BACKGROUND RADIATION, MICROWAVE S.P. BOUGHN AND R.B. PARTRIDGE The microwave background radiation (MBR) is a relic of a hot, early phase in the history of the universe. It fills the universe and is thus detected as a uniform background. Its discovery by radio astronomers in 1964 provided the second major piece of evidence in support of the big bang theory of the universe. The first, Edmund Hubble's discovery of the expansion of the universe, may have been more dramatic but studies of the MBR have yielded much more detailed information about the universe, ranging from the physical conditions a few minutes after the big bang to galaxy formation. The current temperature of the MBR together with observed helium abundances of very old stars have even been used to constrain the number of possible families of elementary particles, a result of major importance to particle physics. Between the wavelengths of 1 ** and 1 **** (the microwave region), the overall brightness of the sky is dominated by the MBR. Its spectrum closely approximates that emitted by a blackbody at a temperature of 2.74 k and it is very nearly isotropic; that is, it has the same intensity in all directions. These two properties of the MBR, which will be discussed in more detail, provide the primary evidence of its cosmological origin. If the radiation is left over from a hot, dense phase of the early history of the universe, the spectrum should be that of a blackbody, whereas virtually every other astrophysical source of microwave radiation has a spectrum that is quite distinct from a blackbody. Similarly, if the radiation is left over from an earlier epoch of the universe, it should fill the universe uniformly and therefore be isotropic. Relatively local foreground sources of microwaves, such as the sun or the galaxy, on the other hand, are not isotropic. By eliminating other possible sources for the MBR, astrophysicists have come to accept the cosmological Interpretation. Although the MBR is the cooled relic of thermal radiation that was present when the universe was very young (<100 s), the microwave photons observed today have not propagated directly to us from that epoch. The matter in the early universe was a dense plasma consisting primarily of electrons, protons (hydrogen nuclei), and alpha particles (helium nuclei) in thermodynamic equilibrium with the radiation. In such an environment radiation is scattered many times in a small distance, especially by free electrons, so the universe was opaque. As the universe expanded and cooled to a temperature 10,000 K, the electrons began to combine with the protons and alpha particles to form neutral hydrogen and helium atoms. However, at that epoch the density of ionized matter was still high enough to keep the universe opaque. Finally, when the temperature dropped to about 3000 K (about 300,000 years after the big bang), so few free electrons were left that the universe became transparent and the thermal radiation was able to travel more or less unimpeded from that point on. This is known as the epoch of "decoupling" or of last scattering. The peak wavelength of the radiation was then about 0.6 ** in the red portion of the visible spectrum. As the universe continued to expand, the radiation continued to cool, so that today. it is observed as microwaves with a temperature of 2.74 K. THE DISCOVERY The serendipitous discovery of the microwave background radiation by Arno A. Penzias and Robert W. Wilson in 1964 followed several "forgotten" predictions and measurements that could have led to the discovery but did not. In 1941, Walter S. Adams observed absorption lines from both the ground and excited states of interstellar cyanogen (CN) molecules. From the ratio of the intensities of these two lines, A. McKellar deduced an excitation temperature of 2.3 K. In retrospect, these molecules were almost surely being excited by the MBR and thus the preceding value constitutes a measurement of the temperature of the background radiation. However, the hot big bang model had not yet been introduced and, of course, the proper interpretation was not made. It is interesting to note that refined observations of interstellar CN absorption lines have recently yielded some of the most accurate determinations of the temperature of the MBR (see the following section). In 1946, George Gamow proposed a hot, radiation-dominated, early universe in which heavy elements were produced by neutron capture. Two of his colleagues, Ralph A. Alpher and Robert Herman, deduced that the relict radiation would today be at a temperature of 5k. Although this work was well publicized at the time, neither Penzias and Wilson nor Robert H. Dicke's group at Princeton were aware of it. In 1964, Dicke independently proposed a hot, early universe, not to produce heavy elements but rather to destroy them at the end of every cycle of a hypothesized oscillating universe. Members of his research group were already building a microwave radiometer to search for the relict radiation when they heard of the excess signal observed by Penzias and Wilson with their 7-cm wavelength radiometer at the Bell Telephone Laboratories in Holmdel, NJ.In 1965, the discovery was reported by Penzias and Wilson and interpreted by the Princeton group. THE SPECTRUM Measurements of the intensity of the MBR are indicated in Fig. 1 along with the spectrum of a 2.74 K blackbody. It is clear that from -1 mm to - 50 cm, a factor of 500 in wavelength, the data are remarkably consistent with a blackbody spectrum, which is essential to the cosmological interpretation as mentioned previously. The longer-wavelength measurements were made with ground-based microwave radiometers. At shorter wavelengths (**3 mm), the atmosphere becomes more emissive and as a consequence direct observations in this region of the spectrum must be made with high-altitude balloon-borne rocket, or satellite detectors. It should be noted that one of the most accurate determinations was deduced from interstellar CN absorption at 2.6 mm. They appear to be ruled out by measurements made by the COBE satellite and reported in 1990. The first quantity of cosmological significance coming from these measurements is simply the temperature of the radiation, which is T= 2.74**0.02 K (an accuracy rare in cosmology). Initially, in 1965, it was an estimate of the baryon density in the universe and the primordial helium content that allowed Peebles to estimate the current temperature of the MBR. Now that the temperature is so well determined, this argument can be reversed to yield a value for the baryon density of the universe, a crucial cosmological datum. A similar calculation involving T and the abundances of light elements limits the number of possible neutrino families to **, an important result for particle physics (this limit could be made tighter still if the lifetime of the free neutron were known to higher precision). Another cosmological parameter that comes directly from the value of T is the ratio of the number of microwave photons to baryons, or entropy per baryon in the universe. This ratio is on the order of ****, and until recently there was no compelling explanation for it. It is regarded as a considerable success of the new grand unified theories of elementary particles that they make possible a natural explanation of the value of this parameter. ISOTROPY Within a few years after the discovery of the MBR, it was shown to be isotropic to within less than 1%. Although this isotropy was an essential test for a cosmological interpretation as mentioned previously, astrophysicists knew that the MBR had to be anisotropic at some level and began immediately to search for such anisotropies. It is generally accepted that, on the scale of galaxies and larger, the lumpiness of matter in the universe is due to aggregation by gravity. If this is so, the mass in the universe at the epoch of decoupling cannot have been perfectly smooth but rather must have exhibited small fluctuations that would later grow in amplitude to become galaxies and clusters of galaxies. A microwave map of the sky is a snapshot of the universe at this epoch more than 10 billion years ago and thus provides us with a unique opportunity to see these "gravitational seeds" from which galaxies grew as spatial fluctuations in the MBR intensity. Dozens of isotropy measurements have been made with many different instruments, including ground-based radiometers only a few centimeters across, the 1 mile Very Large Array in New Mexico, a variety of high-altitude balloon-borne radiometers, and even radiometers operating aboard satellites. Because of diffraction effects in radio telescopes, the smaller instruments can only be used to search for large-scale anisotropy, whereas the very large telescopes are capable of investigating very small scales. There is still no convincing evidence that any deviation from isotropy has been detected with the exception of the "dipole anisotropy," which will be discussed in the following paragraph. Figure 2 is a map of nearly the entire sky made with a balloon-borne radiometer with a ruby maser receiver operating at a wave-length of 15.6 mm. Figure 2 is plotted in galactic coordinates and the strip across the center of the map is the microwave emission from our own Milky Way. The 2.74 K isotropic part of the MBR has been subtracted, so only the deviations are represented in the map. Signal levels correspond to temperatures that range from -********** of the MBR. Aside from the plane of the Milky Way, the only other obvious feature in the map is the "dipole anisotropy" which is generally characterized by the hot area to the right and cold area to the left (the remaining bumpiness in the map is due to instrumental noise). It is generally accepted that this feature is due to the motion of the Earth with respect to the "comoving frame" of the universe in which the MBR is at rest. The MBR is Doppler-shifted to shorter wavelengths in the direction of the Earth's motion and thus appears hotter, whereas in the direction opposite to the Earth's motion the radiation is shifted toward longer wavelengths and appears cooler. This anisotropy certainly has nothing to do with the "gravitational seeds" out of which galaxies form, but it is of fundamental cosmological importance nonetheless. From the magnitude of the "dipole anisotropy" (about 0.1% of T) and the motion of the Earth with respect to the Local Group of galaxies, it is found that the Local Group as a whole has the rather large velocity of 600 km s - 1. This number is being actively exploited to tell us about the large-scale distribution of matter in the universe, because the velocity of the Local Group seems to be at least in part produced by the gravitational acceleration caused by nearby clumps of matter (such as the Virgo cluster of galaxies). Such arguments can also be used to set constraints on the density parameter. There is no convincing evidence of any other anisotropy of the MBR. The upper limits on the deviations from isotropy on angular scales from an arcminute to 90ø are on the order of or less than 1 part in ****. However, even these null results have provided important evidence in evaluating cosmological models and theories of galaxy formation. For example, the limit on the "quadrupole anisotropy" (90ø scale\ places important constraints on anisotropic expansion of the universe and, in turn, provides strong support for the isotropic Friedmann model of the big bang. Limits on anisotropy on the scale of a few arcminutes are already in conflict with many standard models of galaxy formation (especially those not including some form of "dark matter") and such models are no longer considered viable. The lack of a measurable anisotropy on the scale of several degrees and larger has caused astrophysicists discomfort for some time. According to the standard big bang theory, the microwave radiation incident from parts of the sky that are further than a few degrees apart was emitted (at the time of decoupling) from regions of the universe that were not causally connected; that is, there had not been time from the beginning of the universe for light signals to have been propagated between them. How then could these regions possibly have the same temperature to within less than 1 part in ***? It is considered one of the strengths of the inflationary theory of the universe that it offers a natural resolution to this puzzle: The present universe as a whole was inflated from a tiny causally connected region very early in the history of the universe. FUTURE DIRECTIONS Although no investigations have yet found any asymmetry of the MBR (other than the dipole), astrophysicists are very optimistic. Satellites including NASA's COBE mission that was launched in November 1989 and the Soviet Union's Relict II mission to be launched in the 1990s promise to improve the sensitivity to largescale anisotropy, whereas several ground-based missions seek to improve the situation at smaller scales. The goal is a sensitivity of ****** of T. Astrophysicists agree that the lumpiness of the MBR must be larger than this level if we are to preserve our most basic ideas about cosmology and galaxy formation. Additional Reading Dicke, R.H., Peebles, P.J.E., Roll, P.G., and WilKinson, D.T. (1965). Cosmic blackbody radiation. Ap. J. 142 414. Ferris, T. (1983). The Red Limit. Quill, New York. Harrison, E.R.(1981). Cosmology: The Science of the Universe. Cambridge University Press, Cambridge. Mather, J.C. et al. (1990). Ap.J. Lett. 354 L37. Penzias, A.A. and Wilson, R.W.(1965). A measurement of excess antenna temperature at 4080 Mc/s. Ap. J. 142 419. Silk, J.(1989). The Big Bang. W.H. Freeman, New York. Weinberg, S.(1984). The First Three Minutes: A Modern View of the Orgin of the Universe. Bantam Books, New York. Wilkinson, D.T.(1986). Anisotropy of the cosmic blackbody radiation. Science 232 1517. Wilkinson, D.T. and Peebles, P.J.E.(1989). Discovery, of the *** radiation. The Cosmic Microwave Background: 25 Years Later, N.Mandolesi and N.Vittorio, eds. Kluwer Publishing Company, Dordrecht. See also Cosmology, Big Bang Theory; Cosmology, Inflationary Universe; Cosmology, Observational Tests; Galaxies, Local Group.