Seeing Signs of Dark Matter Annihilation

Matt Strassler [April 18, 2012]

It’s not easy to see dark matter, which makes up most of the matter in the universe. It’s dark. And yet, there is one way that dark matter might, in a sense, shine.

How? If dark matter is made from particles that are their own anti-particles (as is true for photons, Z particles, and [assuming they exist] Higgs particles, and perhaps neutrinos), then it is possible that two dark matter particles might encounter each other and annihilate (just as an electron and a positron can annihilate, or two photons can annihilate) and turn into something else that we can potentially detect, such as two photons, or indeed any other known particle and its anti-particle. Whether this is an effect that we could hope to observe depends on a lot of things that we don’t know… but there’s no harm in looking for it, and good reason to try.

How would we hope to find it?

Fig. 1: The stars of a spiral galaxy form a disc with spiral arms centered on a bulge (white), and they sit inside a roughly spherical clump or `halo' of dark matter (grey), whose region of highest density is inside the galactic bulge. Also within the halo are smaller clumps of dark matter, within some of which are found dwarf galaxies with a relatively small number of stars. Very little is known about the detailed structure of the halo around our own galaxy, the Milky Way.

First, we may want to look toward the center of our galaxy, the Milky Way. Just as the most likely place to see an automobile accident is in heavy rush-hour traffic, the place where collisions of dark matter particles are most likely to occur would be wherever the density of dark matter is highest. And that density is largest in the centers of galaxies. The reason (see Figure 1) is that galaxies of stars form in and around large clumps of dark matter — indeed, most of the mass of the Milky Way galaxy is dark matter, distributed in some fashion that is very roughly a sphere, though with a detailed structure that is unknown and possibly very complicated. The stars, and the big clouds of atoms out of which they form, form a rotating disk with spiral arms, sitting within that big sphere, with a ball of stars (the “bulge”) at its heart. The stars in the disk and bulge are presumably centered on the highest concentration of dark matter. So collisions, and consequent annihilations to particles that we can potentially detect, may be occurring near the center of our galaxy, and for this reason we might want to design scientific instruments that can look in that direction, seeking a hint that these annihilations are taking place.

Unfortunately, hints are not so easily obtained, because there aren’t many types of known particles that, if produced in dark matter annihilation near the center of the galaxy, can travel from there to Earth. The only particles that live long enough to reach the Earth are electrons, anti-electrons (positrons), protons, anti-protons, some other stable atomic nuclei (such as helium), neutrinos, anti-neutrinos and photons. But neutrinos (and anti-neutrinos) are extremely difficult to detect, while almost all of the others are electrically charged, so their paths bend and loop in the galaxy’s magnetic field, causing most of them never to reach Earth at all and assuring that we can’t tell, if they make it here, whether they came from the galactic center or not. That leaves photons as the only particles that both can travel straight from the region of the galactic center to Earth and can be easily detected.

Fig. 2: Dark matter annihilation may directly or indirectly produce photons along with other particles; but only the photons may travel in a straight line to Earth and be easily observed (by specialized detectors mounted on satellites.)

So a good hint of dark matter annihilation could come from an unusual class of high-energy photons that are streaming from the galactic center but not from anywhere (or almost anywhere) else; see Figure 2.

However, there’s still a big challenge for that strategy. There are a lot of unusual astronomical objects at the galactic center, and they make high-energy photons also. How can we tell the difference between photons that come from dark matter annihilation and photons that are coming from some kind of unknown class of stellar processes that might be more common at the center of the galaxy than elsewhere?

The answer is that it isn’t easy, except in one special case. If dark matter particles (which have some definite mass, let’s call it M) can sometimes annihilate to two and only two photons, then both of those photons will have motion-energy equal (to a very, very good approximation) to the mass-energy Mc2 of the dark matter particles. The reason is very simple.  It is the same as described in this article on particle/anti-particle annihilation, and as seen in Figure 3.

Fig. 3: Suppose dark matter particles are their own anti-particles; then they may annihilate if they collide. Dark matter particles inside the galaxy are expected to move at speeds far below the speed of light and therefore to have motion-energies that are tiny compared to their mass-energies. If two dark matter particles of mass M are (nearly) at rest, they have only mass-energy and (almost) no motion-energy. If they come sufficiently close to each other, they may perhaps annihilate to two photons, both of which will carry off motion-energy (nearly) equal to the mass-energy of the initial dark matter particles.

If a particle and anti-particle are (nearly) at rest, then the energy of each is (almost) entirely mass-energy and (nearly) equal to Mc2.  Both have momentum (nearly) zero.  Energy and momentum are conserved, so the total energy is (nearly) 2Mc2 before the annihilation and after it too.  When the particle and anti-particle annihilate to a different particle and anti-particle, both the new particle and new anti-particle will have energy (nearly) equal to  Mc2. In general, this will be a mix of mass-energy and motion-energy.  In the specific case in which the final particle and anti-particle are photons, which have no mass and consequently no mass-energy, all of the energy will be in motion-energy.

Now we don’t know what the mass M of the dark matter particles is, and we don’t know therefore what the energies of the resulting photons will be. But since, just as every electron has the same mass and every proton has the same mass, every dark matter particle has the same mass M, every single dark matter annihilation will produce two photons of energy just about equal to Mc2. And that means that if we measure, with a special purpose telescope, the high-energy photons coming from the region near the center of the galaxy, and we make a plot of the number of photons that we detect with a given energy, we should expect astrophysical processes to generate lots of photons at lots of different energies, forming a smooth background, but the dark matter processes will add a bunch of photons that all have the same energy — a bump sticking up above that background.  See Figure 4.  It’s almost impossible to imagine any astronomical object, such as a bizarre star, that would be simple enough to generate a bump of this sort, so a signal in the form of a narrow bump would be a smoking gun for pairs of dark matter particles annihilating.

Fig. 4: A plot of the number of photons observed from the galactic center, as a function of their energy E, will show a broad distribution that falls with E that gets contributions from many types of astronomical sources. Any dark matter annihilating to two photons will produce a narrow bump on this plot, at the energy E equal to the mass M of the dark matter times c-squared. For the reason, see Figure 3.

This gives us a very powerful way to look for dark matter.  It won’t work if dark matter particles aren’t their own anti-particles and can’t annihilate at all. It won’t work if dark matter particles don’t often make photons when they annihilate. But it might work. And so there are efforts ongoing, most notably using the Fermi Large Area Telescope, a satellite experiment, which is out in space now, measuring photons coming from all across the sky, including those coming from the galactic center.

40 responses to “Seeing Signs of Dark Matter Annihilation

  1. Love the simplicity of your reasoning. Always look for the most basic answer. Bet its not that simple though, when ever has it been in your field. Hope its right though. Next stop Dark Energy?

    • Unfortunately, we have far fewer strategies for learning about dark energy. As of now, we really don’t have any way of getting at it that isn’t through its gravitational effects.

  2. Thanks for the article.There is some debate on the Bad Astronomy blog (all non-experts, of course) over the existence of Dark Matter. Quite a few people seem to think that something is wrong with our concept of gravity, and that inventing new particles is not the answer. On the other hand, I tend to think that the standard model of cosmology must be standard for a reason. Anyone without deep knowledge of cosmology is not qualified to argue one way or the other.

    Everything (credible) I’ve read on the subject seems to agree with the existence of Dark Matter. Could you briefly summarize why that theory is much preferred over some type of modication to current theories?

    • Debate on such issues is healthy, but indeed the evidence for dark matter, which was moderate at best when I was a graduate student, is now very strong, and comes from many types of observations, not just one or two. Examples: galactic rotation curves, motions of galaxies inside galactic clusters, Einsteinian gravitational lensing (strong and weak) on various distance and size scales, details of the cosmic microwave background radiation, simulations of galaxy formation, dwarf galaxies with very few stars, etc. This doesn’t make it impossible to think up an alternative to dark matter, but it does make it very much more difficult than it was 20 years ago.

      However (a) to present all the evidence in a complete fashion will be a long article of its own, which perhaps I’ll write in future; (b) an expert who knows more of the astrophysics and cosmology than I do ought to write it, because I’ll end up leaving important things out [in fact I would check out whether Sean Carroll or someone else at Cosmic Variance (a very reputable blog) has a long article on this already]; and (c) it’s not relevant to today’s article.

      The reason it’s irrelevant is that today I’m just explaining to you how you could find certain types of dark matter if they exist. If they don’t exist, well, then when you look for these photons from dark matter annihilation, you won’t see any… that’s all. This is no big deal; we look for all sorts of things, plausible and not so plausible, that turn out not to be present in nature.

  3. Four observations:

    1) You are still assuming that all the dark matter particles have the same mass, M. If there is a distribution of masses then the background plot will remain smooth, annihilation without the “bump”.

    2) If, (I say if because I don’t know if it is valid, but I am sure you can confirm either way), the galaxy black hole is also centered about the dark matter “sphere” then the black hole would be another common denominator that we could use to make any sense of the different properties between visible and dark matter. Ex: the process of two dark particles annihilating to create two photons could be working in both directions. The visible particles could be annihliating, (sucked into the black hole) at the black hole event horizon while the two photon creation process is done further out from the center.

    3) This is by extension from the example I gave above. Apart from the gravitational effects of the stars orbits and the bending of light, the fact that galaxies are not (have not yet, anyway) collapsing into the black hole could suggest that there should be a “balance” of visible and dark matter particles flip flopping back and forth.

    4) I am sorry but my fourth observation is a doomsday scenario. The dark matter particle live in the region of the EMF spectrum that our instrument cannot read yet. Since the dark matter is particularly at rest, then could this indicate that dark matter is our “death” phase in the Big Bang scheme of the evolution process?

  4. A great article as usual. Particularly as it comes hot on the heels of the report by ESO that they can’t see any signs of dark matter in the local area of our sun according to their article yesterday at http://t.co/VICx1gVy . I guess that is just one piece of conflicting evidence that may explain itself if we do finally discover what it is though. Exciting times to be alive.

    • Thanks for pointing out that link. I will look into that carefully.

      Remember two things though.

      1) It’s one group’s result. It may be wrong. Remember OPERA. It is premature to get too excited until a second group agrees.

      2) It is sometimes forgotten that dark matter may well not be smoothly distributed. In fact I indicated that in Figure 1. We do not know if dark matter near us is roughly smooth or in smaller clumps. There’s a lot of work being done on this. So even if this ESO result is right, it might just mean dark matter is very clumpy. We’ll need a lot more information before we could discard all of the diverse lines of evidence in favor of dark matter.

      But of course we should take this result seriously, try to find any mistakes first, then think through possible explanations, and see if we can find ways to get more insights into what it means.

  5. Thank you, Professor. Is Figure 1 hypothetical, particularly relating to the lumps? To my mind, lumps rule out some kind of effect generated by galaxy-sized singularities.

    • Figure 1 is hypothetical in its details; only its rough form is believed to be correct. In particular we do expect there to be lumps in the dark matter distribution (from numerical simulations) but we don’t know how large they are, how many they are, how they are distributed around our galaxy and other galaxies, etc., partly because we don’t have sufficient data from observations and partly because the existing numerical simulations are not yet sufficiently reliable.

  6. Dirck Uptegrove

    A google of “annihilation galactic center” brings up a number of hits proposing a similar idea. For example there is a paper using the Fermi Gamma Ray Space Telescope that seems to do just what you propose: arXiv.org > astro-ph > arXiv:1110.0006

    • I didn’t propose anything here; this is an educational website, and I was explaining a well-known technique. I first learned about this technique when I was a graduate student, over 20 years ago. I think it was probably proposed 30 years ago. Let’s see.

      Here’s an experimental attempt from 2006.

      H.E.S.S. Collaboration, F. Aharonian et. al., HESS observations of the galactic center region and their possible dark matter interpretation, Phys. Rev. Lett. 97 (2006) 221102, [astro-ph/0610509].

      Here’s theoretical calculations from 1989.

      Rate for Annihilation of Galactic Dark Matter Into Two Photons. Gian F. Giudice (Fermilab), Kim Griest (Chicago U., Astron. Astrophys. Ctr. & Fermilab). Published in Phys.Rev. D40 (1989) 2549

      which refers to earlier theoretical work from 1988.

      More related theory (not quite the same process, but with a similar signal) from 1986:

      Probing the Structure of the Galactic Halo with gamma Rays Produced by WIMP Annihilations. Michael S. Turner (Fermilab & Chicago U., Astron. Astrophys. Ctr.). Mar 1986. 14 pp. Published in Phys.Rev. D34 (1986) 1921

      based on work by Silk and Srednicki from even a couple of years earlier.

  7. Why would normal and dark matter differ in terms of large-scale structure? Why would dark matter form a halo and normal matter structures like spiral arms?

    If there was dark matter in the solar system, why wouldn’t it exist in the form of (possibly unseen) planets? Or even as part of the sun, since it would also attract dark matter by gravity?

    • For the first question: It’s pretty complicated. The reason normal matter shrinks down and forms a disc inside a vaguely spherical ball of dark matter has to do with the fact that normal matter can lose energy by emitting photons, while dark matter cannot. Without the dark matter ball, by the way, it is hard to understand why you get a disc of ordinary matter. Here are some related lectures by an astronomer

      http://www.ifa.hawaii.edu/users/shadia/lectures/habbal_astro110-01_spring2009_lecture34.pdf

      For the second question: we know almost nothing about dark matter. There may be a bit of dark matter in the sun. There may be small and dense clusters or aggregates of dark matter in the solar system. Nobody would know at this point. But if the chunks were as massive as planets, we’d easily see unexplained deviations in the orbits of the visible planets. And if there were a very large amount of dark matter in the sun, it would change the temperature at the center of the sun, which (because of our measurements of neutrinos and of sun-quakes) we know very accurately.

      Also, there’s no reason to think that there should be exceptionally large amounts of dark matter in the solar system. The solar system is relatively dense with ordinary matter only because of the details of star formation, which isn’t just due to gravity; there is a lot of complicated physics in which cooling by emission of photons, and strong magnetic fields, apparently play a role. Dark matter would be immune to these effects.

  8. Do we have a rough idea of what the upper and lower mass limits are for the dark matter particle?

  9. Beautiful article. But I have a question: in a virtual creation-annihilation particles, as well occured in a space-time at each second, how you can separated it from a dark-matter ocurrence? You dont quoted anythig about virtual particles. Thanks.

    • I am not sure quite what your question is asking, but if you are asking whether virtual particles would have any sort of effect the answer is no. Virtual particles are note nicely behaved like real ones. While often imagined as particle-antiparticle pairs that ‘pop up’ then annihilate a short while alter, this is not an accurate description. Notably, virtual particles cannot produce real particles that we can then detect.

  10. Thanks for the article. I would like to ask:

    1) How do the dark matter particles making up the galactic halo prevented from simply collapsing into the center of the galaxy? Do the dark matter particles undergo orbital motion about the galactic center?

    2) How would the dark matter particles be able to annihilate into photons if they did not carry electromagnetic charge?

    Thanks.

  11. Hi Matt, thank you for this great website! I notice that you do not seem to mention the possibility that the dark matter particle could be found with the LHC (or maybe you do somewhere, and I have overlooked it….) Why is this, do you think this is probably not going to work out? If yes, why?

    • Take a look at this post: http://profmattstrassler.com/2012/03/05/news-from-la-thuile-with-much-more-to-come/

      We don’t have any way to know if dark matter particles can be produced in abundance at the LHC. It may be a very rare process, perhaps very hard to detect, perhaps impossible. (Obviously you don’t see the dark matter particles, but you see other particles that recoil against them.)

      • Thank you! This is interesting.
        Here are two more questions (if you have time):

        1. If you have a collision, and all sorts of particles flying off in all sorts of directions, how will you know that one of them experienced a recoil because it flew into the newly produced dark matter particle? Or is that not what you mean?

        2. What if the dark matter is not weakly interacting, but only gravitationally? Does this mean that there is zero chance to see it at the LHC?

        Thank you!

        • I should write a short article on this; point is, if you create undetectable particles (even, say, neutrinos) in a proton-proton collision, and you don’t make anything else with signficant energy, there’s no way for you to know that you did so. But if at the same time you make the neutrinos (or dark matter) you produce an energetic gluon that goes flying sideways, what you’ll see in your detector is a high-energy “jet” (a spray of hadrons, http://profmattstrassler.com/articles-and-posts/particle-physics-basics/the-known-apparently-elementary-particles/jets-the-manifestation-of-quarks-and-gluons/ ) heading, let’s say, upward, while you’ll see nothing in the detector moving downward. Since momentum is conserved, and the initial protons came in sideways, the only way something could be going up is if something else went down. So that tells you that you must have produced some undetectable particles.

          To figure out they are something new, and not neutrinos, requires more in depth study of exactly how often this happens and how, in a large class of similar events, the energy of the jet is distributed.

          And to figure out that the new undetectable particles are dark matter? Oh, that’s not easy at all! That will require combining this discovery with other discoveries using other methods.

  12. I guess the assumtion of dark matter being it’s own antiparticle is made to ensure enough events since ‘distinct’ antimatter is far less abundant. However, do we know enough about dark matter to assume such asymetry exist for it as well? Couldn’t there be enough antidarkmatter even if it weren’t it’s own antiparticle?

  13. Could dark matter contain its own anti-particle, but never interact because of a virtual particle field that keeps them in perfect orbit of each other? Could that be why it has a non-interacting disposition?

    • Such a thing is very difficult to make work. A similar situation exists with electrons and positrons, they will create a ‘stable’ ‘atom’ of positroium, both particles ‘orbiting’ each other. Sadly because of the laws of physics they actually spend their time more in a sort of cloud and there is a chance that while both in that cloud they will get too close to each other and annihilate.

      In order to work your dark matter would need a new force to hold it together and it would need to work in a way that the particles got close, but not too close. Even then energetic dark matter particles wouldn’t want to settle into that arrangement and could annihilate.

  14. All,

    Dark matter annihilation is happening everyday inside the earth. The charged particles help create the auroras and our ionosphere and magnetohere. It is all on my blog at darkmattersalot.com

  15. Pingback: Old Problems, New Techniques « Space « Science Today: Breaking science news from around the world

  16. Great article professor! I remember reading somewhere that there may be a blackhole at the centre of the milky way. If so, why doesn’t the dark matter and all the matter around the centre of the universe get sucked in?
    Thanks.

    • Black holes are not ‘cosmic vacuum cleaners’; they cannot ‘suck things in’ any more than our sun can. As such nearly all the matter in our galaxy is orbiting its central black hole, only a tiny bit is falling in. (And we have seen it swallowing matter rather recently: http://www.nasa.gov/mission_pages/nustar/news/nustar20121023.html )

      Black holes are only inescapable monsters close up, most of the stuff in the galaxy is at a safe distance.

  17. May I ask something..
    why do we believe that DM particles may interact weakly? Have we got any actual data on it or is it just speculation?
    Thank you

  18. isn’t it that dark matter particles don’t interact with each other or visible (normal) matter?

  19. Ordinary matter can be transformed to ordinary energy. So dark matter can be transformed to dark energy. Therefore dark matter is concentrated dark energy. Balloon inside balloon theory of matter and antimatter universe on opposite entropy path producing dark energy at common boundary by annihilation and injected into both the universes causing a swirl and whirl of gravitoetherton soup taking galaxies,stars etc in its rotation we know as rotational orbits. The strength of this soup is varying across universe so that Newtons equation is F=P.G.M.m/R.R where P is permeability. Now there is no HIGGS FIELD but a Dark energy field of gravitoethertons. GRAVITY is the mono magnetic push of gravitoethertons on molecules of matter and we see the effect of GRAVITY OR EVEN AVOGADROS LAW IN ENCLOSED GAS ETC ETC. The revised atomic model says that protons are very close to central neutron as charge cloud but not merged with neutrons. As such the residual charge and mono pole magnetism is very interesting to cause gravity because the flow of gravitoethertons are towards center of earth for creating hot core and earths magnetism. Similarly our galaxy center is also getting focussed gravitoethertons to cause a dark matter center we call black hole. BLACK HOLES ARE CONCENTRATED GRAVITOETHERTONS CAUSING BLACK HOLE. AT CENTER OF ALL GALAXIES. Pear shaped atomic model is now known to investigate further Rutherford and Bohr ideas for NEW ATOMIC MODEL. Matter and antimatter universe on OPPOSITE ENTROPY PATH causes the RECYCLIC ,RE BOUNCE MODEL WHEN ONE UNIVERSE REACH TENDS TO ZERO ENTROPY. As such the rebounce starts with DR.GUTH exponential expansion ,then accelerating expansion. BIG BANG IS WRONG. ALL THESE THEORIES ARE EXPLAINED AND CERN CONFIRMED THAT TWO NEW PARTICLE THE DISCOVERED ARE GRAVITON AND ANTI GRAVITON OF GRAVITOETHERTON SOUP CREATED AT THE COMMON BONDARY. More we can discuss if interested to know further.

  20. What is a “gravitoetheron” ?

    • Gravitoethertons is a pack of force carriers including weak and gravity etc. LHC has cracked two we believe graviton and anti graviton. May be with higher energy,we will see other force carriers. But revised atomic model does not require strong force . We have to recalculate standard model with modified relativistic approach considering a scaffolding of dark energy we have not taken into account so far. The process is complex mathematical along with results from LHC and we have to wait for a new physics for recalculation. We have to go beyond Einstein and standard model.

  21. Intriguing ideas here about DM and the aurora, but no mention of the Majorana Fermion which is a candidate particle as a constituent of DM, as well as useful for quantum computing. If Matt is right about annihilation, this would be a great particle to collect and make an interstellar drive out of!

    Better than science fiction.

  22. Professor Strassler,

    First, please let me thank you for all the time you spend maintaining this web site and particularly for extending your incredible knowledge and expertise to those, like myself, who never became physicists but always had a fascination or passion for it. I was wondering if you had a chance to read this recent paper posted on arXiv regarding the recently discovered X-ray line at ~3.5 keV and the authors’ hypothetical explaination for why sterile nuetrino DM with a mass ~7 keV may account for it. To a layperson such as myself, it seems very convincing. I’ve posted the link below:

    http://arxiv.org/abs/1402.5837v1

    Thank you for your time sir.

    Matt Chambers

  23. I wish you well. I find it surprising that plots like figure 4 haven’t already been generated from pointing a telescope towards the center of the galaxy… it’s such a big target, and close too. Admittedly, the idea of the existence of dark matter has never really sat well with me; I understand that astrophysicists don’t really understand the mechanism for the accelerating expansion of the universe and came up with the idea of dark matter as a fudge factor. Under your assumption that dark matter can self annihilate, if you can find a little bump in the photon energies, well, I would be little less skeptical about the whole thing (not that my skepticism really has any bearing on the astrophysical community at all).

    Here is another idea in particle physics that has gnawed at me… in production reactions where photons combine to form matter, antimatter is always generated as a product as well. So, at the beginning, during the big bang if all of the matter in the universe was generated from photon reactions, then there should be a lot of antimatter floating around somewhere as well. If I understand correctly, we don’t detect of the correct energy signature for gamma radiation that should be spiking all over the place if antimatter were to react with matter somewhere within our light radius. So, assuming that antimatter was generated during the big bang, is it possible that most or almost all of it is outside of the light radius?

    Does antimatter have the same gravitational pull inwards that matter does? Does that gravitational pull become a push when reacting with matter’s gravitational field? Is it only charge that gets reversed between matter and antimatter? What is the mechanism that causes matter and antimatter to annihilate, anyways? Why do they do that — annihilate, I mean? Are there cross-force reactions that can occur in nature — like the weak force interacting with the electromagnetic force, is this possible?

    1 – Electro-Magnetic force carriers
    2 – Gravitational force carriers
    3 – Weak nuclear force carriers
    4 – Strong nuclear force carriers

    We have lots of physics and equations to describe interactions like 11, 22, 33, 44… but do weird interactions like: 12, 13, 14, ever occur? Do interactions between groups of 3 fundamental forces at a time ever occur like: 123, 124, etc? If they did, I suppose they would be pretty difficult to isolate, wouldn’t they?

    Goodnight Dr. Strassler

  24. To be clear:
    22 -> represents a thing with mass feeling a pull from another thing that has mass.
    12 -> represents a thing with charge whose EM field somehow interacts with the Gravitational field of another thing to cause a push or a pull.

    It would be really interesting if these kinds of interactions could occur.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s