The end for supermassive population III stars?

Snapshot showing fragmentation and multiple star formation in a medium of primordial composition (Clark et al. 2011)

Are the first stars really very massive? Some 10 years ago, the idea that the first, metal-free stars would be very massive, became popular. Simple theoretical arguments about radiative cooling and complex numerical simulations both seemed to point to the formation of metal-free stars of masses of several 100s solar masses. Because of their zero metallicity they were dubbed Pop III stars. Early simulations of their formation are commonly associated with Tom Abel and Volker Bromm.
Also observationally this idea appeared attractive, as no metal free stars could be found in the halo of our Milky Way and massive stars would indeed not have survived to the current age.
Although the simulations actually only showed the formation of a very massive pile up of gas, as they were unable to follow the formation of the star, the idea of supermassive pop III stars became a paradigm and even made it into the popular literature.
This paradigm is now starting to erode. New simulations at higher resolution are showing that the dense pile up of gas actually fragments that each form a lower mass star. This is found by a range of different simulations (Turk et al. 2011; Greif et al. 2011; Clark et al. 2011, Science). Lower means less than 100 solar masses. In addition other simulations indicate that the radiative feedback can limit the growth of an isolated star to something like 50 solar masses (Hosokawa et al. 2011).
Observationally, it has been realized that the lowest metallicity stars in the Milky Way have abundance patterns consistent with supernova explosions of lower mass stars, around 40 solar masses and the recent discovery of a very low mass star with extremely low abundances of elements heavier than helium also shows that the theoretical ideas about low mass star formation in low metallicity gas are incomplete (Caffau et al. 2011, Nature)
So, suddenly the idea of very high mass first stars appears to be less attractive. The jury is still out, as even the new simulations cannot properly follow the formation of a star, but the idea which ruled for a while may hardly be remembered 30 years from now.

One thought on “The end for supermassive population III stars?”

Leave a Reply

Your email address will not be published. Required fields are marked *