Tag Archives: dark matter annihilation

Gamma Rays from Dark Matter at the Center of the Galaxy: Stronger Evidence

Evidence has been growing for the detection of dark matter more directly – at the center of the Milky Way Galaxy. Normally, we detect dark matter through its gravitational effects only, although there have been many attempts to detect it more directly, both through laboratory experiments here on Earth and from astronomical measurements. The Earthbound experiments are inconclusive at best, with some claims of detection being contradicted by other experiments.

But the evidence for astronomical detection of dark matter is growing. Expected sources include dwarf galaxies that are found near our Milky Way. The low luminosity of dwarf galaxies due to stars and supernovae can make it easier to extract evidence of dark matter due to its self annihilation.

Our own Milky Way Galaxy has a higher concentration of normal matter at the center, and is expected to have a higher concentration of dark matter as well. For the past 5 years or so, there has been evidence for possible dark matter annihilation at the Galactic Center. See

The mechanism is dark matter self-annihilation, resulting in the creation of decay products of ordinary matter and gamma rays (highly energetic photons). See one of my prior blogs at:

The leading dark matter candidate is some sort of WIMP (weakly interacting massive particle). WIMPs interact only via gravity and perhaps the weak nuclear force. WIMP self-annihilations can produce quarks, neutrinos, gamma rays and other ordinary matter particles.

There is a known gamma ray signal in the Galactic Center (the center of our Milky Way) that extends to 5 degrees away from the center, corresponding to roughly a kiloparsec in extent (a kiloparsec is 3260 light-years, and our Sun is 8 kiloparsecs from the Center). The major alternatives for this signal appear to be dark matter annihilation, cosmic ray interactions with interstellar gas, or emission from rapidly rotating neutron stars (millisecond pulsars).

A recent paper from T. Daylan and co-authors from Harvard, MIT, Princeton, the University of Chicago and the Fermi Laboratory is titled “The Characterization of the Gamma-Ray Signal from the Central Milky Way: A Compelling Case for Annihilating Dark Matter”. They have reanalyzed observations from the Fermi Gamma Ray Space Telescope and confirmed that the distribution of gamma rays in the Galactic Center (GC) is largely spherically symmetric and extended. This spatial distribution likely rules out neutron stars as the source, since these are preferentially found in the galactic disk. 


1-3 GeV residual gamma ray image. From Fig. 10 of T. Daylan et al., this is corrected for galactic diffuse emission and has point sources subtracted. The image extends over a 10◦ by 10◦ region.

Dark matter, on the other hand, would be expected to have a roughly spherical distribution around the GC. Interstellar gas is also largely confined to the galactic disk, so this explanation is disfavored. Their study also confirms that the emission extends beyond the GC to what is known as the Inner Galaxy, further ruling out the two alternatives other than dark matter annihilation. The emission falls off in intensity away from the GC, in a manner consistent with a spherically symmetric dark matter density distribution that is in accordance with a Navarro-Frenk-White profile often used successfully in modeling dark matter halos. No evidence is found for any significant deviation from spherical symmetry for the GC and Inner Galaxy components, the latter extending out to around 2 kiloparsecs.

There are various possible annihilation channels for dark matter and the authors’ analysis appears to favor a dominant channel to primarily b quarks (and b antiquarks). In this scenario the WIMP mass appears to lie in the range of 36 to 51 GeV (by comparison a proton or neutron mass is about .94 GeV). Recall that there are 6 types of quarks, and protons and neutrons are composed of u and d (up and down) quarks. The others are b, t, c, s (bottom, top, charm and strange). The quarks other than u and d are unstable and will decay to u and d.

The spectral (energy) distribution peaks at gamma ray energies of around 1 to 3 GeV and is a good fit to the predictions for annihilation to a b quark pair (b and anti-b). In addition, the cross-section for annihilation calculated from the gamma ray intensity is consistent with that expected from the required rate of thermal production of dark matter particles in the early universe, of order 10^-26 cm^3/sec (actually a value of the cross-section multiplied by the average velocity). The observed dark matter abundance freezes out from thermal equilibrium in the early universe as it expanded and cooled, and implies a cross section of that order.

There is also the possibility for other decay channels, including decays to u, d, c, s and t quarks and to tau lepton particles. The spectral shapes disfavor decays to tau leptons and u, d quarks in particular. After decays to b quarks, the c (charm) and s (strange) quark channels are the most likely.  Either a c or s quark channel implies somewhat lower WIMP masses, around the 20 to 40 GeV range. Annihilations to other fermions appear less likely.

In summary, quoting from their paper:

“This signal consists of a very large number of events, and has been detected with overwhelming statistical significance. The the excess consists of ∼ 10,000 gamma rays per square meter, per year above 1 GeV (from within 10◦ of the Galactic Center). Not only does this large number of events enable us to conclude with confidence that the signal is present, but it also allows us to determine its spectrum and morphology in some detail. And as shown, the measured spectrum, angular distribution, and normalization of this emission does indeed match well with that expected from annihilating dark matter particles.”

“There is no reason to expect that any diffuse astrophysical emission processes would exhibit either the spectrum or the morphology of the observed signal. In particular, the spherical symmetry of the observed emission with respect to the Galactic Center does not trace any combination of astrophysical components (i.e. radiation, gas, dust, star formation, etc.), but does follow the square of the anticipated dark matter density.”

There are also possible detections, marginally significant, of gamma ray emission due to dark matter in nearby dwarf galaxies, and in the direction of the Virgo cluster. We look forward to additional observations and theoretical work to confirm dark matter annihilation signals in our own galaxy and nearby galaxies.

NEW BOOK just released:

S. Perrenod, 2016, 72 Beautiful Galaxies (especially designed for iPad, iOS; ages 12 and up)



Black Holes Destroy Dark Matter (and Emit Gamma Rays)

Black holes can cause dark matter to annihilate in their vicinity by concentrating the dark matter and enhancing the collision rate between dark matter particles. The best observational candidates are supermassive black holes, such as the 4 million solar mass black hole found at the center of our Milky Way galaxy. Some galaxies have much larger supermassive black holes, reaching as high as several billion or even tens of billions of solar masses. Most massive galaxies appear to have supermassive black holes in their centers.

Artist's conception of a supermassive black hole (public domain; courtesy NASA JPL)

Artist’s conception of a supermassive black hole (public domain; courtesy NASA JPL)

We infer the existence of supermassive black holes through their effect on nearby stellar or molecular cloud orbits. And we more directly detect supermassive black holes (SMBHs) by the radiation emitted from ordinary matter that is near the black hole (BH), but has not yet fallen into the BH’s event horizon (from which nothing, not even light, can escape). Such matter will often form a hot accretion disk around the SMBH. The disk or other infalling matter can be heated to millions of degrees by the strong gravitational potential of the BH as the kinetic energy of infall is converted to thermal energy by frictional processes. Ordinary matter (OM) heated to such high temperatures will give off X-rays.

Now if OM is being pulled into a SMBH, so is dark matter, which pervades every galaxy. Dark matter (DM) responds to the same gravitational potential from the SMBH. The difference is that OM is collisional since it easily interacts with other OM via the electromagnetic force, whereas DM is generally collisionless, since it does not interact via electromagnetism.

Nevertheless DM – DM collisions can occur, rarely, via a ‘direct hit’ (as if two bullets hit each other in mid-air) and this leads to annihilation. Two DM particles meet directly and their entire energy content, from their rest mass as well as their kinetic energy of motion, is converted into other particles. The cross-section strength is not known, but it must be small due to observational limits, yet is expected to be non-zero. The most likely candidates for decay products are expected to be photons, neutrinos, and electrons.

The leading candidate for DM is some sort of weakly interacting massive particle with a mass of perhaps 5 to 300 GeV; this is the range where DM searches from satellites and on Earth are focused. (The proton mass is a little less than 1 GeV = billion electron Volts.) So if two DM particles mutually annihilate, there is of order 10 GeV to 600 GeV of available rest mass energy to produce highly energetic gamma rays.

The likelihood of a direct hit is proportional to the square of the density of the DM. A SMBH’s gravitational potential acts as a concentrator for DM, allowing the density to be high enough that there could be a significant number of annihilation events, resulting in a detectable flux of escaping photons reaching Earth. Relativistic effects work to further increase the annihilation rate. And it is possible that the annihilation signal could scale as M³ (mass of the SMBH cubed), and thus the most massive SMBHs would be very strong gamma ray emitters. These would be highly energetic gamma rays with well over 1 GeV of energy.

Movie from NASA Goddard showing Dr. Jeremy Schnittman’s simulation

Dr. Jeremy Schnittman of the NASA Goddard Space Flight Center has investigated possible annihilation rates and the nature of the observable gamma ray spectrum for some simple dark matter models. He used a compute cluster to simulate hundreds of millions of DM particles moving in the general direction of a SMBH. One of his remarkable findings is that much higher gamma ray energies can be produced than previously believed, in the case of SMBHs which are rapidly spinning.

This is a result of something known as the Penrose process, which allows energy to be extracted from a rotating BH. There is a region called the ergosphere outside of the event horizon and when two DM particles annihilate in this region and produce two gamma rays, one gamma ray photon would fall into the event horizon (into the BH), and the other photon would escape to infinity, possibly in the direction of Earth. Dr. Schnittman’s simulation indicates that the energy boost can be as high as 6 times or more. The faster the SMBH is spinning, the greater the potential energy boost.

He also has looked at DM particles on bound orbits, which are likely to form into a (donut-shaped) corotating torus around the SMBH, aligned with its spin vector. The bound DM particle annihilations lead to lower energy gamma ray production, as compared to the unbound particles.

One of the important considerations is that the influence radius of the BH is very large. The size of the BH itself (event horizon or Schwarzschild radius) is small, even for SMBHs. The radius is proportional to the mass, via the relation 2GM/R = c² (G is the gravitational constant, c the speed of light and M and R are the BH mass and radius, respectively). A SMBH with a mass of 10 million solar  masses will have a radius of only around 30 million kilometers, or about 1/5 of the Earth-Sun distance (an AU, or astronomical unit).

But the gravitational influence is much greater, since DM particles are typically expected to be moving at around only a couple of hundred kilometers per second far away from the SMBH. Thus DM particles that are 1 million times further away than the SMBH will have their orbits in their galaxy perturbed by the SMBH. And the scale of influence is thus parsecs (1 parsec = 3.26 light-years) or tens of parsecs or even hundreds of parsecs, depending on the SMBH mass.

The most energetic gamma rays can be produced by unbound DM particles. These are on orbits which can approach near to the SMBH after falling from far away (a “swan dive” toward the SMBH) and these DM particles would then typically head out away from the SMBH in the opposite direction. But before they are able to, they have a direct hit with another DM particle and annihilate into gamma rays or some other decay products.

The search for gamma rays from annihilating DM around SMBHs is already underway. There is in fact a possible detection by the Fermi telescope at 130 GeV in our Milky Way galaxy, from the direction of the Sagittarius A* SMBH. Future more sensitive gamma ray surveys may lead to many detections, helping us to better understand both dark matter and black holes.


J.D. Schnittman, 2015. “The Distribution and Annihilattion of Dark Matter around Black Holes”,

J.D. Schnittman, 2014. Phys. Rev. Letters 113, 261102,  “Revised Upper Limit to Energy Extraction from a Kerr Black Hole”

Dark Matter Powered Stars

Gamma Ray Burster 070125

GRB (gamma ray burster) 070125. Credit: B. Cenko, et al. and the W. M. Keck Observatory.

So what is a “dark star”? It is not a Newtonian black hole as suggested by John Michell in the 18th century who used the term while postulating that gravity could prevent light escaping from a very massive, compact star. It is not a “dark energy star”, which is related to a black hole, but rather than having a singularity at the center, quantum effects cause infalling matter to be converted to vacuum state energy, dark energy. It is not a comic book, science fiction comedy film, or Grateful Dead song.

In this blog entry we are writing about dark matter powered stars. These would be the very first stars, formed within the first few hundred million years of the universe’s existence. The working assumption is that dark matter consists of WIMPs – weakly interacting massive particles, in particular the favored candidate is the neutralino. The neutralino is the lightest particle among the postulated supersymmetric companions to the Standard Model suite of particles. As such it would be stable, would not ordinarily decay and is being searched for with XENON, CDMS, DAMA, AMS-02 and many other experiments.

The first stars are thought by astrophysicists to have been formed from clouds of ordinary hydrogen and helium as well as dark matter, with dark matter accounting for 5/6 of the total mass. These clouds, called “dark matter halos” are considered to have contained from about one million to 100 million times the Sun’s mass. The ordinary matter would settle towards the center as it cooled via radiation processes and the dark matter (which does not radiate) would be more diffuse. The stars forming at the center would be overwhelmingly composed of ordinary matter (hydrogen and helium nuclei and electrons).

Without any dark matter at all, ordinary matter stars up to about 120 to 150 solar masses could form; above this limit they would have very hot surfaces and their own radiation would inhibit further growth by infall of matter from the halo. But if as little as one part in a thousand of the protostar’s mass was in the form of dark matter this limitation goes away. The reason is that the neutralino WIMPs will, from time to time, meet one another inside the star and mutually annihilate since the neutralino is its own anti-particle. The major fraction of the energy produced in the annihilation remains inside the star, but some escapes in the form of neutrinos (not neutralinos).

Annihilation of these neutralinos is a very efficient heating mechanism throughout the volume of the star, creating a great amount of heat and pressure support, basically puffing up the star to a very large size. The stellar surface is, as a result, much cooler than in the no dark matter case, radiation pressure is insignificant, and accretion of significantly more material onto the star can occur. Stars could grow to be 1000 solar masses, or 10,000 solar masses, potentially even up to one million solar masses. Their sizes would be enormous while they were in the dark matter powered phase. Even the relatively small 1000 solar mass star, if placed at the Sun’s location, would extend through much of our Solar System, beyond the orbit of Saturn.

We have mentioned the neutralino meets neutralino annihilation mechanism. A second mechanism for heating the interior of the star would be direct impact of neutralinos onto protons and helium nuclei. This second mechanism could help sustain the duration as a dark matter powered star potentially even beyond a billion years.

Eventually the dark matter fuel would be exhausted, and the heat and pressure support from this source lost. The star would then collapse until the core was hot enough for nuclear fusion burning. Stars of 1000 solar masses would burn hydrogen, and later helium, and evolve extremely rapidly because of the high density and temperature in their cores. After their hydrogen and helium fusion cycles completed there would be no source of sufficient pressure support and they would collapse to black holes (or maybe dark energy stars).

It is calculated with detailed simulations that the dark star mechanism allows for much more massive stars than could be formed otherwise, and this provides a potentially natural explanation for the creation of massive black holes. Our own Milky Way has a black hole around 3 million solar masses at its center, and it appears a majority of galaxies have large black holes. The image at the top of this blog is of a gamma ray burst detection that may have come from a large black hole formation event.


Freese, K. et al 2008, Dark Stars: the First Stars in the Universe may be powered by Dark Matter Heating,

Freese, K. et al 2010, Supermassive Dark Stars,