Pretext

The inner regions of the Milky Way Galaxy, when observed in low-energy gamma-rays, reveals a sharp emission line at photon energy 511 keV, which has been observed since the 1970s. A closer look at the energy tells you that it is exactly the mass of an electron. In fact, we know that the 511 keV line in the galaxy is caused because low-energy positrons ‘annihilate’ with low-energy electrons in the Interstellar Medium (ISM) to produce a pair of gamma-ray photons, each of energy 511 keV. The estimated number of such annihilation events in the Milky Way Galaxy is \(5\times 10^{43}\) every second. In more meaningful units, this corresponds to an energy emission that is 10,000 times as compared to that by the Sun.

This enormous annihilation rate also means that an incredibly large number of positrons are produced in the Galaxy (the number of electrons is not a problem, given that most of ordinary matter contains electrons) every second. Identifying such sources astrophysically is difficult, not just because the rate is so high, but also because the emission comes from a completely unexpected region. While most of the suspected sources of positrons reside in the planar disk of the Galaxy, a significant amount of 511 keV emission is produced by a roughly spherical region called the ‘bulge’ that extends both above and below the disk.

Morphology of the 511 keV line emission in the Milky Way Galaxy. Credits: ESA (G. Weidenspointner et al.).

The most plausible source of positrons in the interstellar medium is \(\beta\)-decay of radioactive nuclei, which are produced in stellar explosions such as supernovae. However, one would expect supernovae to have occured at places where stars are, because afterall they are end-products of massive stars. Most of the star-formation activity of the Milky Way takes place in the Galactic disk, which should then be expected to be the main region of positron annihilation. Now, although most of the 511 keV emission in the Galactic disk can be explained using positrons from supernovae, emission of 511 keV from the bulge (which contributes to nearly 40% of the total emission) evades such an explanation because of the lack of strong supernova activity in the bulge. It is worthwhile to note that the bulge is populated mostly by some of the oldest remaining stars in the Galaxy.

Explaining the 511 keV line in the bulge

There have been several attempts to explain the observed excess of 511 keV emission in the bulge by introducing a source of positrons. For example, Crocker and collaborators in 2017 showed that thermonuclear supernovae (also called Type Ia supernovae, which are super-important in modern Cosmology), which are expected to form in old stellar populations such as in the bulge, can explain the 511 keV emission in the bulge, if present in sufficient numbers. Similarly, pair production of electrons and positrons in dense and energetic jets of compact astrophysical objects such as pulsars, microquasars and neutron star binaries has also been proposed as a source of these positrons.

Researchers have also attempted to explain this anomaly using models beyond known astrophysical sources. One such model includes annihilation or decay of dark matter particles into electron-positron pairs, given that the mass of this dark matter particle is sufficiently higher than the electron mass. Similarly, Hawking evaporation of black holes with masses around \(10^{16}\) g (also called primordial black holes because black holes of these masses cannot be produced by stellar collapse, since stars are way heavier than this mass) can produce positrons in sufficiently high numbers in the bulge. Similarly, a dark matter particle transitioning to its ground state from an excited state can produce positrons and electrons from the emitted energy, if the energy gap is again large enough. The distribution of dark matter in the Galaxy is expected to be shaped roughly like the Galactic bulge, and hence these models have been invoked to explain the 511 keV line in the bulge.

There is, however, a strong constraint associated with positron production from dark matter (or any pair production process in general). Our theory of electromagnetic interactions predicts that whenever a process produces a pair of electron and positron, there is a small chance of producing a photon alongside the pair. This phenomenon is termed ‘internal bremsstrahlung’ or ‘final-state radiation’. The emission of this photon would appear in observations of the ‘continuum’ emission in the Galaxy, that is, as an excess emission at energies away from the 511 keV line. Even without pair production of electrons and positrons, one would still expect a broad continuum emission, if positrons are produced initially with high energies. This process, called ‘in-flight annihilation’, takes place when a high-energy positron annihilates with a low-energy electron, and produces a continuum of gamma-rays instead of a line. A line is produced when both the positron and electron are at low energies. Observations of the gamma-ray continuum at MeV energies have not found evidence of such continuum emission from internal bremsstrahlung or from in-flight annihilation and thus, this imposes severe limits on how large the injection energy can be. Previous studies have found that these positrons are injected at most at around 7 MeV energies; anything higher than that would overestimate the continuum emission compared to observations.

Gamma-ray continuum emission from high-energy positrons would over-produce the observed continuum emission if the positrons were injected at very high energies. This can be used to put an upper limit on the injection energy of positrons at a few MeV. Credits: Beacom, Bell and Bertone (2005)

Cosmic Ray Propagation

Details of the work

More about this project

I started doing this project after I completed third year of my undergraduate studies. I had received the Future Research Talent award from the Australian National University. This was the first time since COVID that the program was open for students in-person. I applied through my University, Indian Institute of Science, and got selected for the award, along with 7 others from my institute and nearly 100 from other Universities across India and Indonesia. I was working at the Mount Stromlo Observatory near Canberra, Australia, a 30-minute drive from the city center. The observatory had been burned in wildfires, but is now the home of the Research School of Astronomy and Astrophysics of the University.

I worked for 3 months as a visitor at the Observatory under the supervision of Prof. Mark Krumholz and Prof. Roland Crocker. This was the very first project where I got to work with real astrophysical data. Some of the skills and resources that helped me in this project are as follows:

• Coding: I had a preliminary knowledge of python, for example, plotting, matrix calculations and a little bit of statistics. I had to learn a whole lot of new concepts, which included running scripts, parallel computing, advanced statistics, interpolation etc. A particularly useful programming feature that immensely helped with analysing large datasets was Object-Oriented Programming.
• Data science: Several concepts from data science and statistics were useful in the work. To name a few: Bayesian likelihood analysis, which I used in order to obtain constraints on various fitting parameters of a model based on data that the model tries to fit; The concept of posteriors and priors; Statistical tests and criteria to fit models, such as AIC (Akaike Information Criterion), Bayes factors etc.