X-ray astronomy is the observation of celestial objects that emit radiation in the region of the electromagnetic spectrum where the wavelengths are shorter than ultraviolet light. This region extends from about 10 A (angstrom) to less than 0.01 A. The corresponding energies of the X rays that reach the solar system range from a lower limit below 1,000 electron volts (1 keV)--interstellar matter absorbs anything less than this--to an upper limit near 1,000,000 electron volts (1 MeV), determined by the rapidly diminishing intensity. X-ray observations are crucial in attacking a wide variety of fundamental problems with profound cosmological implications, ranging from the large-scale structure of the universe to the properties of compact exotic objects such as pulsars and black holes. HISTORY Because X rays are rapidly absorbed in matter (unlike the signals observed in \Tradio astronomy\t, which can penetrate the atmosphere to reach detectors on the ground), the first observation of nonterrestrial X rays awaited the launching of rockets carrying suitable detectors. As early as 1939 it had been predicted that X rays with intensities greater than that corresponding to the blackbody temperature of the solar photosphere (6,000 K) were required to produce the ionization observed in the Earth's lower ionosphere. The blackening of a film that was protected from visible and ultraviolet light and flown in a V-2 rocket in 1948 gave the first indication of the postulated X-ray emissions. Geiger counters sent aloft in 1949 by Herbert Friedman and others of the U.S. Naval Research Laboratory confirmed the presence of solar X rays at wavelengths near 10 angstroms. Extrapolations based on the observed intensity from the Sun and the distances to other stars, however, suggested that it would not be feasible to detect X rays of nonsolar origin. The first extrasolar source was accidentally discovered in 1962 by Riccardo Giacconi, Herbert Gursky, Francis Paolini, and Bruno Rossi of American Science and Engineering Co. While they were unsuccessfully attempting to detect lunar X rays, they recognized that they had found an unexpectedly strong X-ray source in the sky. Designated Sco X-1, it became one of the first X-ray sources to be precisely identified optically. Even the early observations carried out using rockets revealed that the sky was in fact aglow with X rays. A number of satellites, including the Astronomical Netherlands Satellite (ANS), Ariel V, the 9th Orbiting Solar Observatory (\TOSO\t 9), the 3d Small Astronomy Satellite (SAS 3), and the 3d Orbiting Astronomical Observatory (\TOAO\t 3, also known as Copernicus), subsequently obtained a significant body of X-ray data. The first satellite dedicated exclusively to X-ray astronomy was \TUhuru\t, which was placed in a nearly circular orbit over the Earth's equator in 1970 from a launch platform off Kenya. The satellite reentered the atmosphere in 1979. In the late 1970s the first two of three \LHigh Energy Astronomical Observator\lies (HEAO), added many more entries to the catalog of X-ray sources, and from 1983 to 1986 the European Space Agency's X-ray satellite Exosat conducted its survey mission. The German Rosat (short for Roentgen Satellite) was launched in 1990 to map the sky at long X-ray to very short ultraviolet wavelengths. X-RAY SOURCES AND PRODUCTION MECHANISMS Among the known X-ray emitters, a large number are compact objects, whereas the remainder are either supernova remnants in our galaxy or extragalactic sources. Many have been optically identified, including a number of close \Tbinary stars\t, \Lwhite dwarf\ls, \Lsubdwarf\ls, and \Lflare star\ls. Aside from the diffuse X-ray background, about two-thirds of the sources are concentrated in the plane of our galaxy (see \TGalaxy, The\t), or in the globular clusters (see \Tcluster, star\t) associated with the galaxy. The rest lie in \Textragalactic systems\t, including radio galaxies (see \Tradio astronomy\t, \TSeyfert galaxies\t, and \Lquasar\ls.) Several possible mechanisms can produce X rays from celestial objects. Just as visible light is radiated by the Sun's 6,000-K gaseous surface, higher-energy (shorter wavelength) photons can be emitted in a thermal process by gas that is heated to a sufficiently high temperature. X rays are also produced as "bremsstrahlung," the electromagnetic radiations produced when a charged particle is deflected by a nucleus in a close encounter, thereby changing the incident particle's direction. Finally, charged particles moving at nearly the speed of light in a magnetic field emit synchrotron radiation (sometimes called magnetic bremsstrahlung), which in accelerators is observed as visible light, but which may occur in the X-ray region in astronomical systems. The mechanism that produces the diffuse background could be the inverse Compton effect, whereby relativistic electrons in space interact with microwave and optical photons, thus boosting the energy of the latter into the X-ray region. The soft, diffuse background in the energy range 0.1 to 2.0 keV is thought to be thermal emission for the most part from a local hot plasma (at approximately 1,000,000 K), including plasmas within the solar system. In summary, the emission process may be thermal, nonthermal, or both. The way in which the flux varies with energy--the energy spectrum--provides clues as to which mechanisms are involved in any particular X-ray source. For example, synchrotron radiation is theoretically expected to be characterized by a power-law spectrum, which means that a plot of the logarithm of the intensity against the logarithm of the energy follows a straight line. A vast amount of effort is being devoted to developing theoretical models that can account for all the observed characteristics of any astronomical object, including the available data over the entire spectral range. Binary Systems The brightest discrete source of low-energy X rays, Scorpius X-1 (Sco X-1), is a binary system that typifies this species of astronomical X-ray generator. One member, the X-ray source, is believed to be a compact star, whereas the other is a normal star. Gravitational energy given to the gas accreting from the primary star onto the compact star is apparently the source of energy for heating the plasma to the requisite temperature for X-ray emission. In at least two cases, Cygnus X-1 and Circinus X-1, the compact object is a candidate for a \Tblack hole\t. Most primary stars, such as Cyg X-1--the brightest discrete X-ray source, at energies greater than 70 keV--are early supergiants. On the other hand, the primary of Sco X-1 is suspected to be a red dwarf, which is difficult to identify optically. The ratio of the X-ray luminosity to the optical luminosity is important for determining the features of an X-ray source. It is large for red dwarfs and ranges from 1,500 for Sco X-1 to 0.0000001 for the source in Sirius. Both the overall brightness and the intensity of discrete emission lines of most of the binary sources are variable, as a result of the interaction of the X rays generated in the accreting disk with the nearby gaseous matter. The types of variability of compact X-ray sources include random variations, quasiperiodic oscillations, regular pulsations, and binary period variations. The periods range from about 0.01 sec for Tau X-1 (the \TCrab nebula\t, where the compact object is a neutron star) to 100 million sec for Chi Persei. A flux of 0.0000003 erg/sec-sq cm reaches Earth from Sco X-1, the strongest source of 1-10 keV X rays, which presumably represent thermal bremsstrahlung from free electron interactions in a 1 billion-K plasma. Considering that Sco X-1 is about 100 light-years away, and the measured flux is almost equal to that from the Sun, Sco X-1 must be about 10,000 billion times brighter than the Sun in the X-ray region. The seven quasi-periodic sources detected by Exosat may represent a bloblike instability in the flow of matter from the primary star onto its companion in the binaries. Supernova Remnants The fact that supernova remnants emit X rays from a large fraction of the visible nebulae was first established by Herbert Friedman and his collaborators. Detectors aboard a stabilized Aerobee rocket launched in 1968 as the Moon was about to occult the Crab Nebula showed that the X rays disappeared slowly, rather than in the manner expected for a compact source. Instead, the radiation was coming from a region covering more than a third of a light-year--in other words, from about a third of the visible nebula. Galaxies and Galaxy Clusters Astronomers believe that most galaxies, including our own, emit at least weak X rays. Some galaxies, called active nuclei galaxies, emit X rays many orders of magnitude stronger than those of normal galaxies. Strong sources of X rays among the galaxies include NGC 5128, the quasar 3C 273, and the Seyfert galaxy NGC 4151. Improvements in both the sensitivity and spatial resolution of X-ray detectors have made it possible to identify as individual components objects that had previously been lost in the general X-radiation background. For example, it has been found that X rays emanate from the gravitationally bound hydrogen-helium gas that lies among clusters of galaxies. Such intergalactic gas is heated to temperatures that can reach 100 million K. Even iron ions are collisionally excited and emit spectral lines at 6.7-6.9 keV. In addition, clusters of clusters--superclusters--consisting of 6 to 12 clusters are potent X-ray emitters. The amount of gravitationally bound hot hydrogen and helium gas in superclusters, which may be 150 million light-years in diameter, may be 20 million billion times the solar mass--5 to 10 times the mass of all the materials seen in the galaxies in the supercluster at other wavelengths. This may be the primordial material left over from the "big bang," and it may serve as a kind of "cement" that gravitationally binds the clusters into superclusters. This possibility has profound implications for the evolution of the universe. It appears that much matter has escaped detection by traditional methods--the so-called "missing" mass, the determination of which is crucial to the arguments for a finite versus an infinite universe. Undetected mass is needed to "close" the universe through gravitational collapse after about 80 billion years (see \Tcosmology\t). Bursters An event that is extreme by any standards is an X-ray burst known as a burster. In a rapidly evolving sequence, the radiation intensity increases tenfold at a rate 100 times faster than other cosmic X-ray variations. Some sources are observed only when they are bursting, although for others the much lower background emission can be detected. The first detected burster, 3U1820-30 in the globular cluster NGC 6625, was discovered in 1975 by the ANS satellite. The rise time of the burst was 0.7 sec, it lasted 2 sec, and it decayed in 8 sec. Other bursts were subsequently observed. They seem to be concentrated in the galactic disk and bulge, although some are in the galactic halo. Typically, bursters emit 10 to the power of 38 to 10 to the power of 39 erg/sec for 10 sec, so that the total energy released in an event is 10 to the power of 39 to 10 to the power of 40 erg. A given source is active and bursting roughly 10 percent of the time in our galaxy, and each recurs in a reasonably well-defined interval that is about a thousand times as long as the burst duration. The significance of bursters to X-ray astronomy is similar to that of the discovery of pulsars to radio astronomy. Bursters may be the result of nuclear explosions on the surface of neutron stars, caused by infalling material from companion stars. X-RAY DETECTORS A wide variety of detectors has been used to observe X rays. The simplest of these was a pinhole camera that photographed the Sun in X-ray "light." Photon-counting devices ranging from simple \LGeiger counter\ls to sophisticated systems using combinations of solid-state detectors have been flown on rockets and satellites. Because X rays are rapidly absorbed (for example, the intensity of 1-keV X rays is cut in half by an extremely thin section of aluminum), thin observation windows are required in the rockets and satellites. Energy resolution is attained by interposing filters in the beam or by observing pulse-height distributions from proportional counters, scintillators such as crystals of CsI and NaI that convert the ionization into light, and the more contemporary solid-state devices. Bragg spectrometers, in which reflection occurs selectively in a narrow energy band, are also used. Spectroscopy is complemented by polarimetry, which provides additional information about the production mechanism. X rays produced by nonthermal processes are polarized, whereas thermal processes produce unpolarized radiation. The polarization may be either linear or circular. The character of the nonthermal energy distribution of the radiating charged particles, and the configuration of the magnetic field that is responsible for the polarized emissions, can be deduced from determinations of the degree, direction, and energy dependence of the polarization. A classical method for observing this effect makes use of the fact that Thomson (Compton) scattering in carbon creates an anisotropy when the incoming beam is polarized. Bragg crystal polarimeters, in which reflection at 45 deg depends on the polarization, are also used. The first X-ray telescopes, such as the one launched aboard the HEAO-II satellite, could not utilize the same optical techniques that conventional telescopes use. Because of their very short wavelength, X rays cannot be easily refracted (bent) or reflected. They can be slightly deflected (grazed) and focused by a pair of crossed grazing-incidence surfaces of a paraboloid-hyperboloid shape, but the resulting images are poor and the reflecting surfaces--polished quartz coated with nickel--are both heavy and costly. Recently, however, astronomers have achieved success in designing a concave surface the refracts and reflects X rays. The surface is coated with many alternating layers (each a few atoms thick) of carbon and tungsten. High-altitude tests with a telescope designed around the new optics have achieved X-ray images of startling clarity. The long lapse between the birth of radio astronomy and the coming of X-ray astronomy occurred because the atmosphere rapidly attenuates X radiation, whereas it is transparent to radio radiation over a wide spectrum. X rays at the lower end of the spectrum can only penetrate the atmosphere down to an altitude of about 100 km (60 mi), a level that is accessible only to rockets. At the upper end of the X-ray spectrum, on the other hand, observations are feasible by balloons, which can reach altitudes of roughly 40 km (25 mi). The HEAO satellites, Exosat, and Rosat have enabled astronomers to scan the celestial sphere over the entire X-ray range with high sensitivity. The pointing precision, timing accuracy, and spectral resolution of which such orbiting observatories as Exosat have been capable bring X-ray astronomy to a status comparable to that of present-day radio astronomy. Martin A. Pomerantz Bibliography: Beatty, J.K., "Rosat and the X-ray Universe," Sky & Telescope, August 1990; Elvis, Martin, ed., Imaging X-ray Astronomy: Two Decades of Einstein Observatory Achievements (1990); Fraser, G.W., X-ray Detectors in Astronomy (1989); International Astronomical Union, High Resolution X-ray Spectroscopy of Cosmic Plasmas (1990); Sarazin, C. L., X-ray Emissions from Clusters of Galaxies (1988).