Above is a composite image of the supernova remnant Cassiopeia A (Cas A). Around 300 years ago, a relatively nearby star (3.4 kpc) exploded in a supernova, flinging out the ejecta visible in the image. Like most supernova remnants (SNR), the ejecta visible is highly structured, pointing to a turbulent and violent (and asymmetric) explosion.
But there’s something more interesting lurking in the ejecta profile of Cas A. In the upper left corner, and lower right corner (less visible), there are “jet-like” structures of ejecta that extend beyond the primary forward shock of the SNe — these have long been the subject of debate.
One theory for how these structures came to be is that the progenitor star went supernova via an internal instability that preempted a “jet-driven supernova” — one where the core rips the star apart along two jets (Kocholov et al. 1999).
But there’s a few problems there. That theory predicts the “kick” vector of the central object (probably a neutron star) would be roughly aligned with the jet axes. Yet, in Cas A, the neutron star seems to be on a trajectory nearly 90 degrees offset from the jet axes. Additionally, that theory predicts a predominance of iron in the jet-ejecta (since the core is what drove the explosion). In Cas A, though, we don’t see this large proportion of iron in the outer regions of the jet-structures.
Does a simpler theory exist that could explain the jet-like structures?
This was the question that drove my first research project. In the summer of 2014, I particpated in the URCA (Undergraduate Research in Computational Astrophysics) internship (technically, an NSF-funded “REU”) at North Carolina State University. There, I worked with Prof. John Blondin developing and running a hydrodynamical simulation that tested whether density asymmetries in the circumstellar medium (CSM) of Cas A’s progenitor could have produced the jet-like structures we see today.
Such an asymmetry isn’t to be unexpected. Stars, over the course of their lives — and particularly in the heavy mass-loss stages near the end of their lives — exhibit asymmetries in mass loss direction. In short, more mass loss around the equator, less at the pole (perhaps due to rotation). In particular, the presence of a binary companion, if one existed, would have tended to incite a larger pole-to-equator asymmetry.
To test this theory, I ran a three-dimensional hydrodynamic simulation of the Cas A SNR on the Stampede supercomputer at TACC (Texas Advanced Computing Center).
Well, back up a second. To test this theory, I spent the summer learning Fortran, the hydro code VH1, and hydrodynamics in general. I implemented grid expansion (to track the explosion) into the code, as well as chemical tracers into the ejecta profile to track where the iron ended up. I inserted the CSM density asymmetry, and adapted the 3D version of the code to use Kageyama & Sato’s “Yin Yang Grid,” which, like the two parts of a baseball, has no poles — where the grid element becoming spatially small drives longer time-steps. I started with 1D, then worked my way up. (Hey, I was a freshman.)
Let’s make a long story short. The full production on the supercomputer indicated that a fairly reasonable gradient in CSM density could, hydrodynamically, produce fairly distinct jet-like structures.
I presented this research in a poster at both the UNC Chapel Hill Annual Research Symposium and, by invitation, at the FOE (Fifty-one Erg) conference. I won 1st and 3rd place prize, respectively. As of now, to my understanding, various other avenues of research have converged on Cas A not having been jet-driven as an explosion mechanism. But I’ll always value the education and experience URCA provided, and would highly recommend it to anyone interested in computational astrophysics.