So what was the problem between the Big Bang theory and the observed amount of Helium 3?

The problem, in fact, is between a combination of the Big Bang theory, the predicted yields of stellar evolution and the observed abundances.  The Big Bang mostly produces isotopes of Hydrogen and Helium, and because Carbon is a very difficult to synthesize element in the conditions of the Big Bang; so basically, no Carbon, or heavier elements, were produced there.  We found that the same calculations that were successful in understanding how stars produced the Carbon, Hydrogen, Nitrogen, Iron, and other elements, those conditions were robustly reproduced, particularly in low mass stars Helium 3.  Normally, what one does to test the Big Bang is to go out and look at observed abundancies and you subtract on the changes expected from stellar evolution.  When you subtract out the changes of what has been produced, you should be left with what the Big Bang originally had.  In comparing that, it was good except that the observed abundance of Helium 3 was actually a little below the predicted abundance of Helium 3 in the Big Bang.  That meant that stars had to be destroying Helium 3, but instead, when you looked at the population of stars and you did the same calculations that were producing successfully all the heavy elements, the stars seemed to be producing huge amounts of Helium 3, not destroying it.  That was is conflict between two very successful theories between the theory of stellar evolution and how elements are produced and Big Bang nucleo-synthesis.

Your computer simulations [1] suggest that low mass stars do not release any Helium 3 into the Universe…

The low mass stars in particular were the ones that were the criminals in producing too much Helium 3.  The calculations showed that rather out in the envelopes the stars should be producing Helium 3, and then when they become red giants Helium 3 is mixed to the surface.  As stars evolved up the giant branch, large fractions of the envelopes were rejected and the result was a big increase in Helium 3. We found that there was an instability mechanism that was not recognized in the 1D calculations.  1D codes, one dimensional codes, which is how most stellar evolution is done, cannot mix material unless you tell it to; a 3D code has all the mechanisms for instability that nature has, but the 1D code has to be told when to mix.  Traditionally, it’s told when to mix based on stability criteria for transporting energy.  We found there was an additional mechanism that the 1D codes didn’t recognize and was operational—when you see it it’s really obvious.  But that one mechanism, which is new to us, occurs on the giant branch when the Hydrogen burning shelf in these stars approaches the bottom of the convection zone.  As it does that there’s a stable region just outside the hydrogen shelf just outside the convection zone, that inspite of everyone’s looking for how it might be made unstable, no one had found a way.  What we saw that there was an unusual nuclear reaction:  that when the hydrogen burning shelf approached it, the Helium 3, which has a fairly fragile nucleus, actually burned with itself to produce Helium 4+2 protons.  That is an unusual reaction as most nuclear reactions actually raise the average weight of a particle in the gas while this one lowers the average weight because it produces 2 Hydrogens of a particle of gas.  The bubble of material that you produce with this slight enhancement in Hydrogen is buoyant and starts to float up.  As it floats up, it is replaced by the heavier material above it, which then gets processed, which drives more bubble’s going up.  A kind of equivalent to this is, if you took a 1D code and tried to model a swimming pool in 1D and you had a bubble of air at the bottom, the bubble of air would very happily stay at the bottom—it has enough pressure to hold the material above it up.  But in 1D it doesn’t know about the Rawley-Taylor instability that leads to the formation of a bubble and the bubble rising through the material.

Has the apparent problem with the Big Bang been solved?

Yes, we are working on a somewhat more elaborate version.  We mention in Science that the mixing is robust enough that it destroys the bulk of Helium 3 that was made.  In fact, it probably destroys a little more Helium 3 than was made, so it looks very promising too.  Now, in the future when we model this over a whole range of stars, we will in fact produce a slight decrease but not much.  On top of that—this is not in the Science article, but in the more extensive work we’re following up with—red giants, we have, for about 3 decades, nearly 4 decades, been observing things like the carbon isotope ratios in red giants.  The C12 and C13 and the standard stellar evolution calculations show that when the stars live on the main sequence and become red giants, their C12 and C13 ratios drop from 90, like it is on our sun, to a number like 25 in the calculations.  When the obsrvtns started coming in, they would seem to be more in the neighborhood of 10 or 15, even with a few lower than that.  That was a real puzzle.  People suspected there had to be some other mixing mechanism behind it.  People had proposed things like rotation to solve that problem; if you could solve that problem, you could also solve the Helium 3 problem, but it would require all stars to be rapidly rotating at their core.  That also has a checkered history ]and you don’t like having to introduce a new variable, like “all stars need to be rotating fast in their core.”  What this does is show there is a mechanism that is so straightforward that it is as if, “Why didn’t I look at that before?”

[1]  Lattanzio et al. To be published in Science, 26 October (2006)

David Dearborn works at the Lawrence Livermore National Laboratory.

Interview by Thanh Tam Candice Vu and Gilles Prigent

Interview: Send a private message to the author
Follow this author