Radiatively driven Rayleigh-Taylor instability candidates around a forming massive star system
The formation of the highest mass stars is thought to be dominated by instabilities resulting from gravitation and radiation. Instabilities due to gravitation are commonly demonstrated by observations of fragmentation, but those due to effects of radiation have thus far not been found. Here I report on the NACO adaptive optics and mid-infrared diffraction-limited VISIR imaging data of an extremely luminous ultra-compact HII region G333.6-0.2. Two infrared sources, one bright in the near-infrared (appearing pointlike) and another in the mid-infrared (resolved with an elliptical shape) are uncovered through this data, which are located at the heart of this region. These infrared sources appear to be embedded in the waist of a bipolar-shaped nebula and UCHII region, the lobes of which are separated by a dark patch. Dense filamentary features with finger/hook morphology are found; they appear to be connected to the two bright infrared sources and the sizes of these hook features are sharply limited to <5000 AU. The observed properties of this target and a large amount of previous data obtained from the literature are compared together with the results of various numerical simulations of high-mass star formation. This comparison favours the interpretation that the finger/hook-like structures likely represent radiatively driven Rayleigh-Taylor instabilities arising in the outflow cavity of a forming high-mass binary star system.
In 1971, Larson & Starrfield (1971, A&A, 13, 190) presented the theoretical challenge encountered to explain the formation of massive stars by accretion. They showed that stars about 20 times the mass of the Sun would emit so much light even as they begin to form, that the pressure from the radiation on the dusty gas around them will become stronger than the gravitational pull and therefore capable of reversing accretion flows. Almost all proposed solutions to this problem are based on the idea of beaming the radiation pressure in preferred directions such as an outflow cavity to reduce the radiation pressure in the disk plane, thus allowing accretion to take place. How exactly this radiation beaming takes place is a matter of huge differences and debate among modern theoretical scenarios.
In 2009, simulations using the ORION adaptive mesh refinement radiation-hydrodynamic code, showed for the first time that radiation, which was considered to be the “problem” is also the agent that gives the “solution”. Radiation is so dense that it imitates a light fluid punching out against the denser overlying dusty gas fluid, as it escapes from the system. In this process, radiation-Rayleigh-Taylor-instabilities (rRTI) develop in the outflow cavity. The instabilities reorganise the dusty gas into low and high density regions, with the low density regions acting like chimneys to vent out the radiation pressure. This process then allows accretion of the denser dusty gas without further resistance thus aiding the formation of the most massive stars. This phenomenon occurs only for a few hundred to thousand years and very close to the forming star, over size scales of a few thousand AU (the distance up to which radiation pressure supersedes gravitational pull). Therefore, observing them would require not only some of the best telescope facilities but also an ideal target in the sky where rRTI would be in action.
The ultra-compact HII region G333.6-0.2, located at 3,6 kpc from the Sun, is a region which is forming massive stars at the present time, and is the third brightest infrared source in the sky, after the centre of our Milky way and the Eta Carina sources. The author analysed infrared adaptive optics data obtained with the NAOS-CONICA instrument on the 8.2m Very Large Telescope Yepun to discover filamentary features in G333.6-0.2 which he suspected to be rRTI. By supplementing with further analysis of a battery of archival data in different wavelengths, taken by the Chandra X-ray Telescope and the Australia Telescope Compact Array, the author shows that the filamentary features are excellent candidates of the difficult to observe radiation-Rayleigh-Taylor-instabilities.
The analysis of this data reveals the presence of an outflow cavity with several tiny filamentary structures with finger/hook morphology which are probably tracing layers of hot ionized gas and/or dust with predominant small particle emission (Fig. 1). These filaments have a mean length of 2268±1044AU and a mean width of 576±108AU. Comparison with results of radiation hydrodynamics simulations shows that there is a striking similarity between the observations and the simulations. The author uses several arguments to “rule out” the likelihood of other possible mechanisms to produce the hook-like features and strongly favour the presence of actual radiation-Rayleigh-Taylor instabilities.
This work therefore opens a major window to solve a four decade astrophysical problem to explain the formation of the most massive stars in our Galaxy. Further, more complex data, using integral field spectroscopy from the Very Large Telescope, is being analysed by the author and his collaborators to study the full structural details of these HII regions.