I'd barely adjusted to the cold Sheffield weather after returning from my placement year working at the Isaac Newton Telescope (INT) in La Palma. I was back to finish my Master's degree and after spending a year working as a "proper" astronomer I was hardly relishing the prospect of taking exams again. Luckily I had a research project lined up that I could really get my teeth stuck into. I'd be working with Steven Parsons on the excitingly-named eclipsing double white dwarf binary, SDSS J115219.99+024814.4 (or SDSS J1152+0248 for short). I'd been interested in white dwarf binaries and post common-envelope systems for a while and so made sure that when I returned to Sheffield this would be what I'd be working on.

These double white dwarf binaries are particularly exciting due to them being considered as the likely progenitors of many of the exotic phenomena in the Universe, from single hot subdwarf stars and AM CVn binaries to type Ia supernovae. The likelihood of each outcome is heavily dependent on the bulk composition of the two white dwarfs and so determining these compositions is of key interest. Typically this is done by comparing the measured masses and radii to theoretical models for different compositions, requiring precisely determined parameters.

The group in Sheffield is lucky in that it shares access to a suite of incredibly high frame-rate imaging instruments on telescopes around the world, designed for studying astrophysical phenomena on the fastest timescales: ULTRASPEC, ULTRACAM, and the most recent addition, HiPERCAM. The initial plan was to combine VLT X-Shooter spectroscopy with ULTRACAM eclipse photometry in three bands to determine precise parameters for these two white dwarfs. Hopefully we'd manage to detect the fainter white dwarf spectroscopically, something that had previously remained elusive. If successful, this would make it only the second such system and would allow us to robustly determine both masses. It didn't take long for us to start getting excited as hints of the secondary began to show up. When fitting the spectroscopy, the parameters for the secondary were consistently converging to the same place so it seemed that a signal was there, hidden within the noise. We managed to further refine the radial velocity fitting and get some robust masses out just before the Christmas break, leaving the photometry until after we got back.

The key discovery came during this break. There was a HiPERCAM run on the GTC in early January, Steven and Vik Dhillon were the observers for this run and I'd mentioned how it'd be nice to get some five-band HiPERCAM data for this system. It turns out that Steven had already earmarked this target for HiPERCAM due to the secondary appearing to lie close to the white dwarf instability strip. He was right. These much higher signal-to-noise light curves showed clear pulsations and I'm told that there was quite a lot of excitement in the control room that night.

With the help of Stuart Littlefair, we were able to fit these new and improved light curves using a gaussian process to model the pulsations. Comparing our now well constrained masses and radii with those from theoretical models it seemed that neither of the two typical models for white dwarfs comfortably matched our observations. There are a few possibilities why this could be but the most exciting of these is that these white dwarfs have hybrid core compositions with a significant fraction each of Carbon, Oxygen, and Helium. Such white dwarfs have been simulated but never conclusively observed. The presence of the pulsations along with the precise characterisation of the system should enable an independent determination of the core composition through astroseismological modelling, making this system a very exciting target for future study.

Link to university press release

Header image credit: Dr. Mark Garlick