On June 3rd 2013 at 15:49 UT NASA’s Swift satellite detected an intense flash of γ -rays known as a short γ-ray burst. Follow-up observations by the Hubble Space Telescope revealed infrared emission that was present 9 days after the burst, but had faded away after 30 days. This infrared transient is likely the first ever observed example of a “macro-nova”, emission that is produced by the radioactive decay of very heavy nuclei that have been freshly synthesized in the merger of a compact binary system consisting of either two neutron stars or a neutron star with a black hole. If this interpretation is correct, the observation could have profound consequences for high-energy astrophysics, cosmic nucleosynthesis and detections of gravitational waves.
γ-ray bursts (GRBs) come in two flavors of different duration. Long bursts (longer than about 2 seconds) are produced in the death of a rare breed of massive stars, whereas short bursts (shorter than 2 seconds) are thought to result from compact-binary mergers. To date, we know 10 systems containing two neutron stars— extremely densely packed objects with masses around 1.4 times the mass of the Sun, but only about 12 kilometres in radius, and that consist predominantly of neutrons. As the stars orbit around each other they emit gravitational waves and therefore slowly spiral in towards one another until they finally merge. Such orbital decays have actually been observed2, and they agree remarkably well with the predictions from Einstein’s theory of general relativity.
The last moments of the inspiral and the final merger are the prime targets for ground-based gravitational-wave laboratories such as LIGO3, VIRGO4 or KAGRA5. The gravitational-wave signal will not only contain information about the physical parameters of the merging system (for example, masses and spins of the neutron stars [it is really about the masses and spins of the individual neutron stars]), but it will also carry the imprint of relativistic gravity and of the properties of matter at densities that exceed those inside an atomic nucleus. On the other hand, such a signal leaves us essentially blind as to the astronomical environment (for example, the type of host galaxy) in which the merger occurs. Therefore, accompanying electromagnetic emission that can provide such additional information is highly welcome.
Compact-binary mergers are also interesting from a nucleosynthesis point of view: they eject roughly 1% of a solar mass of neutron-rich matter6–8 and calculations show that this material is entirely made of heavy ‘rapid neutron capture’ elements9–11. Such elements form through the rapid (as compared to radioactive decays) capture of neutrons on existing atomic nuclei. Generally, adding new neutrons to existing nuclei will not produce stable configurations, thus leading to a competition between further neutron captures and radioactive decays. If a lot of neutrons are present as in the case of disrupted neutron-star material, very heavy and neutron-rich nuclei can be produced that are far from the ’valley of stability’, the location where all the stable nuclei reside in the nuclear chart. In the most extreme cases, the neutron capture only stops near the ‘neutron dripline’ where neutrons are no longer bound and can just ‘drip out’ of nuclei. Once all available neutrons have been consumed, the unstable nuclei decay towards the valley of stability with a large variety of radioactive half-lives.
If this rapid neutron capture occurs inside the ejected material and sufficient radioactivity is left when matter becomes transparent, this should cause a radioactively powered, electromagnetic transient12–14. This phenomenon is often called a ‘macronova’ or ‘kilonova’ in the astronomical literature. The delay time between the merger and the peak of the transient emission and its wavelength range depend sensitively on the opacity of the ejected material. [The emission peaks at the time when the matter becomes transparent. But exactly when this happens—whether this is after a few hours or a week—depends on how efficiently the photons are trapped, i.e. on the opacities. Also the radiation properties are also different: if matter becomes transparent later, the emission is less bright and the photons have a lower energy.] Unfortunately, the opacity of such heavy r-process matter is not well known, but recent calculations15 indicate that they are orders of magnitude larger than those of iron group-like nuclei. Therefore macronovae should peak a few days after a merger at infrared wavelengths. The transient observed by Tanvir et al. may now be the first event of this kind ever observed.
If this interpretation is correct, the observation represents a major advance linking different fields of research. It would mean that short GRBs are indeed caused by compact binary mergers — a connection that has been suspected for a long time but never actually proven. It would further imply that one has captured an r-process ‘in flagranti’, or at least its still glowing ashes. The broad agreement with model calculations suggests that the ejecta carry indeed roughly 1% of a solar mass. This, in turn, indicates that compact-binary mergers are one of the major cosmic cauldrons in which the heaviest elements are forged. Because macronovae emit their radiation practically isotropically, chances are reasonable to observe them in connection with gravitational waves. However, for short GRBs which are bright but highly beamed and hence mostly point away from us, the chances to observe them together with gravitational waves are slim. A macronova observation following a claim of gravitational-wave detection would substantially enhance the confidence that this was not a false alarm and it could provide an accurate source position.
It is worth pointing out, however, that the macronova case is still far from being closed. The data are consistent with the interpretation of a radioactively powered transient, similar to recent model predictions16–18, but we have no spectroscopic evidence that truly r-process elements have been produced. If indeed compact-binary mergers are the ‘central engines’ of short GRBs, they involve a wealth of different physical processes that are not well enough understood to securely rule out other production channels for infrared emission. To better understand the macronova phenomenon we need a larger observational sample and, on the theory side, better r-process opacity estimates (current values are based on only a few representative ions) and more detailed radiation-hydrodynamics calculations. Thus, for the moment the results must be viewed as tentative, but they could really be the first step into the multi-messenger era of compact-binary mergers.
Stephan Rosswog e-mail: email@example.com