Kyle M. Cudworth
Globular clusters were originally defined as rich, compact, nearly spherical, groups of hundreds of thousands (or even millions) of stars. Work in the past few decades has shown that the stars in globular clusters are among the oldest stars in the Galaxy, with ages greater than 1010 years. As a consequence, for most astronomers the working definition of a globular cluster now concentrates more on the age of a cluster than on its richness, and a few are quite sparse. About 140 of these clusters are now known in our Milky Way galaxy. The brightness and distinctive appearance of globulars make them relatively easy to detect at large distances (except in directions where dust very severely absorbs starlight), so it is likely that most that exist in our Galaxy have been discovered. Furthermore, globular clusters are found in the galactic halo, well above and below the thin disk of the Galaxy that contains most stars and the younger open clusters. (The galactic halo should not be thought of as a shell, but rather as a roughly spherical volume of space within which globular clusters and some old stars are found.) While globulars are strongly concentrated toward the center of the Galaxy, some are found at very large distances from the galactic center. These characteristics are major reasons why globulars are key objects for the study of distant parts of the Galaxy. Not surprisingly, globular clusters are also seen in and around other galaxies.
The stars in globular clusters have been found to differ in chemical composition from most stars in the galactic disk in that globular cluster stars are depleted in heavy elements (metal poor) by factors ranging from at least 2 up to 200. In most clusters all stars have very similar chemical compositions, but the composition differs from cluster to cluster.
Because the spatial distribution and chemical composition of the globulars are thus distinctly different from those of most stars, these clusters reveal a different aspect of galactic structure than ordinary stars, and because the clusters are the oldest identifiable objects in the Galaxy, these differences contain information regarding the formation and early evolution of the Galaxy. Globular clusters thus provide much, probably most, of the basic observational data on which any understanding of the early epochs of our Galaxy must be based. We seek to learn more about these early stages in our Galaxy's history in order to understand how the Galaxy came to have its current structure and other characteristics, and we expect that much of what we learn about our Galaxy's history will also be applicable to other galaxies as well.
In an early (1915-1919) use of globular clusters in what we would now call a study of galactic structure, Harlow Shapley derived distances for many globulars, and found that the distribution of globulars was centered at about 15 kpc (kiloparsecs, 1 kpc = 1000 pc = approximately 3260 light-years) away from the Sun in the direction of the constellation Sagittarius. Shapley based his cluster distances upon the brightnesses of individual stars in each cluster when possible, and on the size and brightness of each cluster as a whole when individual stars could not be studied. Since many of the globulars that Shapley studied are out of the dusty plane of the Galaxy, the distances that he found were not too severely affected by the lack of a correction for the absorbing affects of dust. Shapley went on to argue that such massive objects as globular clusters were most likely to be centered around the galactic center, and that thus the center of the Milky Way was 15 kpc away from the Sun toward Sagittarius. Similar studies in the 1970s and 1980s with absorption corrections and with much better data from which to determine cluster distances have yielded distances to the center of the distribution of about 8 instead of 15 kpc.
While Shapley's conclusions remained controversial for a few years, they were eventually accepted by essentially all astronomers, and this technique is still considered one of the primary means of determining the distance to the center of the Galaxy. This classic work by Shapley was crucial in that it was the first study of the structure of our Galaxy to indicate a center well away from the Sun. This shift away from a heliocentric (Sun-centered) Galaxy may even be likened to the Copernican shift away from a geocentric (Earth-centered) solar system (and universe) that had occurred a few centuries earlier, but the shock from Shapley's studies was significantly less severe from either an astronomical or a philosophical/religious point of view.
In the 1940s, Walter Baade developed the concept of stellar populations to describe the differences between the stars commonly found in the thin Galactic disk (Population I) and those distributed spherically about the galactic center (Population II). Globular clusters were and are the primary example of Population II, or the halo population. We now understand the fundamental difference between these populations to be age, and that the differences in the types of stars, their chemical composition, and their distribution in the Galaxy, are all related to their ages. Further work has broken the division down into more populations, including a distinction between extreme Population II and intermediate Population II that was noted in the 1950s and has been reemphasized in the 1980s.
Most globular clusters are in the extreme Population II or halo class, having stars metal-poor relative to the Sun by factors of at least 6 or 7 (and in most cases greater than 10), and being distributed essentially spherically about the galactic center. Halo clusters are often found as far as 10 kpc or more above or below the galactic plane. These clusters, like everything else in the Galaxy, are in orbit around the galactic center. Although many millions of years are required for a cluster to complete one orbit, we can learn a good deal about the shapes of the orbits from the current locations and velocities of the clusters. Such investigations have revealed that the orbits of halo globulars are not at all circular, but are generally quite elongated, and are oriented essentially at random. Clusters in such orbits participate only slightly, if at all, in the general rotation of the Galaxy that is the dominant motion of Population I stars in the galactic disk Some halo globulars even show a motion opposite to galactic rotation.
In contrast to the halo objects, about 20% of the globular clusters in our Galaxy are less metal poor and are found within about 1 or 2 kpc of the galactic plane (compared to most Population I stars lying within 0.4 kpc of the plane), and thus belong to intermediate Population II, often called the thick-disk population. These clusters are also in orbit around the galactic center, of course, but their orbits are more nearly circular and are oriented near the galactic plane. These clusters are moving in the same direction as galactic rotation, though at a rate that is slightly slower than the rotation of the thin disk of Population I stars.
Nearly all of the thick-disk globulars lie closer to the galactic center than does the Sun (i.e., within about 8 kpc of the center), but halo clusters are found out to much larger distances, though also with a strong concentration toward the center. Globular clusters of both types are thus found in large numbers in the general direction of the galactic center, but only halo clusters are found well away from the center. An individual globular can usually be assigned unambiguously to one population or the other on the basis of its chemical composition or velocity.
All of these differences presumably reflect changes that happened toward Sagittarius. Similar studies in the 1970s and 1980s with in the Galaxy very early in its history, as stars and star clusters absorption corrections and with much better data from which to determine cluster distances have yielded distances to the center of the distribution of about 8 instead of 15 kpc.
While Shapley's conclusions remained controversial for a few years, they were eventually accepted by essentially all astronomers, and this technique is still considered one of the primary means of determining the distance to the center of the Galaxy. This classic work by Shapley was crucial in that it was the first study of the structure of our Galaxy to indicate a center well away from the Sun. This shift away from a heliocentric (Sun-centered) Galaxy may even be likened to the Copernican shift away from a geocentric (Earth-centered) solar system (and universe) that had occurred formed first from a large, nearly spherical protogalactic cloud of gas. This cloud would have contracted under its own gravitational pull and its small initial rotation would have sped up as the cloud contracted and caused the contraction to end in a disk rather than a small sphere. According to a commonly accepted picture, stars and clusters that formed a little later would have been formed in a thick disk and would have a higher metal abundance, as the chemical composition of the gas from which stars could form was enriched by the products of nuclear reactions in the most massive of the earliest stars.
A major issue in current research on globular clusters is the question of whether there are significant age differences from cluster to cluster, and especially between the two populations of globulars. Such age differences could tell us how long it took the Galaxy to contract from the original spherical cloud to a thick disk, and how long the thick-disk phase lasted. It presently appears that the difference in ages is no more than a few billion years (out of a total age of about 15 billion years), implying fairly rapid changes in the early history of the Milky Way. Our instruments for obtaining appropriate data, and our ability to infer reliable ages from those data, are both improving rapidly, however, so these conclusions regarding ages and age differences could change in the next few years.
As noted earlier, halo globulars exist much further from the galactic center than the Sun's position. The outer limit for normal halo clusters seems to be about 30-40 kpc from the Galactic center, though a few globulars are also found between about 60 and 100 kpc from the center. Most of these outer halo clusters are less dense than typical globulars and most show some peculiarities in their constituent stars that may indicate slightly younger ages. A few dwarf spheroidal galaxies are also found at distances similar to those of the outer halo globulars. These systems contain old metal-poor stars similar to those in globular clusters, and have masses comparable to the most massive globulars, but are different in that the star density within each system is very low compared even to the low density outer halo globulars.
It is generally believed that the outer-halo globulars and the dwarf spheroidal galaxies are in orbit around the Milky Way, and are not merely passing by. As they are the most distant objects associated with our Galaxy, their motions should reflect the effects of the gravitational pull of the entire mass of the Milky Way. Observations of their velocities in recent years have thus yielded some of the best available derivations of the mass of our Galaxy: about 5 × 1011 times the mass of the Sun. Since this is considerably more mass than can be accounted for by adding up the masses of all the stars and gas clouds we observe directly, this is part of the evidence for unseen dark matter (sometimes called missing mass) in the Galaxy.
Because of their importance in questions of galactic evolution and the study of the outer parts of the Galaxy, globular clusters continue to be investigated intensively by many astronomers today. Questions such as the distance scale, chemical compositions, and ages of globulars are all interrelated and are receiving considerable attention from both observational and theoretical perspectives. We can thus look forward to increasing knowledge and, one hopes, a better understanding of these issues in the coming years.
Armandroff, T.E.(1989). The properties of the disk system of globular clusters. Astron. J. 97 375.
Bok, B.J.(1981). The Milky Way galaxy. Scientific American 244 (No. 3) 92.
Freeman, K.C. and Norris, J.(1981). The chemical composition, structure, and dynamics of globular clusters. Ann. Rev. Astron. Ap. 19 319.
Harris, W.E. and Racine, R.(1979). Globular clusters in galaxies. Ann. Rev. Astron. Ap. 17 241.
King, I.R.(1985). Globular clusters. Scientific American 252 (No. 6) 78.
Zinn, R.(1985). The globular cluster system of the Galaxy. IV. The halo and disk subsystems. Ap. J. 293 424.
Adapted from The Astronomy and Astophysics Encyclopedia, ed. Stephen P. Maran