Wednesday 21 May 2014

BICEP2 and Axions. A few comments.

After our paper on Axions and BICEP2 came out (here) we were contacted quite a bit by various media outlets for comment. One article appeared in Nature News. There is another due to appear in Quebec Science tomorrow. All the answering of interview questions made me think quite hard about explaining this business in a manner understandable by the lay person, and I think I got quite good at it. So I've decided to reproduce for you here the transcript of the interview I gave for Quebec Science. Their article only used a few quotes of mine in the end, but this here is the whole shebang!

(P.S. Sorry for the weird formatting: I'm not really sure what happened)


On March 17th, astronomers announced they’ve found some evidence of gravitational waves. Could you tell us quickly what they are and how/why they confirm the theory of cosmic inflation ?


What was actually detected is large angle so-called "B-mode polarisation" in the cosmic microwave background (CMB) (the other "E-mode polarisation" was detected years ago by WMAP). If this polarisation is not due to some foreground, for example dust or magnetic fields within the galaxy, then it is what we call "primordial", or at least "cosmological in origin". A leading theory for what could have caused this large angle B-polarisation is gravitational waves produced during inflation. There are other possibilities to produce large angle B, but the production via inflationary gravitational waves was a key prediction of inflation, worked out by a number of theorists in the 80's and 90's. It is therefore seen as the "simplest" explanation for B, and so seen as a strong confirmation of inflation, if the B is truly primordial/cosmological. Confirmation of this is necessary, and will be provided fairly soon by other experiments measuring the polarisation at different frequencies than BICEP; for example, the European Planck satellite should confirm this in their next data release scheduled for some time in late 2014.



What do these results, if they are confirmed, reveal about inflation ? Is it exactly what the theory had predicted ? Or does it help to define more precisely how powerful inflation was, for instance?

In the simplest models of inflation the gravitational wave production is a direct probe of the energy scale at which inflation operated. The predictions of inflation give the amount of fluctuations in "curvature" or "scalar" modes: fluctuations we see as temperature fluctuations in the CMB, compared to the fluctuations in gravitational waves, as a function of the energy scale inflation operated at. We already knew the amplitude of the scalar fluctuations from as early as COBE in the 1990's, so the inflationary prediction boiled down to saying "if you measure B-modes with a certain magnitude, this implies inflation operated at a certain energy scale". BICEP measured a large amount of B-modes with an amplitude that implies the inflationary energy scale is very high, up near what particle physicists call the "grand unified scale".


If I get it right, you used their measurements to rule out some models about inflation and dark matter. Could you first try to explain what is the link between inflation and dark matter ?

What we did was to work out what this high energy scale of inflation implies for certain theories of dark matter (DM), in particular axion DM. Inflation imprints the seed fluctuations of structure in the universe. If the inflationary field, the "inflaton" is the only game in town during inflation, then these fluctuations have a characteristic shape, called "adiabatic", and this shape is very close to what we observe in the CMB. However, if there is already DM around during inflation, then the inflaton imprints fluctuations in the DM which are non-adiabatic (so-called "isocurvature"). Given that the universe is seen to be largely adiabatic we are left with just a few options: 1) The DM was not around during inflation 2) There was not too much DM around during inflation 3) Inflation did not happen at too high energy (so that the non-adiabatic fluctuations were small). The B-modes imply a high energy for inflation, so we are left with option 1) or option 2).

You work on the axion. What is it ?

Axions are a candidate for the DM in the universe. They come in many flavours. The most minimal scenario posits just one axion. It was put forward in the late 70's by Peccei, Quinn, Weinberg and Wilczek, not as a DM particle, but as a solution to a problem within the standard model of particle physics. Later it was realised the axion could be DM, and so it is seen as a strong candidate, since its existence was predicted for other reasons. Unfortunately, axions interact very weakly (as all DM has to, or it wouldn't be dark!) and we haven't seen any in the lab yet, but many experiments are trying to close in on them. There are also many other possible flavours of axion, in particular they are predicted to be very abundant in string theory. In string theory the axions come from having extra dimensions of space-time, and the number of them comes from the huge complexity of shapes possible for the extra dimensions. This has come to be called the "String Axiverse". Any or all of these string theory axions can contribute to DM, so in this scenario the DM is not just one particle, but possibly many different ones.

What do these measurements about gravitational waves reveal about axions (that they could not have been created before inflation, is that right ?) ? Why ?

In the axion DM scenario, there is an energy scale, f, that controls whether the DM was around during inflation. What we worked out is that if the axions are around during inflation exactly how little of the DM must they be. That is, we worked out the answer to 2) above. For almost all possible axion masses we showed that if this energy scale f is large then the axions cannot be the DM. On the other hand, if f is small then the axions are produced after inflation, that is option 1).

To reiterate, if axions are produced before inflation then they screw up the adiabaticity of the universe as measured by the CMB. This means that any axions produced before inflation have to be just a small amount of the DM. Axions produced after inflation can still be the DM. This "before or after inflation" tells us about the axion energy scale, f. For the "after inflation" scenario, f has to be small and axions are squeezed into quite a small window.

What does it mean for dark matter research ? Is this exciting ?

This is very exciting for many reasons: it is a very strong constraint on axion DM. Firstly, if axions are made after inflation then we now have a really good idea what energy scale they can be at, and experiments can target this. But more exciting for me is how we can now use axions to check on what we think about inflation by looking for them in exactly the opposite place, where they are made during inflation at high f. That is, if we can find high f axions (which are very light, the mass is inversely proportional to f) that should have been made before inflation, then this challenges the simple model where the B-modes were caused by gravitational waves from high energy inflation. 

There are quite a lot of ways to look for these high f axions, and actually some reason to believe they may exist. Firstly, in string theory many models have high f, so if we think these models of string theory describe the world, then we had better find these DM axions, and there had better be something more complicated happening with inflation. There are new experiments that have been put forward that can look for high f axions in the lab. From my own research perspective we can go looking for high f axions out in the cosmos, by looking for the imprint they leave on galaxy formation. There are a few avenues where it is even possible that high f axions "fix" galaxy formation and make it more like what we observe, forming galaxies in a nicer way than other models of DM do. It's possible that in the near future, if high f string theory axions exist, that we will see their imprint in galaxy formation with surveys like Euclid. If we find any evidence for axions formed before inflation in galaxy formation, or in the lab, then it will be back to the drawing board with the simplest inflation models.

And the summary

The clear cut answer is the following, I think. If we assume axions are all of the dark matter, then *the simplest interpretation* of BICEP2 rules out axions that were made before inflation, as the other blog says. On the other hand we can, as you quote from my blog, use this new constraint to limit the fraction of the dark matter that can be made up of axions. The other key point is that there are other ways to detect axions that are independent of inflation, for example in the lab or in galaxy clustering. If any of these searches actually detect them, then that means they are a large fraction of the dark matter, and the simplest interpretation of BICEP2 would have to be wrong.

So I think saying, as the CERN blog does, that "BICEP2 points out where to look for Axions" is too simplistic. It only does that in a way that depends on the model of inflation. We Should still look everywhere for them until That model of inflation can be pinned down.

No comments:

Post a Comment