A Disk-Shattering Discovery

Jeremy Hare

We are very pleased to welcome Jeremy Hare as a guest blogger today. Jeremy is a co-author of a study led by George Pavlov from Pennsylvania Statue University and Oleg Kargaltsev from George Washington University that is the subject of our most recent press release, on a binary system named LS 2883. Jeremy is about to begin his fourth year of graduate school at GWU working under Oleg Kargaltsev. He studies high-mass gamma-ray binaries, mainly in X-rays, and the classification of X-ray sources using machine learning. He tells us that LS 2883 was the first research project he worked on in graduate school and that it has been “very exciting to study!”

High mass gamma-ray binaries are rare objects in the Galaxy. These binaries consist of a massive star (usually with a mass greater than 10 solar masses) and a compact object, a neutron star or black hole. Many high-mass stars have a disk of material around them, which the compact object can interact with as it nears the star in its (often elliptical) orbit. High-mass gamma-ray binaries can accelerate particles to extreme energies of 10 TeV (=1012 electron volts, or eV) or higher, which is comparable to the energies that are currently being produced at the Large Hadron Collider. These particles then scatter off of lower energy photons (packets of electromagnetic energy that make up light) produced by the star, transferring some of their energy and boosting the photon’s energy to the GeV (109 eV) and TeV energy range.

LS 2883 stands out among the other high-mass gamma-ray binaries as it is the only one where the nature of the compact object is known. This is a young pulsar PSR B1259-63 with a spin period of 48 ms (about the typical period of a household washing machine). The orbital period of this system is 3.4 years and the orbit is highly elliptical, similar to that of a comet. This system is also unique because it has been observed flaring by the Fermi Large Area Telescope (in the GeV energy range; http://www.nasa.gov/mission_pages/GLAST/news/odd-couple.html), which is a puzzling phenomenon that has yet to be fully explained or understood. The flaring happens shortly after the pulsar passes closest to the star (periastron) and therefore must be associated with the interaction between the disk and the pulsar wind. This system has become a theoretical testbed for models of wind interaction and intra-binary shocks (i.e., areas where the pulsar wind and stellar wind meet) due to the known nature of the compact object. 

In 2009, this system was observed with the Chandra X-ray Observatory (for the first time in the imaging mode) and hints of extended emission were found. This emission was thought to come from the pulsar’s wind being blown out of the binary by the stellar wind. My first project in graduate school was analyzing the new data for this system, from Chandra observations that occurred in May 2013. We had two new observations that seemed to show some extended emission far away from the binary. At this point, two possible scenarios were entertained. First, it was possible that this emission came from a cloud of matter (either a disk fragment from around the massive star or a clump of the pulsar wind) that was launched from the binary system at an incredibly fast speed (about 7% of the speed of light!). However, it was challenging to find a way for the energetics to work out, especially because the cloud was moving very quickly and did not seem to be slowing down at all. A second explanation was therefore needed. Alternatively, the emission could come from the pulsar's wind “termination shock” region (analogous to a sonic boom from a jet, this region is where the flow of matter from the pulsar slows down), similar to those seen in pulsar-wind nebulae around isolated pulsars (see http://arxiv.org/abs/0801.2602). The varying distance to the termination shock could then come from the variability in the pressure at the shock front. However, in this case (which we considered a favorable interpretation) this distance should not change by a large factor and it does not have to be steadily increasing. 

Our observations of February 2014 completely surprised us. It showed the extended structure still traveling away from the binary. And, not only was it not slowing down, but it actually appeared to be accelerating! This was very strange because we could not figure out what could be supplying the energy to create this acceleration and the observed behavior was difficult to reconcile with the termination shock scenario. After several weeks of struggle and many failed hypotheses, we finally picked one that seems to be plausible. In our prior calculations we assumed the ejected material (a fragment from the disk around the massive star) was traveling through the stellar wind as it flew out of the binary; this should lead to a rapid deceleration of the fragment because the stellar wind moves very slowly and is massive (composed of protons as well as electrons). However, we conceived the idea that maybe the fragment is actually moving in the unshocked pulsar wind. The unshocked pulsar wind represents a highly relativistic magnetized flow, travelling at speeds of 99.995% or more of the speed of light, comprised of relativistic pairs (electrons and positrons) and therefore it could constantly supply energy to accelerate and heat the ejected disk fragment. Under this assumption, a natural explanation for the X-ray emission is the heating and particle acceleration in the shocks developing at the interface between the clump and the pulsar's wind. If supported by future observations, this scenario is exciting because the motion of the disk fragment and its emission properties must then be intimately related to the properties of the unshocked pulsar wind, providing a unique peek into the physics of this otherwise elusive unshocked pulsar wind zone (the unshocked pulsar wind is not expected to produce any detectable emission in X-rays).

LS 2883/B1259-63 is an amazing laboratory for studying high-energy astrophysics and so far this system has surprised us with each new observation. We have two more observations scheduled with the Chandra X-ray Observatory that, we hope, will confirm our most recent interpretation of the observed phenomenon. That being said, it would not be shocking if these new observations surprised us once again.