Thursday, October 16, 2014

An overlooked paper discovering axions gets published

What's the catch?

Sam Telfer has noticed and tweeted about a Royal Astronomic Society press release promoting today's publication (in Monthly Notices of RAS: link goes live next Monday) of a paper we should (or could) have discussed since or in March 2014 when it was sent to the arXiv – except that no one has discussed it and the paper has no followups at this moment:
Potential solar axion signatures in X-ray observations with the XMM-Newton observatory by George Fraser and 4 co-authors
The figures are at the end of the paper, after the captions. Unfortunately, Prof Fraser died in March, two weeks after this paper was sent to the arXiv. This can make the story about the discovery if it is real dramatic; alternatively, you may view it as a compassionate piece of evidence that the discovery isn't real.

Yes, this photograph of five axions was posted on the blog of the science adviser of The Big Bang Theory. It is no bazinga.

This French-English paper takes some data from XMM-Newton, X-ray Multi-Mirror Mission installed on and orbiting with ESA's Arianne 5's rocket. My understanding is that the authors more or less assume that the orientation of this X-ray telescope is "randomly changing" relatively to both the Earth and the Sun (which may be a problematic assumption but they study some details about the changing orientation, too).

With this disclaimer, they look at the amount of X-rays with energies between \(0.2\) and \(10\keV\) and notice that the flux has a rather clear seasonal dependence. The significance of these effects is claimed to be 4, 5, and 11 sigma (!!!), depending on some details. Seasonal signals are potentially clever but possibly tricky, too: recall that DAMA and (later) CoGeNT have "discovered" WIMP dark matter using the seasonal signals, too.

What is changing as a function of the season (date) is mainly the relative orientation of the Sun and the Earth. If you ignore the Sun, the Earth is just a gyroscope that rotates in the same way during the year, far away from stars etc., so seasons shouldn't matter. If you ignore the Earth, the situation should be more or less axially symmetric, although I wouldn't claim it too strongly, so there should also be no seasonal dependence.

What I want say and what is reasonable although not guaranteed is that the seasonal dependence of a signal seen from an orbiting rocket probably needs to depend both on the Sun and the Earth. Their interpretation is that axions are actually coming from the Sun, and they are later processed by the geomagnetic field.

The birth of the solar axions is either from a Compton-like process\[

e^- + \gamma \to e^- + a

\] or the (or more precisely: die) Bremsstrahlung-like process\[

e^- + Z \to e^- + Z+ a.

\] where the electrons and gauge bosons are taken from the mundane thermal havoc within the Sun's core, unless I am wrong. This axion \(a\) is created and some of those fly towards the Earth. And in the part of the geomagnetic field pointing towards the Sun, the axions \(a\) are converted to photons \(\gamma\) via the axion-to-photon conversion or the Primakoff effect (again: this process only works in the external magnetic field). The strength and relevance of the relevant geomagnetic field is season-dependent.

Their preferred picture is that there is the axion \(a\) with masses comparable to a few microelectronvolts and it couples both to electrons and photons. The product of these two coupling constants is said to be \(2.2\times 10^{-22} \GeV^{-1}\) because the authors love to repeat the word "two". Their hypothesis (or interpretation of the signal) probably makes some specific predictions about the spectrum of the X-rays and they should be checked which they have tried but I don't see too many successes of these checks after the first super-quick analysis of the paper.

There are lots of points and arguments and possible loopholes and problems over here that I don't fully understand at this point. You are invited to teach me (and us) or think loudly if you want to think about this bold claim at all.

Clearly, if the signal were real, it would be an extremely important discovery. Dark matter could be made out of these axions. The existence of axions would have far-reaching consequences not just for CP-violation in QCD but also for the scenarios within string theory, thanks to the axiverse and related paradigms.

The first news outlets that posted stories about the paper today were The Guardian, Phys.ORG, EurekAlert, and Fellowship for ET aliens.



    Single photon counting says "zero solar axions since May 2003." If proposed axion mass equivalent is a few micro-eV and it is produced by crappy center of mass collisions, What powers the x-rays?

    Amass some jellium, blast it with a nuclear-pumped gamma ray laser, and feed the effluent into CAST. What could go wrong? More studies are needed.

  2. Solar Axions or Quantum knots or Glueballs also called Dark Matter Blackl Holes.

    IMHO Axions are so called dark matter Quantum Knots or Glueballs, based on three or four interlocking stringy EM photons created by solar flare interference.
    The XMM Newton observatory seem to have recently identified solar Axions by measuring the seasonal variable x-ray result.

    Large Axions with multiple interlocking photons are called Dark Matter Black Holes, incidentaly expelled from the sun
    and assumed to form primordial Comet nuclei.
    See perhaps:

  3. Do not confuse center of mass calculations with the actual frame that the axions were created and interact . The ones leaving the sun should have some energy which they transfer with the scattering.

  4. Would not the Sun's magnetic field also be a source of axions in this case?

  5. This "evidence" for axion dark matter is one of the silliest I've seen in awhile, and there's been a lot of false "evidence" for dark matter recently.
    Does anybody serious consider this seasonal dependence of x-ray signal as evidence for axion dark matter throughout the universe (as required to describe the multiple different lines of "evidence" for dark matter)?
    First, there's a major jump in logic between a seasonal dependence of x-ray signal and axion dark matter. If axions can interact with magnetic fields, then they aren't candidates for dark matter as required to describe the initial clumpiness of the universe during the time period that the Cosmic Microwave Background was being formed.
    Second, why are axions "coming from the Sun"? If this is really a candidate for dark matter, then the axions are completely surrounding us and can't interact strongly with matter in the Sun.

    Is there something I missing here or is this idea just insane?

  6. Apologies, your comment looks like a stream of nonsense and ignorance to me so I guess that the correct answer is that you are missing "something" ("everything"?) here.

  7. I guess it would, but I think the contribution from the solar field should be at least very roughly axially symmetric and thus not detectable as a seasonal signal, shouldn't it?

    One point that worries me in the paper: The measured north/south ratio seems to match the predicted values very badly, if I'm not totally confused. The prediction (Fig 6) has the highest ratio around April, while the measured ratio has its minimum at this time (~A2 in Fig 19). The presentation is quite confusing, however, as they use different ranges and labelling for the time axis for model prediction and data.

  8. Pardon my farrowed brow but aren't axions primarily an invention to make the associated mathematics "work"?

    Dark matter/string theory physicists must be wary of using the prism of mathematics to explain what is observed in the universe. It's a strong attraction if the path to understanding the mathematics has been a difficult one. (BTDTGTTS; on another matter in the dim past.)

    Perhaps™ the universe works under different rules that fit all the observed behaviour. We can stick to using what seems to work but must remain aware that our theories are synthetic and the result of abstractions.

    The consequences of scientists becoming enamoured with their models are potentially devastating.

    BTW: If axions are sensitive to magnetic fields, then the depth through which potentially (inter)galactic axions have to travel _could_ be seasonal if the primary source is from one direction. i.e. if the source is on the far side of the sun, then the (effect of) axion will be diminished and strongest when Earth is closest to the source. If … potentially … could. i.e. the usual caveats.

    OTOH, observed seasonal fluctuations in Sun-origin axion (effects) could be the result of a non-uniform sun.

    My quick look at the paper (can you blame me for looking at the pictures, first?) didn't spot any consideration of the above by the authors. Thus; my first impression.