8
Basic Plasma Experiments

INTRODUCTION AND BACKGROUND

Plasma physics deals with the behavior of many-body systems under the influence of long-range Coulombic forces. Plasmas are inherently nonlinear media and, in the presence of a magnetic field, they are also anisotropic. Consequently, plasmas are capable of sustaining a wide variety of waves and instabilities. Plasmas can support three-dimensional currents and exhibit nonlocal behavior and "memory effects" (e.g., within the particle distribution functions). Plasma instabilities can lead to chaotic particle motions, to intricate wave dynamics, and to turbulence. Thus, understanding plasma phenomena involves fundamental aspects of statistical mechanics, fluid dynamics, electrodynamics, and frequently, atomic physics.

Progress in basic science has historically relied on a close interaction between experiment and theory. This is particularly true of plasma physics, where nonlinear and nonequilibrium phenomena in many-body systems are of central importance. In striking contrast to the central importance of laboratory experiments to this field, it is the finding of the panel that activity in and support for basic experiments has decreased markedly over the last two decades. For example, at the 1973 plasma physics division meeting of the American Physical Society, there were 126 papers on basic experimental plasma physics. In contrast, at the 1992 meeting, there was no general session on basic laboratory experiments, and there was only one poster session on laboratory experiments related to space plasmas. At this meeting, there were only about 30 experimental papers on basic plasma physics that were not related to a particular application,



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



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications 8 Basic Plasma Experiments INTRODUCTION AND BACKGROUND Plasma physics deals with the behavior of many-body systems under the influence of long-range Coulombic forces. Plasmas are inherently nonlinear media and, in the presence of a magnetic field, they are also anisotropic. Consequently, plasmas are capable of sustaining a wide variety of waves and instabilities. Plasmas can support three-dimensional currents and exhibit nonlocal behavior and "memory effects" (e.g., within the particle distribution functions). Plasma instabilities can lead to chaotic particle motions, to intricate wave dynamics, and to turbulence. Thus, understanding plasma phenomena involves fundamental aspects of statistical mechanics, fluid dynamics, electrodynamics, and frequently, atomic physics. Progress in basic science has historically relied on a close interaction between experiment and theory. This is particularly true of plasma physics, where nonlinear and nonequilibrium phenomena in many-body systems are of central importance. In striking contrast to the central importance of laboratory experiments to this field, it is the finding of the panel that activity in and support for basic experiments has decreased markedly over the last two decades. For example, at the 1973 plasma physics division meeting of the American Physical Society, there were 126 papers on basic experimental plasma physics. In contrast, at the 1992 meeting, there was no general session on basic laboratory experiments, and there was only one poster session on laboratory experiments related to space plasmas. At this meeting, there were only about 30 experimental papers on basic plasma physics that were not related to a particular application,

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications and half of these papers were on nonneutral plasmas. It is the conclusion of the panel that the level of activity in basic plasma experiment in the past 20 years almost certainly has been lower than it would have been if there had been in place a well-planned and balanced program in basic plasma science in the United States. The danger is that basic experimental plasma science will disappear in this country, unless one or more funding agencies assume the responsibility to support a critical mass of scientists in this area. A survey of the plasma science community in the United States, conducted by the panel, shows that renewed support for basic laboratory plasma experiments is its highest priority. The panel has come to this same conclusion: The highest priority in establishing a healthy plasma science in the United States is renewed support for basic experimental research in plasma science. This conclusion coincides with the principal findings of the Brinkman report, Physics Through the 1990s:1 Direct support for basic laboratory plasma-physics research has practically vanished in the United States. The number of fundamental investigations of plasma behavior in research centers is small, and only a handful of universities receive support for basic research in plasma physics. A striking example is the minimal support for basic research in laboratory plasmas by the National Science Foundation.… Because fundamental understanding of plasma properties precedes the discovery of new applications, and because basic plasma research can be expected to lead to exciting new discoveries, increased support for basic research in plasma physics is strongly recommended. Support for basic plasma experimental research can be expected to serve an important educational function as well. University-scale experimental research programs in basic plasma science provide an excellent opportunity to train students in a variety of disciplines and techniques that are of importance in modern science and technology. The chapters in Part II describe plasma physics experiments relevant to low-temperature and nonneutral plasmas, beams and radiation sources, and space and fusion plasmas. While many of these experimental studies have contributed significantly to our understanding of basic plasma science, they were often constrained by programmatic goals and by the plasma devices and plasma regimes relevant to a particular application. In this chapter, we focus specifically on what we have termed basic plasma experiments, whose primary goal is to isolate and study fundamental plasma phenomena in the simplest and most flexible situation possible. The objective of these experiments is to test our understanding of fundamental plasma phenomena, quantitatively and over the widest pos- 1   National Research Council, Plasmas and Fluids, in the series Physics Through the 1990s, National Academy Press, Washington, D.C., 1986.

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications sible range of relevant plasma parameters. Although these experiments are not intended to focus directly on any particular application, they can be expected to provide a quantitative understanding of the underlying physical principles and to have a potentially significant impact on an entire spectrum of applications ranging from plasma processing and fusion to astrophysics. Experiments on plasmas in the laboratory began in the 1830s with the work of Faraday to study the role of gas discharges in the chemical transformation of the elements. Further progress hinged on the discovery of the electron and the development of the atomic theory of matter at the end of the last century. In the 1920s, Irving Langmuir discovered the existence of collective oscillations in gas discharges. The understanding of plasma-related phenomena grew substantially with studies of electron beams in the 1940s and 1950s, in conjunction with the development of beam-type microwave devices. Since then, an enormous amount of work has been done in this area, and listing all of it is beyond the scope of this report. To convey the importance of a healthy and vital effort in basic experimental plasma science, we briefly review significant accomplishments in this area since 1980. We then proceed to discuss a number of important areas in which progress could be made in the next decade. These include topics that can be expected to have broad impact in virtually all of the areas of plasma science described elsewhere in this report. By the same token, basic experiments in specific topical areas are described in Part II. Examples include studies of electromagnetic wave-plasma interactions in the chapters on radiation sources and inertial confinement fusion and studies of fluid turbulence and transport in the chapter on nonneutral plasmas. OVERVIEW OF RECENT PROGRESS In this section, the panel presents a selection of areas and topics in which there has been significant progress recently both in experimental studies of fundamental plasma phenomena and in the development of new experimental capabilities. Basic Plasma Experiments Wave Phenomena Bernstein Waves. Bernstein waves are predominantly electrostatic waves that propagate in a magnetized plasma. These waves require a kinetic description, since the dispersion relation is dominated by the cross-field motion of the plasma particles and the wavelengths of these waves are comparable to the gyroradii of the particles. There are branches of the Bernstein wave dispersion relation associated with each of the harmonics of both the electron and the ion cyclotron frequencies. Unlike acoustic and electromagnetic waves, Bernstein waves have

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications no analogue in uncharged fluids, and they are therefore uniquely a plasma phenomenon. They are sensitive to kinetic effects and can be used as a diagnostic of plasma behavior as well as for plasma heating. An extensive body of knowledge has now been obtained from experiments that have used a variety of antennas and boundary conditions to elucidate the unusual properties of these modes. A wide variety of linear and nonlinear phenomena that involve Bernstein waves has been explored in the last decade, and they continue to be an important topic for basic research. Results from such studies have been used to interpret satellite observations of space plasmas. This knowledge has also been used to develop schemes for heating plasmas and for diagnosing plasma behavior. For example, there are potential applications using these waves to improve the stability and confinement properties of tokamak plasmas. However, an improved understanding of the nonlinear behavior of large-amplitude Bernstein waves will be required for such applications. Mode Conversion. Understanding mode conversion has been an important area of investigation in the last decade. In finite-temperature, spatially nonuniform plasmas, there can be degeneracy in the wave dispersion near plasma resonances, and mode conversion can occur near the spatial locations of these resonances. In particular, long-wavelength waves, which are often electromagnetic in character, can convert into electrostatic waves that then convect away the wave energy. Consequently, mode conversion can provide an important physical mechanism for absorption of the energy of electromagnetic waves. A variety of cases have now been studied, including the conversion of electromagnetic waves to Bernstein waves, Langmuir waves, lower and upper hybrid waves, and whistler waves. However, several important issues remain to be addressed. For example, although the linear transfer of energy has been observed, quantitative studies of the converted waves, the efficiency of energy transfer, and the associated electric field patterns have yet to be done, and theories of these phenomena have yet to be tested quantitatively. Understanding mode conversion is of great practical importance because of potential applications to plasma heating and use in plasma diagnostics. Wave-Particle Interactions Magnetically Trapped Particle Instabilities. The ubiquitous spatial nonuniformities of magnetic fields in laboratory and naturally occurring plasmas can cause the generation of two distinct populations of plasma particles: passing particles and mirror-trapped particles. Under very general conditions, the bounce motion of the trapped particles can result in the spontaneous amplification of various plasma modes. Recent experiments, based on an arrangement of multiple mirrors, have now elucidated the fundamental nature of these processes.

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications Lower Hybrid Wave Current Drive. Lower hybrid wave current drive, a fundamental Landau-damping process, describes the transfer of the momentum of traveling waves, which have been excited by an external source, to the momenta of the individual plasma particles. By choosing an appropriate wave, it is possible to induce a dc current in the plasma by trapping particles in the wave. The first experiments were done in a linear device. Subsequent toroidal experiments have investigated this interaction in detail by exploiting the unusual properties of lower hybrid waves. Efficient methods of current drive will be important in developing a steady-state fusion reactor. Beat Wave Excitation and Particle Acceleration. Basic laboratory experiments have demonstrated that when a plasma is irradiated by two electromagnetic waves whose frequency difference matches the local plasma frequency, very intense (GeV per centimeter) electric fields can be generated that travel at a significant fraction of the speed of light. Recently, it has been demonstrated in the laboratory that the controlled acceleration of a tenuous electron beam can result from its interaction with these plasma waves. Such studies suggest that compact particle accelerators based on this principle may be feasible. (See Figure 5.2.) Nonlinear Phenomena Double Layers. A fundamental nonlinear structure encountered in plasmas is the internal nonneutral sheath or double layer. A double layer can be thought of as the boundary between regions of plasmas having different particle distribution functions. An impressive body of experimental data has now been gathered from laboratory experiments on the shape, amplitude, and formation of these remarkable structures. These phenomena are important in space science. There are indications from satellite observations that double layers may form spontaneously in the near-earth plasma. The possible relationship between double layers and the formation of auroral beams is also being investigated. Ponderomotive Forces and the Filamentation of Electromagnetic Radiation. The ponderomotive force is one of the basic nonlinear affects governing plasma behavior. This force can be thought of as arising from the added plasma pressure produced by the oscillatory motion of charged particles in a strong electromagnetic field. When the amplitude of this field varies as a function of position, the spatial variation in this additional contribution to the pressure results in the ponderomotive force. Several experiments have elucidated the macroscopic nature of the ponderomotive force, the limits of fluid-like response, and the limitations set by the requirements for adiabatic behavior. A variety of experiments in magnetized plasmas have explored how to use the ponderomotive force to quench various configurational instabilities and thereby to produce quieter and longer-lived plasmas with improved particle and energy confinement.

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications Experiments have been conducted to study the propagation of a high-power laser beam through a plasma and the resulting plasma response. These experiments have demonstrated the filamentation of the primary beam into high-intensity beamlets, which trigger secondary plasma-wave instabilities and create associated beams of fast electrons. Magnetic Field Line Reconnection. The first magnetic field line reconnection experiments were done more than a decade ago in plasma pinch devices. Recently, a new generation of precise and well-controlled laboratory experiments has been carried out in which the ions are effectively unmagnetized but the electrons are magnetized. The magnetic field topology was mapped in three dimensions, and its dependence on plasma parameters was investigated. Observations include Alfvénic ion flow from the neutral sheet (i.e., a plane in the plasma at which the local magnetic field vanishes) and the formation of a neutral sheet on time scales less than the Alfvén transit time across the sheet. In the case where the current sheet was much narrower than its length, the breakup of the current sheet into a filamentary structure was observed. Other important nonlinear and three-dimensional phenomena were observed and studied, including the spontaneous generation of whistler-wave turbulence, the local formation of double layers, the generation of magnetic helicity, and the observation of highly non-Maxwellian particle distribution functions. Recent experiments have also studied magnetic reconnection in the merging process that occurs when two spheromak plasmas are brought together. (See Figure 8.1.) These plasmas are isolated structures, spheroidal in shape, that are self-sustained by a combination of currents and magnetic fields. In this case, there are local current sheets with magnetized ions. These experiments indicate that the merging process depends qualitatively on the initial helicities (i.e., the ''twists") of the magnetic fields of the plasmas involved in the merger process. Plasma Reorganization. Several experiments have been done in the past five years on the merging of plasma currents and the propagation of currents across magnetic fields. (See Figure 8.2.) One common feature of these experiments is that the current flows are fully three-dimensional. For example, merging currents in a high-beta plasma (i.e., a plasma in which the plasma pressure is comparable to that provided by the confining magnetic field) were observed to spiral about each other as they coalesced. The currents evolved to become nearly parallel to the local magnetic field and hence force free. An elegant experiment in which an electron current was made to propagate across a magnetic field showed that whistler waves played a key role in the evolution of the current channel. The experiments relied on highly reproducible, repetitive, plasma sources and on probes capable of studying the three-dimensional nature of the plasma behavior. These experiments are relevant to space plasma physics (such as the

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications tethered shuttle experiment), solar physics, the study of helicity generation and helicity injection, and the behavior of three-dimensional current systems. Chaos and Turbulence Chaos. Accurate description of the plasma dielectric response relies on integration of the perturbation caused by an applied field along the trajectory of plasma particles. The perturbed currents and densities thus obtained may then be put into Maxwell's equations to determine the wave dispersion. However, even in a uniform magnetized plasma, the application of a single, finite-amplitude plane wave can be sufficient to render the particle orbits chaotic, and no self-consistent theory exists for the plasma dielectric response in this case. Experiments have now determined that non-self-consistent chaos theory correctly predicts several aspects of wave-induced particle chaos, as long as the wave amplitude is sufficiently small. Conservation laws describing the particle orbits, even during chaotic particle motion, have also been identified. Chaotic heating of plasmas has been observed, not only from externally launched waves but also from spontaneous, unstable waves in a plasma that is externally driven. These experiments were made possible by laser-induced fluorescence techniques that have advanced dramatically in the last decade. Quasilinear Effects and Single-Wave Stochasticity. A series of experiments in single-component electron plasmas, which were carefully designed to eliminate the complications arising from ion dynamics, have tested the fundamental assumptions of "quasilinear theory," the standard model of weak plasma turbulence. These experiments demonstrated the importance of mode-coupling effects in modifying the wave-particle interactions described by the theory. In particular, in the presence of a mildly nonmonotonic particle distribution, unstable waves were found to grow and then saturate at the level predicted by the theory. However, the growth rates of individual waves were found to depend on the rates at which other waves grew, and this is not accounted for in the theory. Thus, a complete understanding of this important problem has yet to be achieved. This topic is related to the common assumption of the "random phase approximation" in turbulence theory, which is central to current descriptions of weak turbulence. The potential for new experiments in this area is discussed below in the context of turbulence and turbulent transport. Complementary experiments have observed the evolution of a large-amplitude monochromatic wave to a stochastic signal, via sideband generation and trapped-particle dynamics. A very important, but as yet unresolved, question is the detailed mechanism by which a single, large-amplitude wave is transformed into the background of weak turbulence that can be addressed by quasilinear theory.

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications FIGURE 8.1 Experimental study of magnetic reconnection processes in the merging of two spheromak plasmas. This experiment demonstrated that the three-dimensional structure of the magnetic field is crucial to the merger process in that the difference between the cohelicity and counter-helicity merger process is due to the relative directions of the out-ofplane components of the magnetic field in the two plasmas. A new mechanism of plasma acceleration (perpendicular to the plane of the figure) was discovered in the course of this work. (Reprinted, by permission, from M. Yamada, F.W. Perkins, A.K. MacAulay, Y. Ono, and M. Katsurai, Physics of Fluids B 3:2379, 1991. Copyright © 1991 by the American Institute of Physics.)

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications Collisionless Heat Transport. Laboratory experiments have now explored the important question of how heat is transported in collisionless plasmas. These measurements involve the application of high-power microwave beams to generate hot electron tails in a nonuniform plasma. The qualitative features of this effect and the important scaling properties have been identified. They have helped to clarify the relevant theoretical issues in this area. Strong Langmuir Turbulence. One of the significant advances in the understanding of nonlinear plasma behavior has been the development of the concept of plasma-wave collapse and the associated spiky turbulence that frequently accompanies it. Several laboratory experiments, aimed at uncovering the microscopic dynamics of Langmuir-wave collapse, have used both electromagnetic driving and electron beams to trigger the collapse, of extended wave packets, which in turn produces strongly localized fields and density depletions or cavitons. More recently, the ionosphere has been used to demonstrate the ubiquitous nature of these phenomena and the important role they play when a plasma is driven by large-amplitude perturbations. Experimental Techniques and Capabilities Opportunities for advances in experimental physics are often linked to the development of new technologies. The effect of these technologies is twofold. First, they enable the creation of experimental conditions that permit the demonstration and isolation of important physical effects. In plasma science, this frequently involves both new means of plasma production and new means of plasma confinement. In addition, new technologies frequently lead to new diagnostic techniques and new means of processing data, which not only result in improved accuracy and precision but often result in new perspectives on the underlying physics. Plasma Sources Over the past 20 years, there has been substantial progress in the development of improved, quiescent plasma sources. Some of the first, "high-quality" plasmas used in basic research were created in Q machines (Q stands for "quies-

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications FIGURE 8.2 Experimental study of the penetration of a pulsed current into a magnetized plasma. Shown are the characteristic field lines, sheets, and tubes of the current density, J(r), at different times after a 100-ns (FWHM) current pulse is applied to a disk electrode (shown). These data are extracted from a dataset of 10,000 point measurements at each time step. Typical experiments involve studying 1000 such time steps. At 80 ns, the current penetrates a short distance from the positive electrode into the plasma, before turning back to the negative electrode located at the back endwall of the vacuum chamber. Little helicity is observed in this fountain-like current flow. As the current propagates (120 ns) two distinct current systems are observable: a closed azimuthal Hall current in regions where Jz ÷ 0 and field-aligned solenoidal plasma currents between the positive and negative electrodes. At 150 ns a current tube starts off-axis, where Jz ≠ 0 and JB ≠ 0, and exhibits strong helicity; i.e., it twists and knots in the right-hand direction. After the end of the applied current pulse, at 200 ns, the current lines detach and propagate away from the electrodes, and shown is a closed, singly-knotted, twisted current tube. Experiments like this, which illustrate the fully three-dimensional nature of the dynamics of the plasma response resulting from such a current pulse, have recently been made possible by the advent of fast, relatively inexpensive laboratory computers with large data handling capabilities. (Courtesy of R. Stenzel and M. Urrutia, University of California, Los Angeles.)

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications cent" which were developed 30 years ago. These devices generate a magnetized plasma column, with a diameter of about 10–20 ion Larmor radii, that is well suited for the study of such phenomena as drift waves and ion cyclotron modes. The plasmas in Q machines are such that the electrons and ions have equal temperatures (i.e., Te = Ti). Consequently, these devices are not appropriate for the study of ion acoustic waves, which are strongly damped in such plasmas. The use of large numbers of small, permanent magnets to produce surface magnetic confinement, together with a variety of different electron sources, has provided a way to produce unmagnetized, collisionless plasmas that are both isotropic and quiescent. Such devices have Te/Ti ≈ 10, and they are well suited to the study of the linear and nonlinear behavior of ion acoustic waves. These plasma devices have also permitted experiments on plasma sheaths and on a variety of other linear and nonlinear waves. Combination of two or three of these plasmas has resulted in so-called double and triple plasma devices that have been used to study beam-plasma interactions, solitons, and ele ctrostatic shocks. In the past decade, dc discharges based on oxide-coated cathodes have resulted in the ability to produce large, quiescent, magnetized plasma columns, of the order of 50 cm in diameter (which is equivalent to 500 ion Larmor radii) and 10 m in length. Efficient, microwave-generated plasmas are now also conveniently available. Electron cyclotron resonance sources provide another way to study highly collisional plasma phenomena, with ion-neutral mean free paths of several centimeters or less. Inductive sources have recently shown considerable promise in producing uniform, unmagnetized and magnetized plasmas in the pressure range greater than 5 mtorr and, for example, have already been employed in studies of double layers. During the last few years, "helicon" sources (bounded whistler-wave sources) have produced steady-state plasmas with densities as high as 1014 cm-3. Such sources, which operate between the lower hybrid and the electron cyclotron frequency, do not have a high-density cutoff; they are therefore useful in producing plasmas with high densities. Plasmas consisting of negative and positive ions, with very low concentrations of electrons, have also been created, both with and without a magnetic field. The production of these plasmas relies on the large electron-attachment coefficient of gases such as SF6 for cold electrons. For sufficiently low values of the electron density, waves and instabilities in these plasmas can differ qualitatively from those in electron-ion plasmas, since the dominant charge species now have comparable masses. Mechanical Probes Refinement of probe techniques has occurred hand in hand with plasma source development. These include directional velocity analyzers (with resolu-

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications development of compact, subpicosecond terawatt lasers with beams that can be focused to intensities greater than 1018 W/cm2. At these intensities, an electron is accelerated to relativistic energies in one period of the laser light. This permits the study of highly nonlinear, fast, relativistic processes in laser-plasma interactions. If the laser is focused on an overdense plasma target, dc magnetic fields of the order of 109 G are predicted to occur. It will be important to determine whether the nonlinear ponderomotive forces and relativistic effects can reduce the diffraction of these ultrahigh-intensity laser beams, so that the light can be focused to beam sizes smaller than a few Rayleigh lengths, the limit expected at lower values of light intensity and in a linear medium. Chaos, Turbulence, and Localized Structures Nonlinear Particle Dynamics and Chaos. Modern concepts of nonlinear dynamics have created a renaissance in classical physics, bringing new techniques to bear on long-standing problems. One crucial issue in plasma physics is the onset of chaotic particle motion in response to coherent or turbulent wave fields. Of particular interest are a self-consistent description of the system under such circumstances and the evolution of the system from regular particle motion to chaos. It is now possible to conduct precisely controlled experiments in the laboratory to address these important problems, which are of interest in a wide variety of contexts, ranging from fluid dynamics to advanced particle accelerators. Nonlinear Wave Phenomena. With the exception of Alfvén waves, most of the other linear branches of the plasma dispersion relation have been explored. In addition, many nonlinear, three-wave coupling processes have been observed. However, the transition from linear to turbulent wave behavior is not understood. This includes the nonlinear behavior associated with almost every branch of the plasma dispersion relation. Turbulence. Very generally, plasmas are electrodynamic, many-body systems far from equilibrium that are dominated by nonlinear effects. Consequently, plasmas are typically highly turbulent, exhibiting large fluctuations in such quantities as the local density, temperature, and magnetic field, which can vary rapidly in time and space. Important examples of plasmas whose behavior is influenced profoundly by turbulence include essentially all magnetically confined fusion plasmas and many astrophysical and space plasmas. We have no first-principles understanding of turbulence in any plasma, and understanding such turbulent behavior is perhaps the key unsolved problem in plasma physics. This problem presents an important synergism with fluid dynamics, in that plasmas can often be modeled as fluids and understanding turbulence is central to a complete description of fluid systems.

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications Since turbulence is so common in plasma physics, the potential rewards for achieving predictability are particularly high. In the past decade, new plasma sources and measurement techniques have been developed that will allow us to undertake a new generation of precisely controlled experiments to study turbulent plasma behavior. One starting point in achieving a deeper understanding of turbulence will be further study of the questions raised by the observed breakdown of quasilinear theory and experimental tests to determine the range of validity of the random phase approximation. Turbulent Transport. Profound consequences of plasma turbulence include the transport of both particles and energy and the acceleration of particles that can be induced by turbulence. Such transport can completely dominate plasma behavior. For example, transport by turbulence, in the form of both convection and enhanced diffusion, is the dominant transport mechanism of particles and energy in present-day tokamak fusion plasmas. Turbulent transport presents an excellent opportunity for carefully controlled laboratory experiments. At least in low-temperature laboratory plasmas, techniques are now available to study both the transport of particles and energy and the fluctuations responsible for this transport. To establish the causal connection between turbulence and transport, it will be necessary to make precise, spatially resolved measurements of fluctuating quantities such as plasma temperature, density, velocity, and magnetic field and to establish the correlations between these quantities and local measurements of the particle and energy fluxes. Because turbulence and turbulent transport are not understood in any plasma, careful experimentation in flexible, small experiments is likely to make significant contributions to testing existing theoretical predictions and to guide further theoretical work in this important area. Given the fundamental lack of understanding and the important practical consequences that would derive from a deeper understanding of turbulence and turbulent transport, a sustained program of both theoretical and experimental research is extremely important. Sheaths, Boundary Layers, and Double Layers. Plasma sheaths (i.e., regions where the plasma is not charge-neutral) have been an important topic throughout the history of plasma physics. All plasmas in the laboratory and in space have boundaries at which there are sheaths, and probes and antennas immersed in plasmas are surrounded by such sheaths. One important area for future research is the nature of sheaths in magnetized plasmas. To probe the structure of such sheaths requires detectors smaller than an electron Debye length. Such probes and probe arrays can be expected to be available in the next few years. Double layers are a class of sheaths that are detached from a physical boundary and are supported by locally non-Maxwellian conditions. Although there has been some research done on double layers, there has been little work on situations in which the ions are magnetized and situations involving the transition

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications regime between collisionless and collisional plasmas. This is of importance for solar physics (in regard to coronal holes) and in ionospheric heating experiments. Double layers are responsible for localized particle acceleration and could also play an important role in the aurora. Shock Waves. Much laboratory work has been done on shocks in unmagnetized plasmas. Shocks also have been studied in pinches and exploding wires. However, careful experiments on Alfvénic shock waves in magnetized plasmas have yet to be done. Of particular interest is the propagation of large-amplitude (i.e., δB ≈ B) magnetic pulses. Work would include studies of wave steepening, particle reflection and heating, and a search for a new class of shocks (the ''intermediate shock") that has been predicted but not yet observed. Such shock wave phenomena are of importance in space and astrophysical plasmas. Striated Plasmas. Plasmas with nonuniformities, such as density or temperature striations in the direction parallel to the magnetic field, are of fundamental interest. They occur, for example, in the ionosphere and in the aurora. If the gradient in the plasma properties is steep compared to the wavelengths of interest, these structures can trigger the mode conversion of whistler waves. The reflection, refraction, and interaction of waves with plasma structures that have steep gradients has not been studied in the laboratory and presents a difficult "plasma scattering" problem. Striated plasmas are not limited to those with density perturbations but also include local "hot spots" and magnetic field perturbations. Topics of interest include the interaction with lower hybrid waves, refraction and reflection, fast-particle generation, and minority-species heating. Flows in Magnetized Plasmas. It now is possible to generate highly magnetized laboratory plasmas in which the diameter of the plasma column is much larger than the ion Larmor radius (e.g., by factors of as much as 103) and in which magnetic Reynolds numbers of 104 to 105 are attainable. (The magnetic Reynolds number is the time scale for transport of the magnetic field by the flowing plasma relative to the time scale for diffusion of the field due to the finite resistivity of the plasma.) By using multiple sources (so called "double-plasma" configurations), flowing plasmas can be generated with drift velocities comparable to the Alfvén wave velocity and with Mach numbers (i.e., plasma flow velocities relative to the ion sound speed) of the order of 500. Large currents can be entrained in these plasmas. Such situations are predicted to lead to shocks and turbulence. Insights into dynamo action (discussed below) are also likely to be achieved in such experiments. These experiments also would be relevant to the solar wind and to other solar and astrophysical processes. Plasmoids. "Plasmoids" are plasma structures that propagate as recognizable entities through a background plasma. Satellite data suggest that plasmoids may

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications occur in the Earth's magnetotail and become detached and move away from the Sun during magnetic storms. The propagation of plasmoids has been studied with small plasma guns. Larger structures of interest to fusion physics (i.e., spheromak plasmas) have also been investigated. (See Figure 8.1.) The latest generation of diagnostics, coupled with the recently developed ability to generate plasmoids easily, now permits a new generation of experiments. For example, one now can study in detail how plasmoids are generated, how they propagate, and the details of their internal field structure and associated plasma currents. Magnetic Effects Magnetic Field Line Reconnection. Magnetic field line reconnection is one of the principal means by which magnetic field energy is converted to thermal energy and plasma motion. For example, it is thought to be responsible for the high temperature of the solar corona and to be important in the Earth's magnetotail and in many astrophysical situations. Magnetic reconnection also is of importance in understanding the so-called sawtooth crashes that occur in the hot core of tokamak plasmas when a certain type of magnetohydrodynamic instability is present. In this case, reconnection has the effect of expelling hot plasma from near the plasma center and is detrimental to plasma confinement. Experiments on magnetic reconnection in plasmas with unmagnetized ions and magnetized electrons have already been done. Areas for further study include cases where the magnetic Reynolds number is greater than 100. Important issues include the three-dimensional nature of this phenomenon, the connection between global and local time scales, the acceleration and heating of the plasma particles, and the generation of plasma flows. Dynamo Action. The dynamo is a process by which the kinetic energy of a conducting fluid is transformed into magnetic field energy. (See Figure 8.4.) In a dynamo, a "seed" magnetic field from a small current fluctuation can be stretched and reconnected by the turbulent fluid motion. In principle, this can lead to amplification of the magnetic field to a level where the magnetic field dominates the dynamics of the fluid flow. The dynamo is a fundamental process in magnetohydrodynamics, and dynamo action is crucial to understanding many aspects of space physics and astrophysics. For example, it is believed to be the origin of the magnetic fields of such diverse objects as the Sun and the accretion disks of stars and is intimately connected with the physics of novae and supernovae. The conditions for dynamo action require large-scale flows in highly conducting media, and up until now, such conditions have proven difficult to achieve in the laboratory. The criterion for dynamo action is the achievement of Reynolds numbers on the order of 100. Possibilities now exist to carry out well controlled

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications FIGURE 8.4 Demonstration of the dynamo effect in a laboratory plasma. The dynamo effect is the spontaneous self-generation of magnetic fields within plasmas, a process common in astrophysical bodies such as the Sun. Shown is a "dynamo event" in which magnetic flux is suddenly generated in a toroidal laboratory plasma. By measuring the local electromotive force, ⟨ v × B ⟩/c, generated by fluctuations in plasma velocity and magnetic field, it was established that this force is responsible for generation of the magnetic flux. The fluctuating flow and field were measured with Langmuir and magnetic probes inserted into the plasma edge. This particular dynamo mechanism has long been thought to be an important source of astrophysical magnetic fields. (Reprinted, by permission, from H. Ji, A.F. Almagri, S.C. Prager, and J. S. Sarff, Physical Review Letters 73:668, 1994. Copyright © 1994 by the American Physical Society.) dynamo experiments, for example by establishing a rapid fluid flow in a large vat of liquid sodium. Such experiments can provide new and fundamental insights into the nature of dynamo action and provide a quantitative basis for refined theories of many important physical processes. Magnetic Reconfiguration. Many plasma instabilities and processes are inherently three-dimensional. Laboratory experiments now allow these processes to be explored in detail. Important topics include the dynamics of three-dimensional current systems, current and wave filamentation, current sheet formation, the effect of magnetic forces on plasma currents, helicity generation, and helicity conservation. These processes have far-reaching consequences in the understanding of solar flares, solar magnetospheric physics, and fusion physics.

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications New Experimental Capabilities In any experimental science, and particularly in physics, advances in diagnostics consistently have led to new discoveries and frequently have opened up entirely new areas of research. Several new tools for plasma research are now becoming available. Some of these techniques have not been developed with plasma diagnostics in mind, but they can be expected to have significant impact on experimental plasma science. In many cases, progress is likely to require the collaboration of plasma physicists, solid-state physicists, and engineers. Much of this work will have important applications in other fields. Use of Nanotechnology Advances in nanotechnology are likely to have a profound impact on experimental plasma physics. Typical devices are miniature valves and mass analyzers. Techniques widely used in the semiconductor industry will enable the production of particle detectors, mass-sensitive energy analyzers, and magnetic and electric field probes with overall scale sizes smaller than 1 mm and active sensor areas less than 10 mm in diameter. These detectors will be capable of providing spatially resolved measurements on the scale of the Debye length and electron cyclotron radius in research plasmas with densities of the order of 1012 cm-3, electron temperatures of tens of electron volts and magnetic fields of 0.1 T. Such probes would produce a minimal perturbation of the plasma if their connections and supports were also microscopic (&2248;0.2 mm). It is likely to be possible to position many (e.g., 104 to 106) of these detectors on a lattice that could be moved within the plasma. Optical Diagnostics The recent discovery of giant Faraday rotation in magnetoactive crystals now enables the construction of magnetic field probes as small as 10 µm in diameter and less than 1 mm long. As a light beam traverses one of these small crystals, the plane of polarization of the light is rotated. Preliminary tests have demonstrated sensitivities of 1 G per degree of angular rotation and response times faster than 10-9 s. Such probes are immune to electrical pickup and are nonmetallic. Another important capability has been created by the discovery of the quantum-well effect in crystals. Quantum-well devices can now be fabricated into microscopic probes to measure the local amplitude of the electric field. Arrays of these optical probes could be used to diagnose the space-time behavior of the electric fields associated with plasma waves and currents.

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications New Regimes of Plasma Parameters As described in Part II, advances in laser technology now make possible laboratory experiments in previously inaccessible regimes of plasma parameters. Both short-pulse, high-power lasers and multiphoton ionization using tuned sources can be used to produce liquid or solid density plasmas, in which both quantum and classical many-body effects are important. The creation of these high-energy-density plasmas also opens up the possibility of studying plasmas with highly ionized ions (i.e., high-Z plasmas). Data Acquisition In the past 10 years, experimental physics has benefited greatly from advances in digital technology. Analog-to-digital converters and microprocessors have decreased drastically in price. Workstations are now available with 128 Mbyte of memory and 8 Gbyte of disk storage, and this trend shows no sign of saturating. A system with 106 channels of acquisition is capable of acquiring on the order of 1 Gbyte of data per second. Such a data acquisition system might consist of many parallel processors sharing a fast network and have 10 Gbyte of random access memory and 10 to 100 Tbyte of mass storage. This system would permit the study of nonuniform and fully three-dimensional plasma phenomena and plasma processes occurring on more than one spatial scale with unprecedented spatial and temporal resolution. Massively Parallel Plasma Diagnostics Interactions among plasma particles range from short-range collisions between individual particles to long-range, collective forces; consequently, plasmas frequently contain several different characteristic length scales. Magnetized plasmas are inherently anisotropic and nonlinear, exhibiting nonlocal behavior, chaotic particle motions, turbulence, and self-organization. The fundamental equation describing the plasma behavior of a many-body system of N charged particles is Liouville's equation for the distribution function in the 6N-dimensional phase space of the system. As a practical matter, theoretical descriptions of plasmas are frequently based on much simpler and more tractable equations. However, the assumptions concerning the statistical structure of the plasma, which buttress the derivations of these simpler treatments, have not been tested. In the next decade, a new generation of plasma experiments is likely to be able to make significant contributions in testing the validity of the approximations used to describe plasmas, for example, as fluids or as many-body systems described by kinetic theory or by a particular kind of particle correlation function. With what is now or will shortly become available, experimental plasma science will be able to explore a range of plasma problems with a precision and

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications to a degree of detail previously unattainable. Below, we briefly list some possibilities, assuming the capability exists to create a lattice of thousands of microscopic detectors and/or thousands of channels of optical probes. In plasma physics, the plasma is typically described by a combination of time-averaged and fluctuating fields. In the case of a turbulent plasma, the average of a quantity may be much smaller than its fluctuating component. If one considers an experiment that is repeated many times, all individual quantities, such as the instantaneous values of the electric and magnetic fields, the density, and the particle distribution function, can in principle be measured and recorded. This detailed set of measurements could be used to calculate the higher moments of the distribution function to test the assumptions that go into the derivations of the equations of kinetic theory. For example, our present understanding of three-body correlations is poor, but such quantities could be measured directly. Fine structure in the particle distribution functions could also be measured. Measurements by particle detectors on spacecraft and in the laboratory indicate that when instabilities are present, the distribution functions cannot be regarded simply as functions of the magnitudes of the components of velocity perpendicular and parallel to the magnetic field. Snapshots of the distribution functions could be expected to reveal phase-space structures that go far beyond such a simplified description. Such highly anisotropic particle distribution functions can be expected to have profound effects on the growth and damping of a variety of plasma waves. With such detailed measurements, one could also test the validity of equating temporal averages with spatial ones. The ability to probe fine spatial scales will permit a detailed exploration of plasma sheaths and boundary layers. For example, tiny puff valves and micro-beam sources could be used to tailor the local particle distribution function or to add minute quantities of an impurity ion. Finally, phenomena on scale lengths ranging from less than the Debye length, to the ion cyclotron radius, to the electron cyclotron radius, could be explored simultaneously in one experiment. This would allow exploration of physics from the regime in which single-particle interactions are important to the regime in which kinetic and MHD effects are dominant. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS The panel has great concern that basic experimental plasma science is disappearing in the United States. By its count, there are currently fewer than 20 groups engaged in basic plasma experiments in the United States. Yet an intellectual atmosphere that allows for dialogue, complementary experiments, and in some cases, competition is necessary for any field of modern science to make efficient progress. There is tremendous benefit to be derived if different research groups working on similar and complementary problems can exchange

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications ideas and collaborate. Investigators free to follow where their research leads produce qualitatively new insights and new approaches to the underlying science. This type of scientific environment typically produces new techniques and new ideas on a rapid time scale. Such basic science is the foundation for applied science. These processes do not happen as frequently at large "technology centers," which must operate in a less flexible and more programmatic fashion. In plasma science, the place for new and significant discovery is very frequently the laboratory. When there are significant new theoretical predictions, much of the value of these predictions is lost if they cannot be tested quantitatively by experiment. There is no adequate substitute for carefully planned and precisely controlled laboratory experiments. Many interesting and stimulating observations of plasmas can be made by spacecraft, but space experiments are not a substitute for the well-controlled and repeatable experiments that can be performed in the laboratory. The notion that computers can simulate plasmas so well that laboratory experiment can be replaced is also incorrect and is likely to remain so for the foreseeable future. Laboratory experiments, theory and modeling, spacecraft and astrophysical observations, active space experiments, and experiments on fusion plasmas are synergistic. It is the healthy interplay among all these elements that will lead to a healthy plasma science. The field can "get along" for a while ignoring one element or the other, but it cannot continue for long in the unbalanced manner that has occurred in the last decade in the case of basic laboratory plasma experiments. Without a healthy underpinning of experimental laboratory work, the field of plasma science will not attract talented young scientists and is destined to become sterile and inefficient. As a consequence, the highest priority of the panel is the establishment of a system of sustained support for modest-sized experimental efforts, sufficiently small and flexible that they can make rapid changes in their approach to a research problem, guided by the internal logic of the science and by new experimental and theoretical discoveries as they develop. As discussed in Chapter 4, fundamental aspects of plasma science crucial to fusion physics must be pursued in detail in the fusion-relevant geometries provided by large plasma devices and facilities. However, given the relatively high costs of operation of large facilities and the limited funds that one can expect for fundamental plasma experiments, the number of large devices not motivated by important applications such as space science or fusion is likely to remain small for the foreseeable future. It is the conclusion of the panel that the type of sponsorship necessary for a revitalization of basic plasma experimental science is the support of at least 30 to 40 independent groups at a reasonable level for experimental research in a university, which is of the order of $200,000 to $400,000 per year. For example, funding at the $200,000 level would allow a program of two students, a postdoctoral researcher, and modest expenditures on equipment and supplies. Larger programs would require some technical support and additional personnel

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications and equipment, as appropriate. It is important that such support be granted for periods of at least three years, given satisfactory performance, so that significant research goals can be set and accomplished. Large equipment purchases would have to be funded from separate equipment proposals. Given that the infrastructure for basic experimental facilities has declined so significantly in the past two decades, additional initial equipment purchases, where necessary, would typically range from $300,000 to $600,000 per program. Additional mechanisms that would allow for collaboration between groups on a rapid time scale, compared to the proposal cycle that now exists, would also be beneficial. Placing some resources at the discretion of program managers would be one way to accomplish this. The increased support that the panel recommends for basic experimental research can be expected to serve an important educational function as well. It is generally recognized that small-scale experiments are an excellent setting in which to train students. The training of students, under the guidance of their supervisor, to make qualitative changes in an experiment or even in research direction as results unfold is invaluable in modern research and technological development. In addition, experimental plasma science students typically receive very thorough training in such important areas of modern technology as digital electronics, optics and computational hardware and software. The following of the panel's general recommendations (see Executive Summary) are made to implement the revitalization of experimental plasma science described above: To reinvigorate basic plasma science in the most efficient and cost-effective way, emphasis should be placed on university-scale research programs. To ensure the continued availability of the basic knowledge that is needed for the development of applications, the National Science Foundation should provide increased support for basic plasma science. To aid the development of fusion and other energy-related programs now supported by the Department of Energy, the Office of Basic Energy Sciences, with the cooperation of the Office of Fusion Energy, should provide increased support for basic experimental plasma science. Such emphasis would leverage the DOE's present investment in plasma science and would strengthen investigations in other energy-related areas of plasma science and technology. Approximately $15 million per year for university-scale experiments should be provided, and continued in future years, to effectively redress the current lack of support for fundamental plasma science, which is a central concern of this report. Furthermore, individual-investigator and small-group research, including theory and modeling as well as experiments, needs special help, and small amounts of funding could be life-saving. Funding for these activities should come from existing programs that depend on plasma science. A reassessment of the relative allocation of funds between larger, focused research

OCR for page 131
Plasma Science: From Fundamental Research to Technological Applications programs and individual-investigator and small-group activities should be undertaken. The panel recommends that the National Science Foundation increase its support for individual principal investigators conducting university-scale programs in basic research because this is most closely associated with NSF's mission. Increased support for basic research by the Department of Energy is recommended because DOE is charged with responsibility for both the magnetic and inertial confinement fusion programs, as well as a number of other energy-relevant programs that are critically dependent on the fundamental principles of modern plasma science.