What's wrong with globular clusters?

Globular clusters are amongst the oldest and most massive star clusters in the Universe. Their size and luminosity means that not only can we study the approximately 150 globular clusters in our own galaxy (the Milky Way) in quite a lot of detail, but we can also observe and study globular clusters in other galaxies. This is useful because globular clusters, like all types of star cluster, can provide unique insights into how galaxies form.

For many years astronomers have considered globular clusters to be examples of simple stellar populations, meaning that all the stars in them are thought to have formed at the same time and out of the same gas cloud, meaning that their initial chemical compositions were thought to be very similar. However, recent observations have shown that many globular clusters show evidence for multiple stellar populations with different chemical compositions (e.g., Gratton et al. 2012).

Colour magnitude diagram for the globular cluster NGC 2808.
Each dot represents a star in the cluster. The distribution of
dots into multiple but distinct lines suggests the presence of
multiple populations (Credit: Piotto et al. 2007).

How do astronomers know that there are multiple populations in these globular clusters? Well, if you measure the colour and brightness of all the stars in a cluster and plot their distribution then a single population of stars will form a single distribution in a narrow line, but astronomers have found that globular clusters appear to show multiple distributions.

The image on the right shows one of these plots, referred to as a colour-magnitude diagram (the magnitude of a star is a measure of its brightness), for the globular cluster NGC 2808. The stars are distributed in a narrow band, but closer inspection shows that this band is actually made up of multiple, narrower bands.

This means that the globular cluster is made up of multiple populations of stars, each with a distinct chemical signature that is different from the other populations. Astronomers can measure the chemical compositions in the different populations using spectroscopy, confirming that these discreet bands in the colour-magnitude diagram are caused by different chemical abundances.

The origin of these multiple populations aren't currently known. There are various possibilities that are being considered by astronomers, mostly involving multiple bursts of star formation within the clusters (e.g., D'Ercole et al. 2008), with the second generation of stars being chemically enriched by some process.

This then leads to the question of what could cause the chemical enrichment. There are various ideas that are being investigated, ranging from material being ejected by evolved stars, thrown off by rapidly-rotating stars, or even violent ejections by interacting massive binary stars. Astronomers are currently trying to work out which of these effects are responsible, though its a difficult task because most of this enrichment would have occurred many billions of years ago!

Understanding these massive star clusters is important because they represent some of the oldest star clusters that we can study and their formation appears to be closely related to the formation of their host galaxy.