The Fault in our Star (Populations): Solving the Multiple Stellar Populations Problem with Dark Matter

Title: Dark Matter as a Possible Solution to the Multiple Stellar Populations Problem in Globular Clusters

Authors: Ebrahim Hassani, Seyyed Milad Ghaffarpour Mousavi

First Author Institution: University of Birjand, Birjand, Iran

Status: Submitted for publishing to MNRAS Journal [pre-print available on arxiv]

The more we uncover about the universe, the more mysteries we unearth. In today’s paper, we’ll explore one of the many current questions plaguing astronomers: the multiple stellar populations problem.

Globular clusters are large gatherings of old stars that are packed closely together in a spherical shape. While there is still a lot we don’t understand about the formation of globular clusters, it’s currently believed the stars within all form from the same giant molecular cloud. Within these clouds are dense regions of gas that begin to clump together to form stars. Because they all evolve from the same environment, it stands to reason that all of the stars within a cluster would have similar ages and metallicity (i.e. amounts of metal). Yet observations of globular clusters have revealed this is not the case – we have actually observed multiple stellar populations inside of globular clusters. Stellar populations here refer to stars with the same age and chemical properties; so we are seeing groups of stars with distinct metallicities. While different potential solutions have been theorized, the authors of today’s paper suggest dark matter as the latest topic to explore.

Traditional stellar evolution

The Hertzsprung–Russell diagram (shortened to H-R diagram) shows the traditional way that stars evolve; it charts the temperature of a star (it’s astronomical color B-V) versus its luminosity, or brightness (absolute magnitude V). In the case of a globular cluster, it’s anticipated that stars would follow a narrow evolutionary path along the H-R diagram. Yet, most spectroscopic studies of globular clusters show at least two paths for stellar evolution. 

Figure 1: H-R diagram of ω Centauri GC (or NGC 5139, the largest known globular cluster of our Galaxy). Stars with different [Fe/H] values are color coded. Three separate stellar populations with unique [Fe/H] values are detected in the GC. Figure 1 in paper.

The diagram pictured above shows three distinct generations of stars with different chemical compositions; the [Fe/H] values indicate the abundance of iron detected in the star; aka its metallicity. Clearly, these stars are evolving differently. But how? This brings us to dark matter.

Dark matter: the story so far

When it comes to studying galaxies, we are only seeing one small piece of the galactic puzzle – a huge portion of our universe is made out of matter we cannot see! In the 1960s – 1980s, observations of the rotation of spiral galaxies brought up an interesting question regarding the system’s mass. We can get a general estimate of the stellar mass of galaxies by adding up all of the light we receive and converting that into its equivalent stellar mass. The mass of a system can also be used to predict its speed; for spiral galaxies, it can be as simple as applying the virial theorem. (This can be done for non-spiral galaxies as well, as explained in this bite) However, observations of spiral galaxies revealed that stars far from the center were moving much faster than should be possible given its mass. Ergo, there was much more mass in these galaxies that weren’t emitting light! This mass-to-light ratio today helps us to estimate the quantity of dark matter; a larger mass-to-light ratio indicates a larger abundance of dark matter.

It’s been theorized that stars in the presence of dark matter evolve differently than those not surrounded by dark matter. Let’s consider the main sequence phase of stars. The main sequence is a stage in a star’s lifetime in which it is converting hydrogen to helium via nuclear fusion in its core. This process is a function of temperature. If dark matter particles were to annihilate in stars, this would change the temperature at their core, and in turn affect the rate at which stars consume hydrogen. In this case it would be fair to suggest that the presence of dark matter would alter not only the temperature of stars, but also its chemical composition; stars would spend less time in certain stages, creating populations that have more/less metallicity. From this reasoning, today’s authors theorize that the denser the region of dark matter surrounding a star is, the more extreme deviations we would see from the standard stellar evolutionary path.

The simulations

The authors of today’s paper calculated the rate at which stars would capture dark matter by using MESA, an open-source code used to simulate stellar evolution. Figure 2 below shows the traditional path stars evolve as a thick blue line, where the red lines correspond to stellar evolution for a star with a fixed mass in environments with different amounts of dark matter. Figure 3 keeps the amount of dark matter fixed, and instead varies the mass of the target star. 

Figure 2: Stellar evolution of a star the mass of our sun as a function of temperature and luminosity. The blue line represents the standard stellar evolution in an environment with zero influence from dark matter (ρχ = 0). The red line in graphs (b)-(f) represent stellar evolution in environments with increasing dark matter density (ρχ ≠ 0). Figure 2 in paper.

Figure 3: Stellar evolution of stars with different masses on the H-R diagram. In each graph, the thick blue line represents the standard evolution in an environment with zero influence from dark matter (ρχ = 0), and the red line represents the stellar evolution at a fixed non-zero dark matter density. The mass of the star varies, increasing from 0.5x solar masses in (a) to 1.0x, 1.5x, 2.0x, 2.5x, and 3.0x solar masses in the following graphs. Figure 3 in paper.

The simulations shown above suggest that stellar evolution deviates from the expected path when under the influence of dark matter. If dark matter particles annihilate inside stars they provide a new source of energy within stars, creating extra luminosity and affecting its place on the H-R diagram. Higher dark matter densities correlate with larger deviations. 


It’s possible that dark matter is not distributed evenly throughout a globular cluster. Observations of ω Centauri GC reveal that the central region has a higher mass-to-light ratio than the average mass-to-light ratio of the entire cluster, implying that dark matter is more abundant towards the center. In that case, there will be regions within a galaxy that are more dark matter-dense, and stars that are therefore in contact with more dark matter. Simulations of these environments reveal that stars that are in higher concentrations of dark matter have larger deviations from the standard stellar evolution. So we would therefore see multiple stellar populations in a non-uniform distribution of dark matter. 

There is still so much we have yet to understand about dark matter. Because we currently lack information on the precise physical nature of dark matter, it is hard to come to a solid conclusion on how they interact with stars. The argument posited here hinges on dark matter being able to affect stars in a significant way.  If the theory discussed here is correct, in that dark matter can alter the luminosity and chemical composition of stars, the multiple stellar populations problem could be explained by the non-uniform presence of dark matter. 

Astrobite edited by Graham Doskoch

Featured image credit: Wolfgang Kloehr/

About Ciara Johnson

Ciara Johnson is currently completing her masters in Science Communication at the University of Leeds. She has 5 years of experience working with planetariums and science centers, and has a research focus in co-production in museums with underrepresented groups in STEM.

Leave a Reply