russia, and Japan, and new ones are being constructed in South Korea, China, and India. All the fusion devices being used or being considered require very large superconducting magnets. The United States is now operating only one superconducting fusion device, known as the Levitated Dipole Experiment. However, it has had an extensive superconducting magnet development program under way since the 1970s. In the 1980s the extremely large superconducting mirror machine MFTF-B was started up at Lawrence Livermore National Laboratory, but it never went into full operation for reasons unrelated to magnet technology. Fusion magnets come in many shapes and sizes, including solenoids, toroids, and helical coils. The devices presently in operation use Nb-Ti magnets, but newer machines will use Nb3Sn magnets. The largest project now being planned is the International Thermonuclear Experimental Reactor (ITER), which will be a collaboration between the United States, Europe, Russia, Japan, China, South Korea, and India. The device will cost more than $5 billion and is scheduled to be constructed over 8 years beginning in 2006. This machine will require the commercial production of about 500 tons of high-quality Nb3Sn superconductor over a several-year period, a more than 10-fold increase in world production of Nb3Sn. During the 1990s the parties involved in ITER made several large-scale prototype superconducting magnets. The largest of these was the Central Solenoid Model Coil, built jointly by the United States and Japan. Its coil has an inner diameter of 1.6 m, an operating current of 46,000 A, a peak field of 13 T, and a stored energy of 640 MJ. It can be operated as a DC magnet or ramped from zero field to 13 T in 8 s without quenching. These examples suggest the broad utility and critical importance of high-field magnet science and technology in fields beyond condensed-matter physics, materials science, and magnetic resonance. Although the scale of application for high-field magnets in a high-energy particle accelerator is vastly different than that associated with the study of correlated-electron systems, both communities drive—and benefit from—general advances in high-field magnet technology. It is important to note that these disparate communities of users have not traditionally collaborated on magnet technology. Indeed, the recent convergence of the particle physics and fusion science magnet efforts was precipitated more by their shared source of funding (DOE’s Office of Science) than by any overlap of ongoing research efforts. It is nevertheless the case that advances in magnet design, construction, and performance made by one community can significantly benefit other communities.
Representative terms from entire chapter: