ANTIMATTER IN ASTROPHYSICS GARY STEIGMAN All matter comes in particle-antiparticle pairs. As Paul Dirac demonstrated in 1928, this is required by any theory that is consistent with quantum mechanics, the special theory of relativity and causality. Particles and their antiparticles have the same mass and lifetime. Electrically charged particles, such as, for example, the electron and the proton, have antiparticles-the positron and the antiproton-with equal but opposite electrical charges. Other, electrically neutral, particles such as the photon are their own antiparticles; such particles are called self-conjugate. Five years after Dirac's prediction, the first antiparticle-the positron-was discovered by Carl D. Anderson in 1933. There then followed a hiatus of some 22 years before the antiproton was produced and detected at an accelerator at Berkeley by Owen Chamberlain and collaborators in 1955. Subsequent accelerator experiments at higher and higher energies provided convincing conifrmation that all particles do indeed come in pairs. These experiments suggest further that particles have associated with them certain "quantum numbers" such as electrical charge, baryon number, lepton number, and so forth, which seem to be conserved in all reactions. If these conservation "laws," inferred from the experimental data, are exact inviolable ***??? - then all matter is restricted to appear (creation) or disappear (annihilation) only as particle-antiparticle pairs. That is, if an electron is created in a high-energy collision, then so must a positron be created in the same collision. In addition, an electron can only disappear if it annihilates with a positron. This apparent symmetry in the laws of physics led Maurice Goldhaber in 1956 to speculate on the antimatter content of the universe. At the same time, Geoffrey Burbidge and Fred Hoyle considered some of the possible astrophysical consequences of antimatter. In considering the problem of the amount and possible astrophysical role(s) of antimatter in the universe, it is useful to distinguish between two general categories of questions. First, we may inquire: Must the universe be symmetric? That is, do the known laws of physics require that there be exactly equal numbers of particles and antiparticles in the universe (Must the universe be symmetric?) In contrast to the somewhat philosophical nature of this question, we may pose a more practical question: Is the universe symmetric? That is, empirically, based on terrestrial experiments and astronomical observations, is there evidence that the universe contains exactly equal amounts of matter and antimatter (Is the universe symmetric?)? In addressing this latter question, it will be necessary to consider the possible astrophysical role of antimatter. SEARCHING FOR ANTIMATTER Antimatter is trivially easy to detect. For a detector, the most rudimentary device will suffice. The sample and the detector are placed in contact and, if the detector disappears (matter and antimatter annihilate on contact), the sample was antimatter. Because antimatter cannot survive in the presence of ordinary matter (atoms and molecules made of neutrons and protons and electrons), it is clear that there are no macroscopic amounts of antimatter on the Earth (antiparticles are, of course, produced in high-energy collisions at accelerators and when cosmic rays impact atmospheric nuclei). Aside from the Earth, only the solar system and the galactic cosmic rays provide a sample of the universe that may be subjected to such a direct test. The Apollo series of lunar landings and the probes to Venus provided direct evidence that the Moon and Venus are made of ordinary matter. In a somewhat less direct manner, the solar wind that sweeps through the solar system establishes that there are no antiplanets in the solar system. The reason is that, were any of the planets made of antimatter, annihilation of the solar wind particles that strike them would have turned them into the strongest gamma ray sources in the sky (gamma rays-high-energy photons-are among the products of matter-antimatter annihilations). If the solar system resulted from the collapse of a gaseous, presolar nebula, then any antimatter initially present would have annihilated before any solid bodies condensed because the annihilation rate is faster than the collapse rate. In this case, the solar system should consist entirely of ordinary matter. However, because condensed bodies of antimatter could survive indefinitely in the environment of the solar system, the capture of antiplanets could have led to the presence of macroscopic amounts of antimatter in the solar system. For this reason, the space probes and the solar wind have provided valuable-direct-information: There are no macroscopic amounts of antimatter in the solar system. The only direct sample of extrasolar system material is provided by the galactic cosmic rays (the high-energy nuclei of atoms, mainly hydrogen and helium). Perhaps the debris of exploding stars (supernovae) or the accelerated nuclei of interstellar gas atoms, cosmic rays bring samples of the material content of the Galaxy. However, as they traverse the Galaxy, cosmic rays occasionally collide with interstellar gas nuclei and may be transformed. For example, in such encounters, cosmic ray nuclei of carbon, nitrogen, and oxygen can be broken up into lighter nuclei such as lithium, beryllium, and boron. From time to time, such cosmic ray-gas collisions are sufficiently energetic to produce a proton-antiproton pair (just as in collisions at a high-energy accelerator) or an electron-positron pair (the source of the positrons discovered by Anderson in 1933). Because any antiprotons in cosmic rays may be "secondary" (produced in collisions), they do not provide an unambiguous signal for the existence of "primary" sources of antimatter (e.g., antistars) in the Galaxy. Indeed, the very small flux of antiprotons found in cosmic rays is consistent with a secondary origin. In contrast to antiprotons, no secondary antialpha particles (the nuclei of antihelium atoms) should be present in the cosmic rays. The discovery of even one antialpha particle in the cosmic rays would provide compelling evidence for the presence of macroscopic amounts of antimatter in the Galaxy. None has ever been found. From the limits to the cosmic ray flux of antialphas, it may be inferred that the fraction of antimatter in those parts of the Galaxy probed by the observed cosmic rays is less than 1 part in 10,000 (<10-~). INDIRECT SEARCHES The solar system and the cosmic rays provide the only sample of material in the universe that can be examined directly. To search further afield for antimatter, it is necessary to rely on indirect evidence. The conclusion, from the absence of antimatter in cosmic rays, that the Galaxy is not matter-antimatter symmetric, receives further support from observations of "Faraday rotation." Polarized light (or radio waves) traversing the magnetized interstellar plasma will have its plane of polarization rotated. The amount of rotation (the "rotation measure" RM) depends on the wavelength of the light and on the column density of electrons (the number of electrons per square centimeter) along the line of sight through the Galaxy. For positrons along the line of sight, the sense of rotation would be opposite. There would be no significant net Faraday rotation observed if typical lines of sight through the Galaxy intersected roughly equally many regions and "antiregions." Indeed, by comparing the RM with the "dispersion measure" DM, which depends on the sum of the electron and positron (if any) column densities, it is confirmed that N(e)-N(e+)=N(e-)+ N(e+-). Faraday rotation provides "another nail in the coffin" of a symmetric Galaxy. Because annihilation provides unmistakable evidence for antimatter, ordinary matter is an excellent probe for the presence of antimatter. Observations of the products of annihilation could provide indirect evidence for antimatter; the absence of annihilation products can lead to bounds on antimatter. The primary products of nucleon- antinucleon annihilation are pions (*******); for annihilation in nonrelativistic collisions, approximately 5 or 6 pions (equal numbers of *** and *** due to electrical charge conservation and approximately the same number of **) are produced. The charged pions decay very quickly to muons (** and *-neutrinos (********) and the muons decay to electrons, positrons, *-neutrinos, and e-neutrinos (******) The neutral pions decay primarily to a pair of gamma rays. The end products of a typical nucleon-antinucleon annihilation are -3* which carry off -1/6 of the total available energy (*****), equal numbers of ******* which carry off comparable energy, twice as many ******, which carry off 1/3 of the total annihilation energy, and -* high-energy (-200 MeV) gamma rays which carry off -*1/3 of the total annihilation energy. Neutrinos are so weakly interacting that it is unlikely that any evidence for (or against) antimatter in the universe can be inferred from annihilation neutrinos. Neither are the annihilation *** pairs of much value as a probe for antimatter. The reason is that the cosmic rays contain primary electrons as well as secondary (produced in cosmic-ray-gas collisions) electrons and positrons. It is hopeless to separate a possible annihilation component from this background. Furthermore, ** are "tied" to local magnetic fields and lose energy rapidly via Compton scattering and synchrotron radiation. Therefore, annihilation secondary ** pairs are fated to die where they are born. Indeed, it is this fact that led to proposals that the strongest extragalactic sources observed (quasistellar objects, active galactic nuclei, etc.) might be powered by annihilation. However, rather than provide a panacea for the energy woes of these sources, annihilation merely exacerbates their already strained energy budgets. The problem is that to produce the observed radio/optical radiation from **** *** * requires such enormous magnetic fields (************) that most of the energy in such sources would reside in the magnetic fields. Annihilation gamma rays provide the most useful probe for antimatter in the universe. Any site in the universe where matter and antimatter mix will be a gamma ray source. Because gamma rays are unaffected by magnetic fields and are scattered or absorbed only in environments of extremely high density (column density *** several grams per centimeter squared), the distribution of observed gamma rays yields information on their sources. In particular individual galactic and extragalactic sources can be identified and a diffuse galactic component can be separated from an isotropic (extragalactic) background. Because there are alternate (to annihilation) sources for the production of gamma rays (e.g., decay of cosmic-ray-produced **), the observations can be used to constrain the fraction, *, of the material in various environments that could be antimatter. In our galaxy, the gamma ray emissivity per hydrogen atom restricts * to less than a part in ****(*******). This tiny upper limit is not surprising when it is realized that, in the gaseous interstellar medium, an antiatom will survive for only - 300 yr in before annihilating. Indeed, for a collapsing protogalactic gas cloud, the annihilation rate always exceeds the collapse rate. Although antimatter cannot survive in a gaseous form in the interstellar medium, any condensed object made of antimatter (e.g., antistars) could survive almost indefinitely because most of the interior, is shielded from annihilation. Further more, because in such cases annihilation would be limited to the surface layers, antistars would not necessarily be strong sources of gamma rays. Still, the lack o! galactic gamma ray sources constrains the distance to the nearest antistar to be more than 10 ly. More constraining, however, is the galactic gamma ray t-background which limits the possible number of antistars to