Click for next page ( 127


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 126
7 Scientific Interfaces The boundaries of atomic, molecular, and optical (AMO) physics penetrate far into the neighboring areas of science. Across its borders flows a stream of new techniques and vital data. There is hardly an area of science that has not significantly benefited from these. Geology, geophysics, and planetary physics, for instance, have all been enriched by the maps created with optically pumped magnetometers. Laser surveying, to cite another example, permits monitoring the strains that lead to earthquakes as well as the motion of the continents as they drift and of the moon as it wobbles in its orbit. It also makes possible the high-precision alignment of large particle accelerators and allows tunnels drilled from opposite sides of a mountain to meet exactly. There is a second form of commerce between AMO physics and the neighboring areas: this is the commerce of basic science itself. In this the boundaries disappear as the unity of science asserts its pre- eminence. Among the six interface areas that we have chosen to describe here astrophysics; materials research; surface science; and plasma, atmospheric, and nuclear physics instances of the underlying unity constantly occur. ASTROPHYSICS Most of what we know about the universe comes from information brought to us by photons. To decipher their messages, we must 126

OCR for page 126
SCIENTIFIC INTERFACES 127 understand how the photons came into existence and the histories of their journeys through intergalactic and interstellar space. From this enterprise we can learn about the early universe and the nature of the astrophysical entities quasars, galaxies, stars, pulsars, stellar winds, supernova remnants, nebulae, masers, and molecular clouds. The events that produce photons and the processes that modify them during their long journeys lie squarely in the domain of AMO physics. AMO physics is an essential component of astronomy. AMO physics ranges broadly in its applications to astronomy. The physical and chemical processes that created molecular hydrogen in the early pregalactic universe, which manufactured cyano-octa- tetrayne in interstellar clouds and propane in the atmosphere of Titan, which bring molecular clouds to the brink of gravitational collapse and trigger star formation, which control the abundance of ozone on the planet Earth, and which determine the radiative losses from stellar interiors, are all part of the body of AMO physics. The unity of AMO physics is manifested in the remarkable diversity of environments in which atomic processes play the crucial role. Astronomy and AMO physics are mutually dependent. Because the extraordinary physical environments that arise in astronomical phe- nomena are often impossible to duplicate in laboratories, astronomy provides an extended arena for studying atomic processes. For exam- ple, despite the enormous differences in the scales of laboratory and astrophysical plasmas, the fundamental processes are identical. The same mechanism that is believed to be responsible for the lack of electrons at high altitudes in the atmosphere of Jupiter, as revealed by data from the Voyager spacecraft, has been proposed for producing the negative hydrogen ions that are needed to ignite a thermonuclear fusion plasma. The mechanism involves vibrationally excited molecular hy- drogen. These molecules are ubiquitous. In the interstellar medium their Doppler-shifted lines signify the shock waves that accompany the birth of stars. Provided that the molecular processes are clearly understood, these lines can provide powerful diagnostic probes of the earliest stages in the evolution of a star. Interpreting the abundant data of astrophysics demands a deep understanding of atomic, molecular, and optical processes. In addition, it demands a broad data base of atomic and molecular parameters such as transition energies, oscillator strengths, and photon and particle collision cross sections. Providing these data is a major challenge for atomic and molecular physics. Experimental data flow from all branches of the field, particularly from the discipline that has come to be called laboratory astrophysics. These experimental data are vital,

OCR for page 126
128 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS but more data are required than the experimental community can possibly provide. Thus, theoretical data are also vital. The need to generate theoretical data for astrophysics motivates a major portion of the theoretical effort in the atomic and molecular community. The required data base is huge. Several million emission lines are present in the spectra of the Sun and stars. Models of stellar atmo- spheres have limited success, even in the visible region. Important questions are unresolved: the discrepancy between the predicted and measured solar neutrino fluxes may be due in part to errors in the calculated opacity of the Sun. Stellar explosions, which play a crucial role in the energetics of the galaxy and the formation of new stars, provide another example. These explosions leave a remnant in which elements such as oxygen, sulfur, silicon, iron, and nickel are stripped of all but one or two electrons. The emission lines of these ions, which fall in the x-ray region, can yield not only the element abundances but also the density and temperature of the remnant. Because the relevant atomic data are not available, a comprehensive description of the atomic processes occurring in a supernova remnant has yet to be achieved. Atomic Processes Many atomic processes play crucial roles in astrophysics. To cite one example, in recent years the importance of atomic charge transfer in cosmic plasmas has become evident. In interstellar gases composed of elements in their cosmic abundances, ionizing radiation often produces partly ionized plasmas containing some neutral atomic hy- orogen. Charge-transfer collisions with the hydrogen drastically mod- ify the ionization structure of the gas. The emission spectra offer a highly specific diagnostic probe of the plasma. Provided the charge- transfer processes are understood, the spectra can serve to establish the coexistence of multiply ionized and neutral material, provide a direct measure of the neutral abundances, and give unique information on the nature of the ionizing source. Rydberg Atoms Ionized gases emit photons at all wavelengths. At radio wavelengths, the photons arise from transitions between highly excited Rydberg levels with principal quantum numbers that can exceed 300. Rydberg atoms are large in size and sensitive to disturbances, but the space between stars is nearly empty, and there is room for the Rydberg atoms

OCR for page 126
SCIENTIFIC INTERFACES 129 to survive until they radiate. Because atomic theory can provide a detailed description of the modes for populating and depopulating Rydberg levels, these atoms can be a valuable guide to events occurring in our own galaxy and in external galaxies. They are used to infer the temperature and densities and the hydrogen-to-helium abun- dance ratio. Rydberg atoms can now be generated in the laboratory, and their study has developed into a lively subfield of atomic physics. (This is discussed in Chapter 4 in the section on Atomic Structure and in Chapter 6 in the section on Quantum Optics and Coherence.) The prominence that Rydberg atoms assumed in the laboratory in the mid-1970s was stimulated by their discovery in space in the 1960s. Interstellar Molecules Transitions between low-lying levels in molecules generate radiation at radio frequencies. Because the photons suffer little attenuation by interstellar dust, the radio emission lines can be seen over large distances. More than 50 interstellar molecules have been discovered. To cite one consequence, the distribution of matter throughout the galaxy has been mapped from the emission lines of carbon monoxide. Radiation from interstellar molecules can extract energy from the interstellar clouds, cooling them to the brink of gravitational collapse. Two of the most fundamental astrophysical processes, nucleosynthesis and the chemical evolution of the galaxy, can be studied by observing the spatial distribution of isotopic molecules such as '3C~6O and ~2C~8O, though the task requires the mastery of the basic molecular chemistry. In order to determine reliable isotope ratios, for example, molecular fractionation must be understood. Molecular fractionation substantially enhances the abundances of deuterated molecules- molecules in which a hydrogen atom is re- placed by a deuterium atom. By joining the theory of ion-molecule chemistry in interstellar clouds with observations of the abundance ratio of the deuterated compounds, the electron density in molecular clouds can be inferred. This density is a critical astrophysical param- eter. Gravitational collapse and the fragmentation of molecular clouds to form stars are mediated by free electrons. Despite the deep significance of molecular fractionation, however, one of the essential molecular parameters of fractionation theory remains unknown. As a result, no more than an upper limit can be obtained for the electron density.

OCR for page 126
130 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS Many interstellar molecules are chemically reactive. In the labora- tory they exist only as short-lived transient species, difficult to study at high resolution. In some cases butadinyl and cyanoethynyl are ex- amples- the spectrum can be more accurately measured in space than in the laboratory. In other cases, thanks to recent developments in laboratory techniques and laser technology, laboratory measurements are now superior. Thus, the fine-structure parameters of the reactive neutral atom, carbon-12, were first determined in the laboratory. The results pointed the way to the successful detection of atomic carbon in dense interstellar clouds. A great many other interstellar species with different isotopic constituents await investigation. Astrophysical Chemistry Molecular ions occur at crucial points in the ion-molecular schemes that attempt to explain the formation of interstellar molecules. The measured ion abundances provide a sensitive test of the chemical models. Few of the reaction-rate data are available, still fewer at the temperatures prevailing in molecular clouds. The most important reaction pathways may not yet be recognized; the success of the chemical schemes may be no more than an artifact of unreliable data. The very first laboratory experiments on molecular reactions at low temperatures were carried out recently; these may lead to a quantita- tive description of molecular formation in cold clouds. Molecules have now been detected not only in interstellar clouds but also in the hostile environments of stellar atmospheres and circumstellar shells. It seems likely that they also exist in other astrophysical regimes such as quasars and that they may someday be useful in detecting of x-ray sources and supernovas buried inside dense clouds. Cosmology Atomic and molecular processes can provide vital clues to the nature of the cosmos. For example, the distribution of deuterium in the galaxy provides a direct measure of the matter density in the early universe and bears directly on the question of whether the universe is closed or open. The deuterium is detected as a constituent of different molecular species; the chemistry of deuterated molecules must be understood before the total deuterium content can be obtained. The spectrum of the cyanogen molecule also has direct cosmological significance. From its optical absorption spectrum the relative popula- tions of the two lowest energy levels can be determined, and from this,

OCR for page 126
SCIENTIFIC INTERFACES 131 its temperature. The temperature was found to be 2.8 K. This was the first measurement of the temperature of the universal blackbody background radiation left over from the big bang. Cosmology can provide unique insights into fundamental atomic principles. From absorption-line measurements toward distant objects at large red shifts, for example, limits can be set on how the funda- mental atomic constants can vary in space and time. One result is that the fine-structure constant cannot vary by more than 1 part in 10-~2 per year. Space Physics Astronomy is primarily driven by remote observations, but one component, space physics the study of the local solar system is advanced by local experiments with instruments carried aboard space- craft. The interplanetary medium undergoes violent upheaval where the solar wind collides with the ionized gas in the outer regions of the planets and their satellites, providing a natural laboratory for studying the effects of electric and magnetic fields on the large-scale motions of energetic charged particles. Atomic and molecular physics is essential to understanding the scene. Charged particles are created, scattered, and lost by atomic collisions. Planetary atmospheres respond to solar ionizing and dissociating radiation in a complex array of atomic and molecular processes. The evolutionary paths followed by these atmo- spheres are affected by escape mechanisms driven by energy transfer in atomic and molecular collisions. The interpretation can be subtle and can lead to unexpected conclusions. For example, the Viking lander on Mars measured a '5N/~4N isotope ratio 60 percent larger than the terrestrial value, suggesting the operation of a differential escape mechanism for the two isotopes. On Mars, the process of dissociative recombination of ions of molecular nitrogen generates nitrogen atoms with kinetic energies sufficient to escape the gravitational field of the planet. As the isotopes undergo gravitational separation in the atmo- sphere, the heavier isotope becomes depleted at the high altitudes where escape occurs. A careful accounting of the escape efficiency establishes that Mars once contained a large reservoir of nitrogen gas. Similar mechanisms occur on Venus with a startling corollary. When molecular-oxygen ions recombine on Venus, they produce energetic oxygen atoms that collide with hydrogen atoms and drive the hydrogen out of the atmosphere. The collisions are too weak to drive out the heavier deuterium atoms. As a result, the deuterium/hydrogen ratio on Venus is much larger than anywhere else in the solar system. From the

OCR for page 126
132 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS ratio measured by the Pioneer Venus space probe, one can infer that Venus originally had a large abundance of water. CONDENSED-MATTER PHYSICS AND MATERIALS SCIENCE Numerous links join AMO physics with condensed-matter physics and materials science. The study described in Chapters 4 and 5 of how x-ray and photoionization spectra of gaseous atoms and molecules evolve as they assemble into liquids and solids reflects one aspect of this interface area. AMO physics has generated experimental tech- niques ranging from molecular-beam epitaxy and clusters to laser annealing and sputtering. The impact of AMO physics on surface science one of the liveliest areas in solid-state physics is so large that it is described separately in the next section. In this section, we describe three activities: light-scattering spectroscopy, metal clusters, and the creation of spin-polarized quantum fluids. Light-Scattering Spectroscopy The extraordinary spectral purity of gas lasers has made them an important source of radiation for the observation of thermally excited fluctuations and of fluid flow in condensed-matter systems. The inter- action of the laser radiation with spontaneous molecular motion produces spectral broadening or frequency shifts in the scattered light that generally range from 1 to 105 Hz. (The frequency of visible light is about 10~5 Hz.) The accurate resolution of such small-frequency shifts has become possible using the techniques of optical mixing spectros- copy. These techniques represent the successful extension of hetero- dyne and homodyne detection methods, long employed in radio- frequency and microwave spectroscopy, upward into the optical frequency domain. As a result of these advances, a new form of spectroscopy, known variously as quasi-elastic light-scattering spec- troscopy, photon-correlation spectroscopy, or intensity-fluctuation spectroscopy, has emerged and been applied to a wide range of fundamental and applied problems in physics, chemistry, biology, . . . . engineering, ant met lclne. Order-Disorder Transitions: In physics, light-scattering spectros- copy has provided many of the basic determinations of the critical exponents for the divergences of the equilibrium and transport coeffi-

OCR for page 126
SCIENTIFI C I NTERFA CES 1 3 3 cients of pure fluids and binary mixtures near their order-disorder phase transitions. The profound nature of the theoretical ideas needed for the exploration of these and related experiments in magnetic systems culminated in the creation of the renormalization group theory, one of the major achievements of modern condensed-matter theory. Quasi-elastic light scattering has been the principal experimen- tal tool used to investigate the hydrodynamic modes and phase transitions of liquid-crystal systems, a field of high current interest. Light-scattering spectroscopy has been used to study the relaxation of density fluctuations in gases, the propagation of elementary excitations in liquid helium, and the development of soft modes in the phonon spectrum in solids near phase transitions. It is a principal experimental means of investigating the transition to turbulence (or chaos) in hydrodynamics and has been used to determine the value of important universal numbers in the theory of strange attractors. Applications to Chemistry, Biology, Engineering, and Medi- cine: In chemistry, the method is widely used to obtain important microscopic information on the fundamental interactions between amphiphillic molecules, which self-assemble to produce well-defined geometrical structures: micelles, microemulsions, vesicles, and bilay- ers. These structures are fundamental constituents of the living cell and are of great importance in a wide variety of industrial chemical processes. Quasi-elastic light-scattering spectroscopy has been used to discover scaling phenomena in polymer solutions and to examine the moments of polymer cluster size distributions near the sol-gel transi- tion. In polymer gels it has been used to discover a rich variety of hitherto unexpected first-order phase transitions. The latter phenom- ena are potentially promising for the development of mechano- chemical, mechano-electrical, and electro-optical devices. In biology, quasi-elastic light-scattering spectroscopy has been used to determine quickly and accurately the diffusion coefficients and hence the size and degree of self-association of a wide variety of biological macromolecules including proteins, viruses, and antibody- antigen complexes. These studies have been used to characterize accurately the precise form of the Coulomb and van der Waals interactions between polyelectrolytes in ionic solutions. The method is also used in studies of colloid stability, ordering, and flocculation. Light-scattering spectroscopy has led to the opening of the broad field of laser Doppler velocimetry, which permits noninvasive mea- surements of fluid flow in situations ranging from aircraft wake velocity fields to the in vivo determination of blood velocity in the human retinal vasculature.

OCR for page 126
134 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS Atoms in Solids: The theoretical insights and techniques developed to describe atomic and molecular structure are being applied to problems of condensed-matter structure. Two examples are the de- scription of electronic states of impurity ions and atoms in crystals, for instance Ca+ in crystalline LiC1 or H in amorphous silicon, and the calculation of the band-gap energies of insulators and semicon- ductors. The approach is straightforward, at least in principle. In the inde- pendent-electron approximation, the exchange-correlation interaction is represented by a one-electron local potential. The variational wave function is represented as a linear combination of atomic orbitals, just as in molecular-structure calculations. Carrying out such a calculation is a formidable task. But by representing the exchange-correlation interaction with a simple local density-dependent exchange potential (ignoring correlation entirely), and using a series of Gaussian functions to describe the atomic orbitals, the theory becomes tractable, even for disordered systems for which the standard band-structure methods are not applicable. Density functional theory has established that the one-electron density uniquely defines the ground-state energy of a system, but it is surprising how well the local-exchange theory works. A serious difficulty is that the theory fails to predict accurately the band gap, the energy separation between the uppermost valence levels (top of the valence band) and the lowest conduction levels (bottom of the conduction band) of insulators and semiconductors. Recently, dramatic improvement in the theoretical predictions was obtained by making a simple self-interaction correction to the total energy, thus bringing the local-exchange theory more in line with true Hartree-Fock theory, in which the correction is implicit. It has also been shown that the correction leads to much improved energies for isolated atoms. While correlation effects may still prove to be important in some circumstances, the self-interaction correction appears to be a signifi- cant improvement in the quality of the theory. Clusters Chemical and physical processes often occur in a state of aggregation that lies midway between a dilute gas and condensed matter. The entities of this state are aggregates of small numbers of atoms or molecules called clusters. The properties of clusters are intermediate between those of single atoms or molecules and those of solids or liquids. Many of the processes that occur in the cluster regime are

OCR for page 126
SCIENTIFIC INTERFACES 135 important to technology and industry and to environmental issues. These include catalytic reactions; the formation of fog, smog, and aerosols; and the formation of particulates in combustion reactions. Clusters play a role in solution chemistry because they can retain their identity even in the liquid phase. In contrast to the detailed spectral information that exists for atomic and molecular dimers, information on the electronic properties of trimers and heavier clusters is scarce. Recently, the electronic absorp- tion spectrum of the sodium trimer was determined over the complete visible region of the spectrum in a two-photon photoionization exper- iment. The experiment provided the first unambiguous measurement of the absorption spectrum of a gas-phase triatomic metal cluster. At present, spectral or structural information about gaseous clusters beyond the trimer are lacking. These data are critically needed to provide the link between the dimer and the bulk phase. The one continuous property known today for heavier clusters, from 2 to 15 atoms, is the photoionization potential. The earliest measurements of photoionization potentials were on alkali clusters; however, more recently photoionization thresholds as a function of cluster size have been reported for other species including rare-gas clusters, metal clusters, and a few molecular clusters such as (COW, (CS2)n, and (H2S)n Laser-induced fluorescence has been used to determine the spectra of dimers of large organic molecules. These studies provide informa- tion on the energetics of cluster formation, for instance the bond dissociation energy, and information on the transfer of energy between the two moieties of the dimer via the weak van der Waals bond. Other studies have determined the spectra of an organic molecule bound to an increasingly large number of rare-gas atoms. Since the rare-gas atom acts as a weak perturbe-r of the energy levels of the host molecule, these studies approximate matrix isolation studies, allowing the de- tailed determination of the ejects of the matrix on the spectra of the host molecule. One can also approach the cluster region from the solid state. Here the goal is the size at which the collective properties of the solid disappear as the particle diameter is reduced. Experimental data have been reported for melting point, superconductivity, valence-band narrowing, photoelectric yield, plasmons, Mie optical absorption, magnetic moments, Compton profile, superparamagnetism, far- infrared absorption, specific heat, and crystallographic structure. Clusters can also be used to study surface physics, as described in the following section on Surface Science.

OCR for page 126
136 A TOMIC, MOLECULAR, AND OPTICA ~ PHYSICS Ultranarrow Optical Transitions Within the last few years it has been found that certain optical transitions of impurity ions in solids (the praseodymium ion in lanthanum trifluoride is one example) display extremely narrow linewidths, 1 kilohertz or less. These optical transitions, the zero- phonon lines, are optical analogs of the Mossbauer effect: the optically excited impurity ion suffers no recoil effect because its momentum is transferred to the lattice as a whole. Furthermore, at cryogenic temperatures there is virtually no second-order Doppler broadening. These systems are prime candidates for studying the interactions that broaden optical transitions and possibly for establishing secondary optical-frequency standards. The method has been applied to study the optical Bloch equations, the starting point for many theories in quantum optics. It was found that intense laser fields can inhibit the line-broadening effects of nuclear magnetic interactions. The phenom- enon has spurred reconsideration of microscopic theories of nuclear magnetic interactions. This research has provided the first experimental test of the optical Bloch equations, the equations of motion that were initially devised by F. Bloch to describe nuclear magnetic resonance. These equations are widely applied in quantum optics and laser spectroscopy, particularly in gases and liquids; they are the starting point for work in these fields. However, in solids it has been discovered that they fail because the laser field amplitude increases because of a coherent averaging effect that reduces the optical linewidth. A microscopic quantum theory, a modified form of the Bloch equations, has been devised to deal with this situation. Spin-Polarized Quantum Fluids All forms of matter solidify at sufficiently low temperature except for one class of systemsthe quantum fluids which remain in liquid or gaseous states as the temperature approaches zero. Within the past few years two new quantum fluids have been created using techniques from AMO physics: spin-polarized gas 3He and spin-polarized atomic hy- drogen. Because 3He has a total spin of one half, the atoms obey the Pauli principle and there is an effective repulsion between them when their nuclear spins are parallel. As a result, diffusion, viscosity, and thermal conductivity of the gas all depend on the nuclear polarization. The

OCR for page 126
140 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS 1.6 1.4 1.2 1 t.o 0.8 0.6 =. 0.4 ~ 0.2 In - LiF (001) Bi = 64.2 k' = 6.061 - Tt =295K 1 1 1 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 Time of flight Emsec] FIGURE 7.1 Surface Scattering with a Supersonic Helium Beam. An atomic beam of helium constitutes a powerful probe for studying surfaces. In fact, it is believed that the role of helium scattering in the study of surfaces may be comparable with the role played by neutron scattering in the structure of solids. An intense monoenergetic flux of atoms is required; this is provided using supersonic-beams methods developed from research i -

OCR for page 126
SCIENTIFIC INTERFACES 141 the details of such reactions; clusters may provide the key to under- standing what happens. Studying Surfaces with Laser Light Laser light makes it possible not only to study surfaces in ways never before possible but also to change the surface in new ways. Short intense laser pulses can reveal dynamical surface phenomena; coherent UV light can produce new types of nonlinear surface effects when it strikes adsorbed molecules. Laser light can trigger chemical changes on surfaces, in the substrate, and in the overlying gas. How a surface affects a photochemical reaction depends critically on whether the molecule decomposes immediately in the light or whether the reaction takes place slowly; with pulsed laser techniques the two alternatives can be distinguished. Chemical reactions of particles adsorbed on a semiconductor can be triggered in a controlled fashion- essentially catalyzed- by using laser light to generate electron-hole pairs within the material. The holes drift to the surface and trigger the reaction. If the process can be made to occur at a gas-solid interface it could provide an immensely useful new catalytic technique. The surface-sticking coefficient for vapor-phase metal atoms can change by decomposing a thin adsorbed layer of metal alkyls with laser light, offering for the first time a precise way of controlling the interchange of energy between a gas and a surface. Finally, photoreactions on the surface can trigger the growth of new materials with novel properties. The technique has important applications to semiconductor electronics and to electro-optics. A dramatic discovery from the study of surfaces with laser light is that Raman spectra on surfaces can be enhanced by a magnification of the local optical electric field. The enhancement can be enormous as much as a factor of 106. A typical experiment uses green laser light to illuminate a silver surface containing microscopic spheres or ellipsoids. These particles exhibit a plasma resonance that magnifies the electric fields in their vicinity. The plasmon resonances are of considerable interest in their own right. The technique provides an extremely on molecular scattering. The data show clearly resolved structure in the speeds of atoms scattered from a lithium fluoride crystal. The spacing of the peaks provides detailed information from which the structure and motions of atoms on the surface can be determined. (Courtesy of Max-Planck-Institute for Fluid Dynamics, Gottingen, Federal Republic of Germany.)

OCR for page 126
142 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS VAPORI ZATION PROBE LASER LASER ~ =],= 1 MAIN L CHAMBER SUPERSONIC METAL CLUSTER BEAM APPARATUS hex I 1 ~ I I LL Lit W. x NUMBER OF ATOMS IN CWSTER FIGURE 7.2 Metal Clusters. Clusters are small groups of atoms or molecules in a state of matter intermediate between a dilute gas and condensed matter. Supersonic beams of metal clusters are made by vaporizing the metal in an intense pulsed supersonic beam of helium. Using a high-power pulsed laser it is possible to vaporize even the most refractory metals. The data show mass spectra for clusters of iron, nickel, tungsten, and molybdenum. Most of the atoms in the clusters lie on the surface, even for clusters as large as 100 atoms. Because the physical and chemical properties of the clusters are dominated by surface phenomena, supersonic metal atom clusters provide an important new arena for surface science. The technique is particularly valuable for the study of catalysis, a subject of high scientific interest with potentially important applications in chemistry, in manufacturing, and in energy programs. (Courtesy of Rice University.) sensitive tool for studying surfaces because small quantities of adsor- bates in the vicinity of the metal particles are easily detected. It has been found that silicon can reproducibly change from a crystalline solid to the amorphous state and back again to a crystal with successive picosecond laser pulses. Picosecond-pulse probes have been used to study the solid-state plasma that is formed in silicon when it is illuminated by a pulse of light from a short-wavelength laser. Such studies provide a new and novel way to study the dynamics of crystal growth and to understand the mechanisms that underlie the many applications of laser annealing. These are but a few examples of how laser light can be used to study surfaces and surface chemistry. The opportunities are great, and the field is growing rapidly. The Role of Atomic, Molecular, and Optical Data in Surface Science In addition to contributing experimental and theoretical techniques to surface science, AMO physics provides basic data that are essential

OCR for page 126
SCIENTIFIC INTERFACES 143 to the interpretation of much surface research. For instance, Auger electron spectroscopy and x-ray photoelectron spectroscopy, widely used techniques for determining surface chemical composition, de- mand extensive data from AMO physics: ionization cross sections for electron and photon excitation, electron binding energies, and Auger transition probabilities. Other data are needed to relate the Auger line- shapes to the chemical states of atoms in molecules and to interpret the probabilities of multielectron excitations in core-level spectroscopies. Laser studies of surfaces require multiphoton ionization probabilities for atoms and molecules, fluorescence lifetimes and probabilities, and photoabsorption cross sections. Vibrational and rotational emission and absorption spectra for hot molecules are also needed. Stimulated desorption studies and sputtering spectroscopy require impact cross sections for ionization, branching ratios for dissociative ionization processes in small molecules, and spectroscopic data on ions and highly excited neutral species. PLASMA PHYSICS Plasmas are systems of ionized gas whose behavior is determined in large measure by collisions of electrons and ions with each other and with any neutral material that may be present. Plasmas range in physical conditions from hot fully ionized magnetohydrodynamic plas- mas occurring in planetary and interstellar environments to various experimental devices. As our understanding of the physics of plasmas has deepened, the importance of atomic processes in plasmas has become apparent. Atomic processes play a basic role in the creation of most plasmas. Neutral-particle injection into magnetically confined plasmas is used to raise the plasma temperature and bring it to ignition. Atomic processes are crucial diagnostic probes of the physical condi- tions in a plasma. The temperature, density, and fraction of different ionization stages can be derived from measurements based on atomic physics. Atomic processes ultimately destroy the plasma after the initiating source is terminated. The ionization is lost by atomic recom- bination, and the gas cools by atomic radiative processes. There is an enormous variety of plasmas in nature: most of the universe is one form of plasma or another. A multitude of atomic and molecular processes occur in plasmas, and there is an expanding need for reliable data on these processes. National plasma facilities should be available to AMO physicists so that the relevant atomic experiments can be carried out. Results of the experiments will be an important element in the physics of dense plasmas. For example, the effects of

OCR for page 126
144 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS high electron densities on atomic and molecular processes are surely important but largely unknown. Laboratories with tokamaks, mirror machines, and other devices intended for the controlled thermonuclear fusion program have become important centers of atomic-physics research. The plasmas produced in these machines contain an abundance of radiation, electrons, and ions in many states of ionization and with energies that are far from uniform. All of this complexity makes a plasma not only a fertile ground for applications of diagnostic atomic-physics techniques but also a valuable source of new information about the interactions of radiation and particles. In such a complicated environment, it is, of course, not easy to study specific individual interaction processes, but observations in plasmas have nevertheless yielded rewarding results for such areas as the rich spectroscopy of satellite lines of highly ionized atomic species. Conversely, electron-ion beam collision exper- iments have recently substantiated the main features of the dielectronic recombination process that commonly occurs in plasmas. In di- electronic recombination a free electron is captured by an ion, with the simultaneous excitation of an atomic electron. Thus it is evident that atomic physics and plasma physics support each other significantly at their common frontier. Weakly ionized plasmas containing molecular gases raise a new set of questions concerning the influence of atomic and molecular pro- cesses on the evolution of the plasma. Molecular plasmas, energized by some external source, can by virtue of internal excitations modify the course and change the products of molecular reactions. Fusion research provides a major arena for the interplay of atomic and plasma physics, as described in Chapter 8 in the section on Fusion. ATMOSPHERIC PHYSICS Electrical and chemical processes in the atmosphere effect us vitally: they govern massive climatic patterns through their influence on the energy flow from Sun to Earth and from Earth to space; they determine the ultimate fate of industrial pollutants; they control the quality of local and worldwide radio communication. Atmospheric physics, which attempts to understand the complicated web of physical and chemical processes in the atmosphere, draws heavily on AMO physics for vital data and for theoretical and experimental guidance. At its most elementary level, atmospheric physics deals with the physical processes that occur when our atmosphere is subjected to radiation, electromagnetic and corpuscular, from the Sun. A complex

OCR for page 126
SCIENTIFIC INTERFACES 145 sequence ot atomic and molecular processes determines the distribu- tion of the absorbed solar energy into ionization, dissociation, lumi- nosity, and heating; these processes drive dynamical and plasma interactions and are modified by them. The importance of each individual process depends on the location in the atmosphere and on the time: latitudinal, diurnal, seasonal, and solar-cycle variations are all substantial. The quality with which radio waves propagate is determined by a balance between photoionization and recombination in a large electri- fied region of the atmosphere. Sunlight creates atomic ions; these recombine after being converted to molecular ions by ion-molecule reactions. A successful model has been constructed for the chemistry governing recombination, though important questions remain about the role of metastable species and vibrationally excited neutral and ionic molecules. A separate group of physical processes governs the history of the photoelectrons in the atmosphere. Initially energetic, these lose their energy first by exciting and ionizing the atmospheric constituents and finally through elastic collisions with the ambient electron gas. The ambient gas is preferentially heated, and its temperature rises above that of the neutral atmosphere. The hot electron gas is cooled by excitation of the fine-structure levels of atomic oxygen and by excita- tion of the rotational and vibrational levels of molecular nitrogen and oxygen. The processes that lead to the day and night airglow of the atmo- sphere have been broadly categorized, but they are not understood in detail. This is also true for the atomic and molecular processes that follow auroral bombardment at high altitudes. Light from the aurora is a potentially powerful diagnostic probe of the exciting source and of the acceleration mechanism that appears to occur. High-latitude auroral events and polar-cap absorption events produce thermal gra- dients in the high atmosphere, driving the upper-atmosphere winds. They modify the composition of the atmosphere, and they may be related to climatic variations. The chemistry of the mesosphere and stratosphere has undergone rapid development, particularly since it was recognized that the release of fluorocarbons into the atmosphere could attack the ozone layer. Further studies of the molecular processes are needed to understand the potential hazards from fluorocarbons and other pollutants. The penetration of solar radiation is not yet adequately known, nor is the intensity of ultraviolet radiation at low altitudes. The terrestrial atmosphere has evolved markedly since the formation of the planets owing to many influences, including life. Molecular

OCR for page 126
146 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS physics is crucially involved in the attempts to reconstruct the history of the early atmosphere as it responded to changes in solar luminosity and to understand the interactions today of the physical and biological processes that are determining the future of the atmosphere. The effects of an increase in the abundance of carbon dioxide are of crucial importance to our future. Basic problems presented by atmospheric science often have imme- diate consequences. For example, the Space Shuttle was found to glow in the dark even at altitudes as high as 300 km (200 miles). The origin of the glow has not been discovered: it may be produced by collisions of the oxygen atoms of the atmosphere with material on the surface of the spacecraft. The emitting species appears to be molecular in character, but its identity is uncertain. A similar glow observed on orbiting satellites has been attributed to the hydroxyl radical, but the limited data on the Space Shuttle glow suggest that a different molecule must be responsible. It is essential to identify the source of the glow, not least because of its potential impact on the durability and effec- tiveness of instruments in space such as the Space Telescope. NUCLEAR PHYSICS Atomic and nuclear physics are closely related. Here we focus on three areas of contact. The first is the role of atomic spectroscopy in measuring the fundamental static characteristics of nuclear states. This has been an indispensable tool of nuclear physics for decades but now contributes more information than ever. The second is the use of atomic techniques to provide polarized nuclei for sources and targets in nuclear experiments. The third is the study of the dynamical interac- tions between nuclei and their atomic environments. The physics at this interface between the two disciplines, still in its infancy, has already provided nuclear physics with new insights and tantalizing clues. Optical Studies of the Nucleus The mass, size, shape, and internal structure of the nucleus at the center of each atom slightly alter the positions of the atomic energy levels. These energy-level shifts can be found from careful determina- tions of the optical spectral lines of the atoms. The name "hyperfine structure," given to a large class of these effects, emphasizes their intrinsic smallness, but atomic spectroscopy stands out in physics through its extreme precision, and these tiny effects are accessible to

OCR for page 126
SCIENTIFIC INTERFACES 147 measurement and analysis. For example, although nuclear diameters increase with the mass of the nucleus, even the largest nuclei are about one hundred thousand times smaller than most electron orbits in the atom. Nevertheless, atomic spectroscopy provides a powerful tool for accurate measurement of very small changes in the nuclear radius. The complex nature of the nucleus is revealed to the atomic electrons that surround it through the electric and magnetic fields that arise from the nuclear protons, neutrons, and pions and, at a more fundamental level, the quarks. These electromagnetic properties of the nucleus can be parameterized in terms of electric and magnetic moments that describe the size, shape, and charge and current distri- butions of the nuclear constituents. The spins, the magnetic dipole moments, and the electric quadrupole moments of a wide variety of nuclei can be found from measurements of the hyperfine interaction made by two atomic techniques: high-resolution optical spectroscopy and atomic-beam magnetic resonance. Such data provide valuable input to nuclear theorists, who use them to test the still-evolving theories of nuclear structure. Hyperfine measurements have been fundamental to our understand- ing of the most basic properties of the ground states of stable nuclei. It has now become possible to apply these techniques to a large class of unstable and short-lived nuclei, and the horizons of the field have expanded dramatically. In addition to the ground states of vast numbers of nuclear species, there are many relatively long-lived excited states whose size, shape, and moments can now be measured. Every bit of such information adds a further constraint on possible theories of nuclear structure. To appreciate the potential of this new development, one only has to note that while there are about 190 stable or naturally occurring isotopes, there are close to 1600 known unstable nuclei, and it is expected that many more will yet be discovered. Lasers, with their sharp wavelengths and high intensities, have been the principal tool opening up new vistas in the study of nuclei by atomic spectroscopy. The field has been enormously enhanced by the installation of lasers on-line at nuclear reactor and accelerator facilities, where the unstable nuclear isotopes are produced. (See Figure 7.3.) Systematic studies of chains of isotopes, especially at the Isotope Separator On-Line (ISOLDE) facility at the European CERN labora- tories, are providing vital information for nuclear-structure physicists. Techniques from atomic physics are extensively employed to study the sizes and shapes of the nuclei, their spins, and their electric and magnetic moments. These techniques include atomic-beam magnetic

OCR for page 126
148 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS it. \ \ ~ \~.,? FIGURE 7.3. Atomic Physics at ISOLDE. The earliest evidence that atomic nuclei possess spin and magnetic moments came from atomic physics, and as new experimental techniques have been developed, the range and precision of these nuclear studies have increased steadily. The drawing illustrates a number of atomic experiments on nuclear properties being carried out at ISOLDE, the on-line isotope separator at CERN (Geneva). Devoted to research in nuclear physics, ISOLDE is capable of producing essentially any isotope. The isotopes are formed by bombarding a target with a 600-MeV beam from a proton synchrocyclotron (right-hand side of drawing). The radioactive ions are accelerated and mass selected by a bending magnet and then distributed to the experiments by a "switchyard." Among the equipment shown on the experimental floor are an atomic-beam magnetic resonance apparatus and a setup for laser spectroscopy (the laser can be seen two floors above). Optical pumping is also employed. A similar experimental facility is being developed at the University Isotope Separator (UNISOR) at Oak Ridge. (Courtesy of CERN, Geneva, Switzerland, and Laboratoire Aime Cotton, Orsay, France.) resonance, optical pumping, and laser spectroscopy. Nuclei with lifetimes as short as 10 milliseconds have been studied. Out of this research have emerged discoveries such as shape staggering, in which adjacent isotopes alternate between ablate and prolate forms, and shape isomerism, in which a nucleus with two nearby excited levels can assume widely varying shapes. These discoveries have played an important role in the development of nuclear models. As laser spectroscopy continues its rapid advances one can expect

OCR for page 126
SCIENTIFIC INTERFACES 149 corresponding advances in our ability to obtain spectroscopic informa- tion about nuclear states. It will become possible to measure the properties of extremely rare and short-lived nuclear states, including highly excited collective states with very high spin values. The remarkable recent progress made by atomic physicists in trapping ions and atoms for long periods of time is certain to be exploited for further high-precision measurements of nuclear moments. Polarized Nuclear Sources Nuclear physics relies on techniques from atomic physics for pro- ducing the spin-polarized projectiles and target atoms that are being used increasingly in nuclear-reaction experiments. Nuclear physicists require the most intense available beams of nuclei with their spins oriented in a particular direction in space, rather than being randomly oriented. A number of different atomic methods are used for producing polarized nuclei, such as protons, deuterons, 3He nuclei, and lithium nuclei. The spins of nuclei are aligned through their magnetic moments, which offer a "handle" for using a magnetic field to rotate and orient them. However, the nuclear moment is only about one thousandth as large as the magnetic moment of the electron. Therefore, almost all nuclear polarization schemes rely on first polarizing one of the atomic electrons that surround the nucleus. The hyperfine interaction, which couples the magnetism of the electron to the magnetic moment of the nucleus, can then be used to align the nuclei. (Two recent advances in the production of polarized protons and 3He are described earlier in this chapter in the section on Condensed-Matter Physics and Materials Science.) In this important area of ion-source technology it is essential to know the atomic collision cross sections and other atomic parame- ters that determine the efficacy of proposed new polarization mecha- nisms. Dynamics at the Atom-Nuclear Frontier Atomic and nuclear physics have a common frontier that has recently become the site for research into questions that previously could never be addressed but that now, thanks to experimental and theoretical advances, we can hope to answer. What happens to the electrons in an atom when a nuclear particle, such as a proton or a heavy ion, penetrates through the atomic electron shells on its way into or out of the nucleus, where it initiates a nuclear reaction? How does the course of the nuclear reaction and in particular its duration

OCR for page 126
150 ATOMIC, MOLECULAR, AND OPTICAL PHYSICS affect the atomic electrons? Can lessons for nuclear physics be learned by studying these atomic effects? Or, conversely, can useful atomic information be gained? In a number of accelerator laboratories that use all types of accelerators from tandem Van de Graaffs to the most advanced heavy-ion linear colliders, the effects of nuclear reactions at millielectron volt to gigaelectron volt energies on the participating atoms have come under study. The research has begun to yield valuable information on nuclear reactions and atomic-collision pro- cesses. One of the most direct measurements at the atomic-nuclear frontier is the determination of the lifetime of a compound nucleus, before it comes apart again with re-emission of a proton. On its way into or out of the nucleus, the projectile may knock an inner-shell electron out of its orbit, leaving a vacancy, which eventually leads to the emission of an x-ray photon. From such x-ray measurements, and with the aid of results from calculations in atomic-collision theory, it is possible, from purely atomic observations, to determine the lifetime of the compound nucleus, using the atom as a clock. Lifetimes in the range from 10-~6 to 10-'8 second have been measured by such atomic techniques. Life- times of even much shorter-lived nuclear states (10-'8 and even 10-2 second) can now be determined by crystal blocking techniques that exemplify the overlap of nuclear physics with both condensed-matter and atomic physics. Finally, we note here a new development: an experiment that is being carried forward at a laboratory of particle physics but that represents a confluence of particle, nuclear, and atomic physics. This is the attempt to make protonium, an atom composed of a proton and an antiproton. The antiproton storage ring at CERN produces enough of these particles to provide the hope of making protonium, using hydrogen negative ions as the source of protons. Observation of this atom would provide an important new avenue for the study of quantum electrodynamics and quantum chromodynamics.