Flickering laboratory flames have ignited a team of researchers’ understanding into what might spark the colossal explosions of Type Ia supernovae, which may not only help us to grasp a better understanding of these violent variables, but shine a light on the questions posed by the evolution of the Universe.

Originating from the nimble dance of a white dwarf circling a bloated companion star like fighters in a boxing ring, the dwarf delivers a blow to its component by stealing material. After winning too many rounds, the white dwarf starts to struggle and sweat under the mass it has accumulated and, as a result of biting off more than it can chew, reignites its collapsed stellar furnace and detonates into a violent stellar eruption that is so bright that it briefly outshines all of the stars surrounding it. Due to their momentary brilliance and standard intrinsic brightness, Type Ia supernovae can be used to calculate cosmic distances to other astronomical objects.

In an attempt to fire-up a more solid understanding of the conditions that drive Type Ia supernova, Aaron Jackson, an NRC Research Associate working in the Laboratory for Computational Physics and Fluid Dynamics at the Naval Research Laboratory in Washington D.C, along with his colleagues Dean Townsley from the University of Alabama at Tuscaloosa and Alan Calder of Stony Brook University on Long Island, New York, got to work on new 3-dimensional calculations modelling the run-away turbulence which forced a slow-burning flame to its limits, causing it to rapidly blast in what the team have referred to as a deflagration-to-detonation transition (DDT). “We know DDTs happen – just put low octane gas in a high compression engine and you get a detonation, for example, that may be a DDT,” says Calder. “These things have been seen in laboratory experiments and mine explosions as well. We are interested in the basics of the process and Type Ia supernovae are a big, fun laboratory!”

“We are focusing on computing simulations of the first phase of the supernova explosion, the ‘deflagration phase‘ before the detonation takes place,” says Townsley. “We have two general goals; first, to better understand how our estimation of interactions between the turbulence and flame, of which there is a fair amount of uncertainty, affects the results of the simulations such as the predicted properties of the supernova explosion. Second we wish to better characterise the turbulence affecting the flame just before the detonation occurs in order to better understand the conditions under which the transition to detonation might take place.”

Performing the simulations on large computers at national facilities such as the Ranger supercomputer at the Texas Advanced Computations Center, the Intrepid BlueGene/P supercomputer at the Argonne Leadership Computing Facility and computers at the New York Center for Computational Science, the team pushed through the poorly understood DDT mechanism which is continuing to be hotly debated by astronomers. “What is happening is that transition from a reaction front [fusion reactions in this case] being moved through the star due to heat conduction [a deflagration or flame] to being moved by a propagating shock [a detonation],” Townsley tells Astronomy Now. “This is a poorly understood phenomenon in its details but the astrophysics community currently has some hypotheses about how this transition might take place. The leading hypothesis has generally to do with strong interaction of turbulence with the flame.” While there is a prevailing hypothesis that the strength of the turbulence causes DDT, Townsley states that it could be that astronomers are well off the mark meaning that it could be some other phenomenon that is driving the explosive reaction.

With the lack of knowledge about the process comes a variety of outcomes from the explosion. “We mainly discuss variation in terms of the amount of radioactive nickel that is synthesised in the incineration of the white dwarf,” says Townsley. “The decay of this nickel has a half-life of about 20 days and its products is what powers the bright supernova that we see from a distance. What is also important is how fast this ejected material is moving, which also affects the brightness and decline time of the supernova. Type Ia supernova are observed to produce anywhere from about 0.3 to 0.9 solar masses of radioactive nickel.”

While there are holes in the understanding of DDT, the researchers believe that the extreme turbulent intensities inferred by the white dwarf point to the mechanism being a likely cause. Drawing comparisons between their simulation and real observations of the explosions thrown out by supernovae could also indicate conditions for the mysterious mechanism. “By looking at observations of Type Ia supernovae and their remnants, we can get spectra and determine the composition of the ‘ash’,” says Calder. “Because the composition of the ‘ash’ depends on the density at which the white dwarf material burns, from these observations we infer what the density structure of the dwarf must have been when it blew up. Models that invoke a pure deflagration or detonation do not produce the stratified ejecta seen in the observations; heavy iron-group elements in the centre with intermediate-mass elements surrounding.”

However, it seems that the problem mentioned by Calder can be solved with the introduction of a model that incorporates DDT. “It would work well,” adds Calder. “The star has time to expand during the subsonic deflagration so that a good fraction of the material reaches lower densities while the core density is high. Then when the detonation goes, the expanded star produces the right stratified ejecta to match the observations.”

As the team simulated how these explosive stars might operate, they were well aware of the disadvantages and advantages that they met along the way. “In order to simulate the whole star on the computer, it is necessary to leave much of the detailed physics ‘unresolved’; that is, the simulation technically cannot get this physics correctly based solely on simple physical laws,” says Townsley. “Instead we put in models intended to approximate what is going on in the unresolved physics. This is an imperfect art, as there is more than one way to do such a thing, and one has to make choices about which features of the unresolved physics are the salient ones. Also, not all of the emergent physics occurring on unresolved scales is well understood [turbulent combustion is a good example] and therefore we cannot determine from first principles whether the approximate models are correct.”

“Our goal is to provide a more realistic simulation of how a given supernova scenario will perform,” concludes Townsley. However, for now, the researchers can only speculate the fruits of their work which could yield more precise distance estimates when it comes to astronomers using these dazzling outbursts as standard candles. “It is a long-term goal and involves many different improvements that are still in progress.”