Although natural synchrotron radiation from charged particles spiraling around magnetic-field lines in space is as old as the stars—for example the light we see from the Crab Nebula—short-wavelength synchrotron radiation generated by relativistic electrons in circular accelerators is only a half-century old. The first observation, literally since it was visible light that was seen, came at the General Electric Research Laboratory in Schenectady, New York, on April 24, 1947. In the 50 years since, synchrotron radiation has become a premier research tool for the study of matter in all its varied manifestations, as facilities around the world constantly evolved to provide this light in ever more useful forms.
From the time of their discovery in 1895, both scientists and society have recognized the exceptional importance of x rays, beginning with the awarding of the very first Nobel Prize in Physics in 1901 to Röntgen "in recognition of the extraordinary services he has rendered by the discovery of the remarkable rays subsequently named after him." By the time synchrotron radiation was observed almost a half-century later, the scientific use of x rays was well established. Some highlights include
· 1909: Barkla and Sadler discover characteristic x-ray radiation (1917 Nobel Prize to Barkla)
· 1912: von Laue, Friedrich, and Knipping observe x-ray diffraction (1914 Nobel Prize to von Laue)
· 1913: Bragg, father and son, build an x-ray spectrometer (1915 Nobel Prize)
· 1913: Moseley develops quantitative x-ray spectroscopy and Moseley’s Law
· 1916: Siegbahn and Stenstrom observe emission satellites (1924 Nobel Prize to Siegbahn)
· 1921: Wentzel observes two-electron excitations
· 1922: Meitner discovers Auger electrons
· 1924: Lindh and Lundquist resolve chemical shifts
· 1927: Coster and Druyvesteyn observe valence-core multiplets
· 1931: Johann develops bent-crystal spectroscopy
The theoretical basis for synchrotron radiation traces back to the time of Thomson's discovery of the electron. In 1897, Larmor derived an expression from classical electrodynamics for the instantaneous total power radiated by an accelerated charged particle. The following year, Liénard extended this result to the case of a relativistic particle undergoing centripetal acceleration in a circular trajectory. Liénard's formula showed the radiated power to be proportional to (E/mc2)4/R2, where E is particle energy, m is the rest mass, and R is the radius of the trajectory. A decade later in 1907, Schott reported his attempt to explain the discrete nature of atomic spectra by treating the motion of a relativistic electron in a homogeneous magnetic field. In so doing, he obtained expressions for the angular distribution of the radiation as a function of the harmonic of the orbital frequency. Schott wrote a book-length essay on the subject in 1912.
When attempts to understand atomic structure took a different turn with the work of Bohr, attention to radiation from circulating electrons waned. Interest in the radiation as an energy-loss mechanism was reawakened in the 1920s after physicists began contemplating magnetic-induction electron accelerators (betatrons) as machines to produce intense beams of x rays by directing the accelerated beam to a suitable target. (Commercial betatrons are still used as x-ray sources.) The first betatron to operate successfully was a 2.3-MeV device built in 1940 by Kerst at the University of Illinois, followed at GE by a 20-MeV and then a 100-MeV machine to produce high-energy x rays for nuclear research. Meanwhile, in the Soviet Union, Ivanenko and Pomeranchuk published their 1944 calculations showing that energy losses due to radiating electrons would set a limit on the energy obtainable in a betatron, which they estimated to be around 0.5 GeV.
Subsequent theoretical work proceeded independently by Pomeranchuk and others in the Soviet Union and in the U. S., where by 1945 Schwinger had worked out in considerable detail the classical (i.e., non-quantum) theory of radiation from accelerated relativistic electrons. Major features demonstrated for the case of circular trajectories included the warping of the globular non-relativistic dipole radiation pattern into the strongly forward peaked distribution that gives synchrotron radiation its highly collimated property and the shift of the spectrum of the radiation to higher photon energies (higher harmonics of the orbital frequency) as the electron energy increased, with the photon energy at the peak of the distribution varying as E3/R. Schwinger did not publish his complete findings until 1949, but he made them available to interested parties. Quantum calculations, such as those by Sokolov and Tersov in the Soviet Union, later confirmed the classical results for electron energies below about 10 TeV.
After testing of GE's 100-MeV betatron commenced in 1944, Blewett suggested a search for the radiation losses, which he expected from the work of Ivanenko and Pomeranchuk to be significant at this energy. However, two factors prevented success: whereas, according to Schwinger's calculations, the radiation spectrum for the 100-MeV betatron should peak in the near- infrared/visible range, the search took place in the radio and microwave regions at the orbital frequency (and low harmonics) and the tube in which the electrons circulated was opaque. Although quantitative measurements reported in 1946 of the electron-orbit radius as it shrunk with energy were in accord with predicted losses, there was also another proposed explanation with the result that, while Blewett remained convinced the losses were due to synchrotron radiation, his colleagues were not.
Advances on another accelerator front led to the 1947 visual observation of synchrotron radiation at GE. The mass of particles in a cyclotron grows as the energy increases into the relativistic range. The heavier particles then arrive too late at the electrodes for a radio-frequency (RF) voltage of fixed frequency to accelerate them, thereby limiting the maximum particle energy. To deal with this problem, in 1945 McMillan in the U. S. and Veksler in the Soviet Union independently proposed decreasing the frequency of the RF voltage as the energy increases to keep the voltage and the particle in synch. This was a specific application of their phase-stability principle for RF accelerators, which explains how particles that are too fast get less acceleration and slow down relative to their companions while particles that are too slow get more and speed up, thereby resulting in a stable bunch of particles that are accelerated together.
At GE, Pollack got permission to assemble a team to build a 70-MeV electron synchrotron to test the idea. Fortunately for the future of synchrotron radiation, the machine was not fully shielded and the coating on the doughnut-shaped electron tube was transparent, which allowed a technician to look around the shielding with a large mirror to check for sparking in the tube. Instead, he saw a bright arc of light, which the GE group quickly realized was actually coming from the electron beam. Langmuir is credited as recognizing it as synchrotron radiation or, as he called it, "Schwinger radiation." Subsequent measurements by the GE group began the experimental establishment of its spectral and polarization properties. Characterization measurements were also carried out in the 1950s at a 250-MeV synchrotron at the Lebedev Institute in Moscow.
Synchrotron light from the 70-MeV electron synchrotron at GE.
The next step came with the 1956 experiments of Tomboulian and Hartman, who were granted a two-week run at the 320-MeV electron synchrotron at Cornell. Despite the limited time, they were able to confirm the spectral and angular distribution of the radiation with a grazing- incidence spectrograph in the ultraviolet from 80 Å to 300 Å. They also reported the first soft x- ray spectroscopy experiments with synchrotron radiation, measuring the transmission of beryllium and aluminum foils near the K and L edges. However, despite the advantages of synchrotron radiation that were detailed by the Cornell scientists and the interest their work stimulated, it wasn't until 1961 that an experimental program using synchrotron radiation got under way when the National Bureau of Standards (now National Institute of Standards and Technology) modified its 180-MeV electron synchrotron to allow access to the radiation via a tangent section into the machine's vacuum system.
Under Madden and Codling, measurements began at the new NBS facility (Synchrotron Ultraviolet Radiation Facility or SURF) to determine the potential of synchrotron radiation for standards and as a source for spectroscopy in the ultraviolet (the wavelength for peak radiated power per unit wavelength was 335 Å). Absorption spectra of noble gases revealed a large number of previously unobserved resonances due to inner-shell and two-electron excitations, including doubly excited helium, which remains today a prime test bed for studying electron- electron correlations. These findings further stimulated the growing interest in synchrotron radiation. Establishment of SURF began the first generation of synchrotron-radiation facilities, sometimes also called parasitic facilities because the accelerators were built and usually operated primarily for high-energy or nuclear physics. However, the NBS synchrotron had outlived it usefulness for nuclear physics and was no longer used for this purpose.
If SURF headed the first generation, it was not by much, as activity was also blossoming in both Europe and Asia. At the Frascati laboratory near Rome, researchers began measuring absorption in thin metal films using a 1.15-GeV synchrotron. In 1962, scientists in Tokyo formed the INS- SOR (Institute for Nuclear Studies-Synchrotron Orbital Radiation) group and by 1965 were making measurements of soft x-ray absorption spectra of solids using light from a 750-MeV synchrotron. The trend toward higher energy and shorter wavelengths took a big leap with the use of the 6-GeV Deutsches Elektronen-Synchrotron (DESY) in Hamburg, which began operating for both high-energy physics and synchrotron radiation in 1964. With synchrotron radiation available at wavelengths in the x-ray region down to 0.1 Å, experimenters at DESY were able to carefully check the spectral distribution against Schwinger's theory, as well as begin absorption measurements of metals and alkali halides and of photoemission in aluminum.
While the number of synchrotrons with budding synchrotron-radiation facilities was growing, the next major advance was the development of electron storage rings, the basis for all of today's synchrotron sources. In the 1950s, the Midwest Universities Research Association (MURA) was formed to develop a proposal for a high-current accelerator for particle physics. As part of the project, Mills and Rowe designed a 240-MeV storage ring, then a new idea, as a test bed for advanced accelerator concepts. Politics intervened, however, and the decision was made in 1963 to build a new high-energy accelerator in Illinois, a facility that became the Fermi National Accelerator Laboratory. With this decision, MURA eventually dissolved, but in the meantime construction of the storage ring proceeded.
Thanks to the rapidly swelling interest in synchrotron radiation for solid-state research that stimulated a 1965 study by the U. S. National Research Council documenting this promise, MURA agreed to alterations in the storage-ring vacuum chamber that would provide access to synchrotron radiation without interfering with the accelerator studies. With MURA's demise in 1967, funding for the original purpose of the storage ring also disappeared, but supported by the U. S. Air Force Office of Scientific Research, the University of Wisconsin took on the responsibility of completing the storage ring, known as Tantalus I, and operating it for synchrotron-radiation research. The first spectrum was measured in 1968. In subsequent years, improvements enabled Tantalus I to reach its peak performance, add a full complement of ten beamlines with monochromators, and become in many respects a model for today's multi-user synchrotron-radiation facilities.
With Tantalus I, the superiority of the electron storage ring as a source of synchrotron radiation became evident. In a storage ring, the beam continuously circulates current at a fixed energy for periods up to many hours, whereas the synchrotron beam undergoes a repeated sequence of injection, acceleration, and extraction at rates up to 50 Hz. Among the advantages stemming from this feature are a much higher "duty cycle" when the beam is available, higher beam currents and hence higher fluxes of radiation, a synchrotron-radiation spectrum that does not change with time, greater beam stability, and a reduced radiation hazard.
A surge of interest in storage rings soon followed. In 1971, synchrotron-radiation work began on the 540-MeV ACO storage ring at the Orsay laboratory in France. With the help of the Wisconsin group, the NBS converted its synchrotron into a 250-MeV storage ring (SURF II) in 1974. The same year the INS-SOR group (now part of the Institute for Solid State Physics) in Tokyo began commissioning a 300-MeV storage ring, generally considered the first machine designed from the start specifically for the production of synchrotron radiation. The first storage ring in the multi- GeV class to provide x rays to a large community of synchrotron-radiation users was the 2.5-GeV SPEAR ring at the Stanford Linear Accelerator Center (SLAC), where a beamline with five experimental stations was added in 1974 under the auspices of the Stanford Synchrotron Radiation Project. Other large storage rings to which synchrotron-radiation capabilities were added early on include DORIS at the DESY laboratory, VEPP-3 at the Institute for Nuclear Physics in Novosibirsk, DCI at Orsay, and CESR at Cornell (the CHESS facility).
The larger storage rings just cited were electron-positron colliding-beam machines that were operated to provide the highest possible collision rates without blowing up the beams, a condition that generally meant low beam currents. Moreover, while studying the then-fashionable J/? particle and its relatives, they often ran at low beam energies. Under these conditions, parasitic operation meant a severely limited output of synchrotron radiation, thereby motivating a clamor for storage rings designed for and dedicated to the production of synchrotron radiation. The Synchrotron Radiation Source (SRS) at the Daresbury Laboratory in the UK was the first fruit of this movement. Synchrotron-radiation research had begun at Daresbury around 1970 with the addition of a beamline to the 5-GeV NINA electron synchrotron. When NINA shut down in 1977, a plan was already approved to build a 2-GeV electron storage ring at the same site expressly for synchrotron radiation. Experiments began at the new facility in 1981.
In the U. S., after a 1976 National Research Council study documented an increasing imbalance between demand for synchrotron radiation and its availability, construction of the National Synchrotron Light Source (NSLS) at the Brookhaven National Laboratory was approved. With construction completed in 1981, the NASALS complex included separate 700-MeV and 2.5-GeV storage rings for production of UV and x rays, respectively. During this same period, the University of Wisconsin Synchrotron Radiation Center built a new 1-GeV storage ring named Aladdin, which replaced the old Tantalus I (part of which was later sent to the Smithsonian for eventual exhibit). In Japan, the Photon Factory was completed in 1982 at the KEK laboratory in Tsukuba. And in Berlin, the BESSY facility began serving users in 1982 with a 0.8-GeV storage ring. And at Orsay, LURE (Laboratoire pour l'Utilisation du Rayonnement Electromagnétique) began operating an 800-MeV storage ring, SuperACO, in 1984.
Elsewhere, some of the first-generation facilities gradually evolved toward second-generation status by means of upgrades and agreements with laboratory managements to dedicate a fraction and sometimes all of the yearly machine operations to synchrotron radiation as the high-energy physics frontier advanced. The Stanford Synchrotron Radiation Laboratory at SLAC and HASYLAB (Hamburger Synchrotronstrahlungslabor) at DESY are prime examples. All of these second-generation facilities provide fine examples of the productivity of a dedicated source of synchrotron radiation. Over the years, for example, the SRS has grown to about 40 experimental stations serving around 4000 users from physics and biology to engineering, and the NSLS has around 80 operating beamlines and more than 2200 users each year. (By this time, the number of synchrotron-radiation facilities has grown too large to mention them all here; the reader should turn to Chapter 8 for a list of current facilities and the spectral ranges they serve.)
Major experimental developments included a major enhancement of photoemission for studying the electronic structure of solids and surfaces (for example, angle-resolved photoemission began at Tantalus), the development of extended x-ray absorption fine-structure spectroscopy (EXAFS) for the measurement of local atomic structure (which got its start at SSRL), and the extension of high-resolution protein crystallography to small, difficult to grow, or otherwise unstable samples (beginning with the work at SSRL and expanding rapidly to DESY, CHESS, and elsewhere).
As the clamor for facilities dedicated to synchrotron radiation expanded in the 1970's, users increasingly appreciated that spectral brightness or brilliance (the flux per unit area of the radiation source per unit solid angle of the radiation cone per unit spectral bandwidth) was often more important than flux alone for many experiments. For example, since the photon beam is most often ribbon shaped with a larger horizontal than vertical size, typical spectroscopy experiments at synchrotron facilities use monochromators with horizontal slits. Because spectroscopy experiments achieve the highest spectral resolution when the slits are narrowed, obtaining a useful flux through the monochromator exit slits requires that the photon beam have a small vertical size and angular divergence so that most of the flux from the source can pass through the narrowed entrance slits and strike the dispersing element at nearly the same angle (i.e., when the vertical brightness is high). Crystallography experiments, especially those with small crystals and large unit cells, also place a premium on brightness, since its is necessary to match the incident beam to the crystal size while maintaining sufficient angular resolution to resolve closely spaced diffraction spots.
Brightness (flux density in phase space) is an invariant quantity in statistical mechanics, so that no optical technique can improve it. For example, focusing the beam to a smaller size necessarily increases the beam divergence, and vice-versa; apertures can help reduce beam size and divergence but only at the expense of flux. The cure therefore is proper design of the source, the electron beam in the storage ring. The size and divergence of the electron beam are determined by the storage-ring lattice—the arrangement and strengths of the dipole, quadrupole, and sextupole magnets. As planning of the NSLS progressed, Chasman and Green designed what has become the prototype lattice (a so-called double-bend achromat) for storage rings with a low emittance (product of beam size and divergence) and hence a light source with high brightness. The Chasman-Green lattice and variations are the basis for most of today's synchrotron sources.
Undulators provide a way to take maximum advantage of the intrinsic brightness of the synchrotron-radiation source. The magnetic structure of today's most common (planar) undulator is an array of closely spaced vertically oriented dipole magnets of alternating polarity. As the electron beam passes longitudinally through the array, it's trajectory oscillates in the horizontal plane. Owing to the relatively weak field, the radiation cones emitted at each bend in the trajectory overlap, giving rise to a constructive interference effect that results in one or a few spectrally narrow peaks (a fundamental and harmonics) in a beam that is highly collimated in both the horizontal and vertical directions; that is, the beam has a high spectral brightness (see Chapter 4). Tuning the wavelengths of the harmonics is by means of mechanically adjusting the vertical spacing (gap) between the pole tips.
The undulator concept traces back to the 1947 theoretical work Ginzburg in the Soviet Union. Motz and coworkers experimentally verified the idea in 1953 by building an undulator and using it to produce radiation from the millimeter-wave to the visible range in experiments with a linear accelerator at Stanford University. The next step came in the 1970's with the installation of undulators in storage rings at the Lebedev Institute in Moscow and the Tomsk Polytechnic Institute. Measurements at these laboratories began to provide the information needed for a comprehensive description of undulator radiation. Owing to the large number of closely spaced dipoles, electromagnets or superconducting magnets are not that best choice for undulators, which became practical devices for producing synchrotron radiation in storage rings in 1981 when Halbach at Lawrence Berkeley Laboratory and coworkers constructed a device based on permanent magnets and successfully tested it at SSRL. Parallel work was also under way in Novosibirsk.
Klaus Halbach shown in 1986 with Kwang-Je Kim discussing a model of an undulator that Halbach designed.
Wigglers are similar to undulators but generally have higher fields and fewer dipoles, with the result that they produce a continuous spectrum with a higher flux and a spectrum that extends to shorter wavelengths than bend magnets. Despite the similarity, wigglers evolved independently from undulators at the start. A decade after the initial suggestion by Robinson, a wiggler was installed in 1966 at the Cambridge Electron Accelerator (a 3-GeV storage ring that was actually the first multi-GeV storage ring to produce x rays before it was shut down in 1972) to enhance beam storage. In 1979, a wiggler comprising just seven electromagnet poles at SSRL was the first to be used for producing synchrotron radiation. Nowadays, wigglers may be permanent-magnet devices following the Halbach design or be based on high-field superconductors that shift the spectrum to the shortest wavelengths. Together, wigglers and undulators are called insertion devices because they are placed in one of the generally empty straight sections that connect the curved arcs of large storage rings, where the magnets that guide and focus the electron beam reside.
The planar insertion devices just described produce radiation that is linearly polarized in the horizontal plane. However, a feature available from bend-magnet sources is the generation of elliptically polarized radiation, with the most obvious applications in the study of magnetic materials. The radiation from a bend magnet is elliptically polarized above and below the horizontal plane of the electron-beam orbit, and this feature has now been exploited at many facilities, including the pioneering work on magnetic materials in the hard x-ray region by Schütz and coworkers at HASYLAB (1987) and in the soft x-ray region by Chen and colleagues at the NSLS (1990). Now, among several designs for both wigglers and undulators that produce elliptically polarized synchrotron radiation, some have been implemented, tested, and are in regular use.
Both undulators and wigglers have been retrofitted into older storage rings, and in some cases, the second-generation rings, such as those at the NSLS, were designed with the possibility of incorporating insertion devices. Nonetheless, even before the second-generation facilities were broken in, synchrotron users recognized that a new generation of storage rings with a still lower emittance and long straight sections for undulators would permit achieving even higher brightness and with it, a considerable degree of spatial coherence. Beneficiaries of high brightness would include those who need spatially resolved information, ranging from x-ray microscopy to spectromicroscopy (the combination of spectroscopy and microscopy) and those who need temporal resolution, as well as spectroscopists, crystallographers, and anyone who needs to collect higher resolution data faster.
Construction of third-generation synchrotron-radiation facilities brings us to the present day. Following the NSLS two-ring model, third-generation facilities specialize in either short- wavelength (high-energy or hard) x rays or vacuum-ultraviolet and long-wavelength (low energy or soft) x rays. The range in between (intermediate-energy x rays) is accessible by both. The European Synchrotron Radiation Facility (ESRF) in Grenoble was the first of the third-generation hard x-ray sources to operate, coming on line for experiments by users with a 6-GeV storage ring and a partial complement of commissioned beamlines in 1994. The ESRF has been followed by the Advanced Photon Source at Argonne National Laboratory (7 GeV) in late 1996, and SPring-8 (8 GeV) in Harima Science Garden City in Japan in late 1997. These machines are physically large (850 to 1440 meters in circumference) with a capability for 30 or more insertion-device, and a comparable number of bend-magnet, beamlines.
Among the long-wavelength sources, the Advanced Light Source at Berkeley (1.9 GeV) began its scientific program in early 1994, as did the Synchrotrone Trieste (2.0 GeV) in Italy, followed by the Synchrotron Radiation Research Center (1.3 GeV) in Hsinchu, Taiwan, and the Pohang Light Source (2.0 GeV) in Pohang, Korea. These physically smaller machines (120 to 280 meters in circumference) have fewer straight sections and therefore can service fewer insertion-device beamlines than the larger machines, but since they are also less expensive, many more of them have been and are being constructed around the world, from Canada in North America; to Brazil in South America; to Japan, China, Thailand, and India in Asia; and to Sweden, Germany, Switzerland, and other European countries, although in some cases, these are not truly third- generation machines in terms of performance. Addition of superconducting bend magnets to the storage-ring lattice in these smaller machines, as some facilities are planning to do, allows them to extend their spectral coverage to higher photon energies without sacrificing their performance at lower photon energies.
The race to develop a new generation of synchrotron radiation sources with vastly enhanced performance has already begun, even as the third-generation facilities enter their prime, which takes us past the present into the future; namely, to the fourth generation. The candidate with the best scientific case for a fourth-generation source is the hard x-ray (wavelength less than 1Å) free- electron laser (FEL) based on a very long undulator in a high-energy electron linear accelerator. Such a device would have a peak brightness many orders of magnitude beyond that of the third- generation sources, as well as pulse lengths of 100 fs or shorter, and would be fully coherent. Research and development on the many technical challenges that must be overcome are well under way at many laboratories around the world. In the United States, effort is centering around the multi-institutional "Linac Coherent Light Source" proposal to use 15-GeV electrons from the SLAC linac as the source for a 1.5-Å FEL, which if successful would lay the foundation for a later sub-angstrom x-ray FEL. In Europe, HASYLAB at DESY is hosting the two-phase TTF-FEL project culminating in a device operating at 6.4 Å several years from now. The project would pave the way to a still more ambitious 0.1-Å FEL (TESLA-FEL) farther in the future.
The articles used in writing this history are recollections and reviews that contain references to the original sources.
G. C. Baldwin, “Origin of Synchrotron Radiation,” Physics Today 28, No. 1 (1975) 9.
K. Codling, “Atomic and Molecular Physics Using Synchrotron Radiation—the Early Years,” J. Synch. Rad 4, Part 6 (1997) 316. Special issue devoted to the 50th anniversary of the observation of synchrotron radiation.
S. Doniach, K. Hodgson, I. Lindau, P. Pianetta, and H. Winick, “Early Work with Synchrotron Radiation at Stanford,” J. Synch. Rad 4, Part 6 (1997) 380. Special issue devoted to the 50th anniversary of the observation of synchrotron radiation.
P. L. Hartman, “Early Experimental Work on Synchrotron Radiation,” Synchrotron Radiation News 1, No. 4 (1988) 28.
E.-E. Koch, D. E. Eastman, and Y. Farges, “Synchrotron Radiation—A Powerful Tool in Science,” in Handbook on Synchrotron Radiation, Vol 1a, E.-E. Koch, ed., North-Holland Publishing Company; Amsterdam, 1983, pp. 1-63.
D. W. Kerst, comment on letter by Baldwin, Physics Today 28, No. 1 (1975) 10.
G. N. Kulipanov and A. N. Skrinksy, “Early Work on Synchrotron Radiation,” Synchrotron Radiation News 1, No. 3 (1988) 32.
C. Kunz, “Introduction—Properties of Synchrotron Radiation,” in Synchrotron Radiation Techniques and Applications, C. Kunz, ed., Springer-Verlag, Berlin, 1979, pp. 1-23.
K. R. Lea, “Highlights of Synchrotron Radiation,” Phys. Rep. (Phys. Lett. C) 43, No. 8 (1978) 337.
S. R. Leone, “Report of the Basic Energy Sciences Advisory Committee Panel on Novel Coherent Light Sources,” U. S. Department of Energy, January 1999. Available on the World Wide Web at URL: http://www.er.doe.gov/production/bes/BESAC/ncls_rep.PDF.
D. W. Lynch, “Tantalus, a 240 MeV Dedicated Source of Synchrotron Radiation, 1968-1986,” J. Synch. Rad 4, Part 6 (1997) 334. Special issue devoted to the 50th anniversary of the observation of synchrotron radiation.
R. P. Madden, “Synchrotron Radiation and Applications,” in X-ray Spectroscopy, L. V. Azaroff, ed., McGraw-Hill Book Company, New York, 1974, pp. 338-378.
I. H. Munro, “Synchrotron Radiation Research in the UK,” J. Synch. Rad 4, Part 6 (1997) 344. Special issue devoted to the 50th anniversary of the observation of synchrotron radiation.
M. L. Perlman, E. M. Rowe, and R. E. Watson, “Synchrotron Radiation—Light Fantastic,” Physics Today 27, No.7 (1974) 30.
H. C. Pollock, “The Discovery of Synchrotron Radiation,” Am J. Phys. 51, No. 3 (1983) 278.
E. Rowe, “Synchrotron Radiation: Facilities in the United States, Physics Today 34, No. 5 (1981) 28.
T. Sasaki, “A Prospect and Retrospect—the Japanese Case,” J. Synch. Rad 4, Part 6 (1997) 359. Special issue devoted to the 50th anniversary of the observation of synchrotron radiation.
H. Winick, G. Brown, K. Halbach, and J. Harris, “Synchrotron Radiation: Wiggler and Undulator Magnets,” Physics Today 34, No. 5 (1981) 50.
H. Winick and S. Doniach, “An Overview of Synchrotron Radiation Research,” in Synchrotron Radiation Research, H. Winick and S. Doniach, eds., Plenum Press, New York and London, 1980, pp. 1-10.