UCL models of lopsided supernova debris reassign four stars once credited to hypernovae

The method starts with the star before it dies. "Just before a star goes supernova, it has layers like an onion, with heavier elements at the center and lighter ones farther out," said Ralph Schoenrich of UCL's Mullard Space Science Laboratory, the second author ^[7]. "Our model looks at how much material from each layer or each region would be needed to fit the observed star." ^[8]

According to the UCL account, the earlier hypernova arguments did not allow for lopsided ejecta ^[2]. Real supernovae throw elements out unevenly. "Oxygen might predominantly fly in one direction, sulfur in another," said Anmol Aggarwal, the first author and a PhD student at the same laboratory ^[3]. So a gas cloud that later forms a star can take in a few pieces of the explosion and miss its average mix ^[3]. "We developed mathematical models that took account of this and found in all cases that stars with a very unusual mix of ingredients were most likely formed from ordinary supernovae," Aggarwal said ^[4].

The red giant needed a second argument. Its silver and uranium can come only from neutron star mergers or, in theory, from the hypernova of a highly magnetized massive star ^[12]. Earlier work ruled out a merger on timing. The star is very metal-poor, and in a Milky Way model that enriched its gas quickly, neutron stars would not have had time to form and collide before it was born ^[13]. Aggarwal and Schoenrich pointed to the star's high speed and its orbit, which runs opposite to most Milky Way stars. They argued it formed in a dwarf galaxy, where metal-poor stars can form much later, and was pulled in afterwards ^[14]. Given that history, the asymmetric-ejecta model favored a merger plus one ordinary supernova ^[15]. That conclusion rests on two inferences, a birthplace read from the star's motion and the ejecta model ^[14][15]. If the star was born in the Milky Way after all, the old timing objection to a merger still applies ^[13].

The sample is four stars ^[11]. For the three in the first paper, the preference for ordinary supernovae is described as statistical favor, and the phys.org report does not say by what margin ^[11]. "All the evidence we see for hypernovae is suddenly gone," Aggarwal said ^[5]. What the papers support is narrower: none of the four stars now has a hypernova as its most likely source ^[16]. The thing this doesn't tell you is whether hypernovae happen at all. They are theorized to occur when a more massive, rapidly spinning star collapses ^[6].

The result also depends on a process the field has not measured well. "We still don't know how well the supernova material gets mixed before it forms new stars. That is despite decades of work in this area. Our research suggests some mixing goes on, but incomplete mixing," Schoenrich said ^[9]. In my view the reanalysis is persuasive if that holds. If ejecta blend more thoroughly than the models allow, each star again inherits the explosion's averaged chemistry, and the case for a bigger explosion returns ^[2]. Schoenrich said the same uncertainty has to go into "our models of how chemicals evolve in galaxies and of how the interstellar medium works" ^[10].