While there is mounting evidence that long Gamma-Ray Bursts (GRBs) are associated with the collapse of massive stars, the detailed structure of their pre-supernova stage is still debatable. Particularly uncertain is the degree of mixing among shells of different composition, and hence the role of magnetic torques and convection in transporting angular momentum. Here we show that early-time afterglow observations with the Swift satellite place constraints on the allowed GRB pre-supernova models. In particular, they argue against pre-supernova models in which different elemental shells are unmixed. These types of models would produce energy injections into the GRB engine on timescales between several hundreds of seconds to a few hours. Flaring activity has {em not} been observed in a large fraction of well-monitored long GRBs. Therefore, if the progenitors of long GRBs have common properties, then the lack of flares indicates that the massive stars which produce GRBs are mostly well mixed, as expected in low-metallicity, rapidly rotating massive stars.
Download