Imagine the Universe!

What Are Clusters Made Of?

Cluster Mass Distributions, Dark Matter, and the Fate of the Universe

Astrophysicists divide the universe up into two types of matter: "luminous matter" -- stars and gas that shine of their own accord, and "dark matter" that is detectable only through its gravitational effects on luminous matter. One of the outstanding problems in modern cosmology is to measure the total amount of material, both luminous and dark, since this determines the ultimate fate of the expanding universe: if the mass density is above a certain `critical' value the universe will eventually recollapse (perhaps as part of an eternal cycle of big bangs and `big crunches'), if not it will expand forever. The sum of all the known luminous matter adds up to less than 10f this critical density, but we know from studies of our own, and other nearby, galaxies that there is more dark than luminous matter in the universe. But how much more?


This is where the study of rich clusters of galaxies comes in. These conglomerations of hundreds, or even thousands, of galaxies spread over millions of light years are so large that the relative proportions of dark and luminous matter are believed to be the same as the universe as a whole. Using X-ray observations, we can measure the masses of both the hot gas (this pervasive medium lying between the galaxies, and not the stars in the galaxies themselves, is the dominant form of luminous mass in clusters) and dark matter.

The amount of hot gas in a cluster is simply related to the total X-ray luminosity emitted (and collected by telescopes on board orbiting X-ray observatories) -- the more hot gas there is, the brighter in X-rays. Measuring the luminous matter is a fairly direct task. The dark matter can be measured, or at least estimated, because clusters of galaxies are approximately `relaxed' systems: these systems have been around long enough to have attained a balance, an equilibrium, that depends only on the present-day structure of the clusters and not how they were formed or on recent dynamical processes. In the same way, the basic stable structure of the Earth's atmosphere today depends only on the Earth's gravity, atmospheric composition, solar heating, etc. and not on how the earth formed or what today's weather is like. And just as the pressure in the Earth's atmosphere is in balance with the gravity of the Earth -- decreasing as one ascends to higher and higher altitudes -- so too is the hot X-ray emitting cluster gas stratified in pressure in such a way as to be in balance with the total gravity of the cluster. The pressure of the hot gas in clusters can be measured - - it is simply related to the product of the hot gas mass density and temperature. The density is inferred from the X-ray brightness (the higher the density of X-ray emitting gas, the more intense the X-ray emission), while the temperature is inferred from the X-ray spectrum (every temperature produces a characteristic spectral shape).


We then infer the total mass distribution from the measured pressure profile by applying the condition of equilibrium. Not surprisingly -- based on what we already know about individual galaxies -- we have found that most of the mass in galaxy clusters consists of dark matter. The total mass in a rich cluster can exceed 1048 grams (as much as a million trillion Suns). However, there have been a couple of surprising developments that have come out of our cluster studies.

Surprising Developments

Modern physics `theories of everything' that attempt to unite the four forces in nature predict that the total mass density in the universe should be just equal to the `critical density' defined previously. This corresponds to roughly 90-99% of the material being dark. While most of the matter in rich galaxy clusters is dark, we find it is so only by about a three-to-one margin, not a ten- or hundred-to-one margin, as expected. This means that if rich clusters are representative samples of the universe -- and there's no reason to suppose otherwise -- then the universal mass density is less than the critical value, and the universe will expand forever.

Another unexpected discovery is that for some (but not all) moderately rich clusters, clusters with a few dozen galaxies, the dark matter fraction is higher -- around 90%. Why these clusters are different than poorer clusters may be telling us something. The most straightforward explanation of this is that some of the luminous matter (that is, the hot gas) has been expelled from these systems, and therefore the relative proportion of dark matter is greater. While not as massive as the richest clusters, the mass in these systems can exceed 1047 grams, and therefore expelling a large fraction of their hot gas requires a tremendously violent and energetic process. As it turns out there is other X-ray astrophysical evidence for such phenomena obtained from measuring the amount of chemical pollutants in the hot cluster gas.

Chemical Abundances in Clusters: Star Formation and Supernova History

Before stars or galaxies formed, the luminous matter in the universe consisted of only the lightest and simplest elements -- mostly hydrogen and helium. Most of the hot X-ray emitting gas we observe consists of this primordial material, but X-ray spectroscopy reveals emission lines from other heavier elements, most universally an iron line at 6.7 keV. Some clusters also show spectral emission lines from atoms of oxygen, neon, magnesium, silicon, and sulfur. Where did these chemical `pollutants' come from and how did they get out into the gas between the galaxies in clusters?

The most common heavy elements are synthesized in nuclear reactions in the hot, dense deep interiors of stars -- some elements steadily as part of a star's natural slow evolution, others (including those we see in X-ray spectra of galaxy clusters) in explosive events called supernovae. These heavy elements escape out of stars through various means -- in supernovae the explosions themselves blow the star apart. However, to get out into the cluster gas the material must escape not only from a single star but out of the entire galaxy of stars. This can occur if many supernovae explode in a sufficiently short period of time, leading to what is known as a galactic wind.

The strength of any particular X-ray emission line is directly related to the abundance of the element that produced that line, and by measuring all the heavy elements in the spectrum we can estimate how many supernovae occurred. It turns out that the number is unexpectedly high -- higher than one would predict based on the stars we see. This has several consequences. Firstly, it means (again if clusters are representative of other places in the universe) that there has been more massive star formation (only stars at least eight times as massive as the Sun explode as supernovae) in the universe than was previous thought. The energy associated with all of these supernovae is so great that it may have not only blown gas out of individual galaxies, but out of some (but not the most massive) clusters as well -- thus offering an explanation for our detection that some clusters have higher dark matter fractions than others, as discussed above.

Thus our observations that, except for the richest systems, clusters show a range of relative proportions of dark and luminous matter; and, our measurement of large amounts of heavy element enrichment of the gas between galaxies in all clusters find a common explanation. At the time when the stars in cluster galaxies formed, there must have been hundreds of billions of supernova explosions that expelled material out of galaxies. And such powerful injections of heavy metals and energy into the cluster gas both enriched it and, in some cases, expelled some fraction of the material out of the clusters despite the fact that each one is about one thousand times more massive than a typical galaxy consisting of ten billion stars.

What X-ray Missions Are Used in Cluster Research?

In order to carry out these areas of research, we need both images and spectra of the X-ray emitting gas. Of course, we want the best spatial and spectral resolution to give us the best possible observables and the most robust data to compare with model predictions.

The best X-ray images, that is maps of X-ray emission, currently available come from the ROSAT satellite. We use ROSAT observations to derive cluster gas mass distributions.

The most useful spectra for cluster work come from the ASCA satellite, because they yield the most detailed spectra over a wide range of X-ray energies. Our group uses ASCA spectra to get the most accurate temperatures and elemental abundances.

We use a combination of the mass density (from ROSAT) and temperature (from ASCA) profiles to find the pressure distribution and therefore the total cluster mass distribution (dark and luminous matter).

What's In The Future?

New missions, such as CXO, and future missions, such as ASTRO-E and XMM, will improve the detail and quality of cluster images and spectra. The major benefit of these increases in capabilities will to be expand the number of clusters whose mass distributions and elemental abundances are known, and to study clusters that are further away and therefore younger. This will not only enable us to refine the above results, but to explore systematic variations and address the questions of why only certain clusters seem to have lost a lot of their hot gas. Looking back to earlier times will help us to trace out the enrichment history and thus the evolution of, not only the clusters themselves, but star formation in the universe as a whole.

Thank you to Michael Loewenstein for contributing to this article.

If words seem to be missing from the articles, please read this.

Imagine the Universe! is a service of the High Energy Astrophysics Science Archive Research Center (HEASARC), Dr. Alan Smale (Director), within the Astrophysics Science Division (ASD) at NASA's Goddard Space Flight Center.

The Imagine Team
Project Leader: Dr. Barbara Mattson
Curator: Meredith Gibb
Responsible NASA Official: Phil Newman
All material on this site has been created and updated between 1997-2014.
This page last updated: Friday, 12-Jan-2007 10:16:19 EST