← Return to Homepage

MPhys Project

Supervised by Dr. Rhys Seeburger

Unlikely Friends: the Hunt for Neutron Stars with Stellar Companions.

Overview

When massive stars (M > 8 M☉) near the end of their lives, they can undergo spectacular explosions and leave behind some of the most interesting and poorly understood objects in the universe: Neutron stars and black holes. As most massive stars can be found in binaries, we expect many of these compact, dark objects to also have a companion. For a while, after one star in the binary has exploded, but the other is still happily shining and fusing elements, we might thus find an unlikely friendship between a star and a stellar corpse. While population models predict millions of such systems in the Milky Way, only a few have been found through extensive searches using both ground- and space-based spectroscopy, astrometry and photometry. Many times, astronomers have been tricked by so-called ‘imposter’ systems that masquerade as containing a compact object, but upon closer inspection revealed a stellar companion, hiding in the light of the primary star.

(Project Proposal. Dr. Rhys Seeburger, 2026)

A link to the code used for the project can be found on GitHub here.

Data

The Gaia DR3 database was queried using ADQL under these parameters to find candidates: the lower bound mass of the secondary star must be higher than the upper bound of the primary, the mass of the lower bound of the secondary star must be >1.5M☉ and <3M☉, the significance of the orbital solution must be >20. This returned 23 candidates. These candidates were then checked against the FEROS and UVES databases finding 18 and 6 candidates with spectra respectively (the 6 UVES candidates were also found in FEROS). UVES was chosen as the primary dataset due to the nature of the FEROS data and the work required to process it. This left us with the following candidates, GAIA_ID: 4295254307232999296, 4039283116865495424, 4239611700214875904, 4973506334166805376, 6551893071549223040, and 6609971199873698944.

Results

While this project is still in its infancy, it has produced some results.

All 18 candidates found in FEROS and UVES databases overlaid on 100,000 random stars from the Gaia catalogue. Not de-reddened.
All 18 candidates found in FEROS and UVES databases overlaid on 100,000 random stars from the Gaia catalogue. Not de-reddened. This indicates that the 18 candidates with spectra lie on or just above the upper main sequence, with none on the giant branch. Some sit 2–3 magnitudes above the main sequence at their colour, consistent with either evolution into the Hertzsprung gap or interstellar reddening, which has not yet been corrected. Regardless, it supports the assumption that the primary star in each of these candidates is a main sequence or slightly evolved star.
Spectra of all six UVES candidates. Barycentric corrections applied and normalised. Plotted before chip gap at approximately 576nm.
Spectra of all six UVES candidates. Barycentric corrections applied and normalised. Plotted before chip gap at approximately 576nm. All show strong Hβ absorption lines except 6551893071549223040. This, along with the turbulence of the spectra, suggests that the star is cool.
Spectra of all six UVES candidates. Barycentric corrections applied and normalised. Plotted after chip gap at approximately 576nm.
Spectra of all six UVES candidates. Barycentric corrections applied and normalised. Plotted after chip gap at approximately 576nm. All show strong Hα absorption lines except 6551893071549223040 and 4039283116865495424. The the lack of this absorption in the former is likely for the reasons discussed in the previous figure whereas the latter shows a double-peaked Hα which potentially originates from an accretion disk. However, further study of this candidate is required to support that conclusion.