Did LIGO Detect Dark Matter?

It has often been said, including by me, that one of the most intriguing aspects of dark matter is that provides us with the best current evidence for physics beyond the Core Theory (general relativity plus the Standard Model of particle physics). The basis of that claim is that we have good evidence from at least two fronts — Big Bang nucleosynthesis, and perturbations in the cosmic microwave background — that the total density of matter in the universe is much greater than the density of “ordinary” matter like we find in the Standard Model.

There is one important loophole to this idea. The Core Theory includes not only the Standard Model, but also gravity. Gravitons themselves can’t be the dark matter — they’re massless particles, moving at the speed of light, while we know from its effects on galaxies that dark matter is “cold” (moving slowly compared to light). But there are massive, slowly-moving objects that are made of “pure gravity,” namely black holes. Could black holes be the dark matter?

It depends. The constraints from nucleosynthesis, for example, imply that the dark matter was not made of ordinary particles by the time the universe was a minute old. So you can’t have a universe with just regular matter and then form black-hole-dark-matter in the conventional ways (like collapsing stars) at late times. What you can do is imagine that the black holes were there from almost the start — that they’re primordial. Having primordial black holes isn’t the most natural thing in the world, but there are ways to make it happen, such as having very strong density perturbations at relatively small length scales (as opposed to the very weak density perturbations we see at universe-sized scales).

Recently, of course, black holes were in the news, when LIGO detected gravitational waves from the inspiral of  two black holes of approximately 30 solar masses each. This raises an interesting question, at least if you’re clever enough to put the pieces together: could the dark matter be made of primordial black holes of around 30 solar masses, and could two of them have come together to produce the LIGO signal? (So the question is not, “Are the black holes made of dark matter?”, it’s “Is the dark matter made of black holes?”)

LIGO black hole (artist's conception)

This idea has just been examined in a new paper by Bird et al.:

Did LIGO detect dark matter?

Simeon Bird, Ilias Cholis, Julian B. Muñoz, Yacine Ali-Haïmoud, Marc Kamionkowski, Ely D. Kovetz, Alvise Raccanelli, Adam G. Riess

We consider the possibility that the black-hole (BH) binary detected by LIGO may be a signature of dark matter. Interestingly enough, there remains a window for masses 10M≲Mbh≲100M where primordial black holes (PBHs) may constitute the dark matter. If two BHs in a galactic halo pass sufficiently close, they can radiate enough energy in gravitational waves to become gravitationally bound. The bound BHs will then rapidly spiral inward due to emission of gravitational radiation and ultimately merge. Uncertainties in the rate for such events arise from our imprecise knowledge of the phase-space structure of galactic halos on the smallest scales. Still, reasonable estimates span a range that overlaps the 2−53 Gpc−3 yr−1 rate estimated from GW150914, thus raising the possibility that LIGO has detected PBH dark matter. PBH mergers are likely to be distributed spatially more like dark matter than luminous matter and have no optical nor neutrino counterparts. They may be distinguished from mergers of BHs from more traditional astrophysical sources through the observed mass spectrum, their high ellipticities, or their stochastic gravitational wave background. Next generation experiments will be invaluable in performing these tests.

Given this intriguing idea, there are a couple of things you can do. First, of course, you’d like to check that it’s not ruled out by some other data. This turns out to be a very interesting question, as there are good limits on what masses are allowed for primordial-black-hole dark matter, from things like gravitational microlensing and the fact that sufficiently massive objects would disrupt the orbits of wide binary stars. The authors claim (and quote papers to the effect) that 30 solar masses fits snugly inside the range of values that are not ruled out by the data.

The other thing you’d like to do is figure out how many mergers like the one LIGO saw should be expected under such a scenario. Remember, LIGO seemed to get lucky by seeing such a big beautiful event right out of the gate — the thought was that most detectable signals would be from relatively puny neutron-star/neutron-star mergers, not ones from such gloriously massive black holes.

The expected rate of such mergers, under the assumption that the dark matter is made of such big black holes, isn’t easy to estimate, but the authors do their best and come up with a figure of about 5 mergers per cubic gigaparsec per year. You can then ask what the rate should be if LIGO didn’t actually get lucky, but simply observed something that is happening all the time; the answer, remarkably, is between about 2 and 50 per cubic gigaparsec per year. The numbers kind of make sense!

The scenario would be quite remarkable and significant, if it turns out to be right. Good news: we’ve found that dark matter! Bad news: hopes would dim considerably for finding new particles at energies accessible to particle accelerators. The Core Theory would turn out to be even more triumphant than we had believed.

Happily, there are ways to test the idea. If events like the ones LIGO saw came from dark-matter black holes, there would be no reason for them to be closely associated with stars. They would be distributed through space like dark matter is rather than like ordinary matter is, and we wouldn’t expect to see many visible electromagnetic counterpart events (as we might if the black holes were surrounded by gas and dust).

We shall see. It’s a popular truism, especially among gravitational-wave enthusiasts, that every time we look at the universe in a new kind of way we end up seeing something we hadn’t anticipated. If the LIGO black holes are the dark matter of the universe, that would be an understatement indeed.

62 Comments

62 thoughts on “Did LIGO Detect Dark Matter?”

  1. So how many black holes are we talking about per galaxy? If there’s a huge spherical cloud of black holes around each galaxy, wouldn’t that sort of ‘blur’ the way it looks from afar, due to the cumulative effect of all that gravitational lensing?

  2. Would primordial black holes distribute like dark matter galaxy halos? And if not, how to explain the galaxy rotation mystery?

  3. Thanks Phillip — I guess I wasn’t really thinking about black holes as sucking up dark matter, but rather that matter and dark matter would statistically congregate near each other, and so there would already be a lot of dark matter hanging around (in direct proximity) when a star collapsed. I hadn’t thought about whether we would expect dark matter to radiate energy like that though, but I suppose I was left wondering why we wouldn’t expect nearly all massive objects in the universe to also have a lot of dark matter fluff hanging near them.

  4. I realize now though I was entirely misreading the post after reading Mike’s comment! Thanks Mike for clarifying we confused souls.

  5. “Why is that bad news?”

    In Sean’s Great Courses “The Higgs Boson and Beyond”, he exclaims how great it would be for physics if particle accelerators, while experimenting with previously unattainable particle energies, were to discover new particles, thus extending the Standard Model and leading to new horizons in physics. The bad news is that it may not happen.

    Also, in the Simeon Bird et al article, shouldn’t the formula Gpc(^−3) yr(^−1) be Gpc(^3( yr(^-1) – cubic parsecs per year?

  6. Could these primordial black holes (aka dark matter) have been produced right after the big bang with an excess of anti-matter over matter, so that this situation would explain not only dark matter, but also where the all the anti-matter went?

  7. “Could these primordial black holes (aka dark matter) have been produced right after the big bang with an excess of anti-matter over matter, so that this situation would explain not only dark matter, but also where the all the anti-matter went?”

    Probably not. As Carlo Rovelli wrote in his book about Anaximander: “But in science it is not difficult to come up with ideas; it is difficult to come up with workable ideas, to find a way to compose and atriculate new ideas as part of a whole that is consistent with the rest of our knowledge, and to convince others that the entire process is reasonable.” 😐

    Note also that since primordial black holes must have formed before nucleosynthesis, you would need something which has matter distinct from antimatter but no baryons.

  8. Bob Murley, assuming that your Q about the Bird article concerns the occurrence rate range in the abstract quoted by Sean-

    I believe the rate in question is a number of occurrences per unit volume, per year. That places the length unit and the time unit each squarely in the denominator, thus each with a negative exponent.

  9. These address the wide binary issue: “Searching for Dark Matter Constituents with Many Solar Masses” http://arxiv.org/abs/1510.00400 and “The Primordial Black Hole Mass Range” http://arxiv.org/abs/1511.08801

    The wide binaries of Monroy-Rodriguez and Allen (2014) is a flawed null experiment.

    Yoo et al. (2003), Carr et al. (2009), and Ricotti et al. (2007) were all written before AGN quasars at z>6 were known, and don’t comport with them.

  10. Phillip Helbig,

    I guess I don’t see the problem exactly. If (for some physics properties unknown to us) anti-quarks and positrons formed into anti-matter at a slightly sooner time than it took for quarks and electrons, then couldn’t it have made up slightly more of the percentage of stuff in these primordial black holes?

    I am not trying to propose new physics (although clearly there are several somethings we do not know about the big bang and the physics of the early universe). But sometimes several birds are actually killed with one stone.

  11. “If (for some physics properties unknown to us) anti-quarks and positrons formed into anti-matter at a slightly sooner time than it took for quarks and electrons, then couldn’t it have made up slightly more of the percentage of stuff in these primordial black holes?”

    Sure, but this is an idea you made up just to solve this problem. There is no supporting evidence. There is no other reason we should believe this. At the very least, you need something which this scenario predicts and the standard scenario doesn’t.

  12. “Sure, but this is an idea you made up just to solve this problem. There is no supporting evidence. There is no other reason we should believe this. At the very least, you need something which this scenario predicts and the standard scenario doesn’t.”

    OK… then my scenario simply predicts that there will be more matter than anti-matter in the current universe (excluding primordial black holes).

    There is abundant supporting evidence. Your argument is (please excuse my bluntness here) pretty weak. We (almost certainly) know that we can never tell what ‘type’ of mass is at the center of a black hole (or endlessly falling towards the ‘center’).

    Does this mean I win the argument?

  13. “OK… then my scenario simply predicts that there will be more matter than anti-matter in the current universe (excluding primordial black holes).”

    “Does this mean I win the argument?”

    No, because you did not predict anything which was not known before.

  14. “No, because you did not predict anything which was not known before.”

    I totally agree. As all of Einstein’s papers at first, and exactly as all of your own papers for ever.

    So we wait.

    (But seriously… I am not actually trying to be a rude troll. It’s just that at least one possibly true possibility will eventually be totally dismissed because there is no way to know the real reality. (If that even could make sense).)

  15. Einstein’s papers did of course make testable predictions. This is not the only valid criterion; something which explains known facts more elegantly also has a lot going for it, but this does not hold for “properties unknown to us”.

    Although I think of myself as a theorist, most of my papers are “measurement papers”. They report a measurement, i.e. an interpretation of observations in light of some theory. So, someone has to make the predictions, and someone has to verify them. (About a quarter of my papers point out errors in other analyses.)

    But a scientific idea, a theory, that neither predicts something other theories don’t, nor explains known data more elegantly, has nothing going for it.

    I’m not trying to be rude, but progress can be made only by doing real science. 😐

  16. After first reading this post, and getting my mind blown by the idea of black holes being the ‘particle’ in CDM, I realized there was something in the back of my head nagging me, though it refused to reveal itself.

    I re-read this post several times, trying to figure out what that nagging thought was, when one clue popped out: What velocity distribution must black-hole CDM (BHCDM) posses?

    It wouldn’t follow a Gas Law, else we’d see the occasional high-speed black hole splashing through gas clouds (long tails). And gas particles bounce in collisions, not get absorbed. No, BHCDM must be very cold indeed. But its velocity distribution can’t be all that much colder than the large-scale velocity profile of the conventional matter it surrounds, since that’s what keeps it gravitationally bound.

    But that was only part of what was nagging me: The idea of things zipping around with velocity distributions. The next thing that popped out was ‘density’, in this case, energy density as mass. The nagging thought had to do with velocity distributions of energetic things.

    Then finally my brain finally closed the loop with LIGO and BHCDM, taking me back decades.

    About 20 years ago I was involved in an early dirt-cheap satellite effort to do real, hard science in LEO on a shoestring budget (a few $K, with a free piggy-back launch). The first requirement was to avoid, literally at all costs, the use of pricey rad-hard circuitry. The second was to figure out how to get reliable performance from COTS (commercial off-the-shelf) components and circuits.

    That effort required delving into the statistics of particulate radiation spectra in LEO and MEO, modeling their effects on commercial silicon, then designing experiments to validate the models, which involved an amazing trip to Brookhaven National Labs to let their Tandem Van de Graff accelerator hurl relativistic heavy ions at my naked (de-lidded) prototype circuits.

    Anyhow, to cut a long (but fun) story short, I found myself wondering if the statistics that modeled that radiation and its effects would be at all similar to that for BHCDM interactions.

    The nagging thought was quickly dropped when I realized I hadn’t modeled radiation self-interactions, and that an entirely different class of stats would be needed. Still, it was a fun to have that old work recalled by such an immediate and important event.

    But I was left with one small question: Why doesn’t every star have one or more small BHCDM black holes at its center? (I was thinking along the lines of diffusion and mixing of clumps of normal matter with BHCDM, especially during galaxy collisions.)

    The underlying issue is to wonder if the BHCDM black-hole population is indeed small, and also restricted to a halo that doesn’t meaningfully interact with the galaxy it surrounds.

    Or, conversely, if it did at one time mix, what happened to get to the present situation of no obvious local small black holes?

  17. I do not agree with the assertion that it’s hard to make primordial black holes of that mass range without exotic physics. Here are some mainstream descriptions of hybrid inflation: http://arxiv.org/abs/1411.6616 , http://arxiv.org/abs/1505.04926 , http://arxiv.org/abs/1412.7619 , and http://arxiv.org/abs/1406.3342 .

    And http://arxiv.org/abs/1503.02317 is some recent string theory on black hole dark matter, describing post-inflation phase transitions which would not be visible in the CMB, and so the black holes involved might not be strictly primordial. With lithium abundances so poorly predicted by Big Bang nucleosynthesis, I doubt this is out of the black hole dark matter mainstream, either.

  18. Looking forward to LIGO and time to see if this theory holds water. Any theoretical ideas on how long we may have to wait before a sufficient amount of mergers could be compared to the map of dark matter and the occurrence of (or lack of) electromagnetic counterpart events?

  19. Black holes being the source of dark matter in the universe is a most intriguing hypothesis. The September 2015 LIGO direct detection of gravity waves is exciting and may be the foundation to learn more about black holes and possibly unravel the mystery of dark matter.

  20. BobC, the self-interaction cross section derived from the bullet cluster and similar observations is within the realm of possibilities for MACHOs because they are so compact, and so most of the time when they get close, they don’t actually merge, two of them go into a mutual orbit until a third comes along and then they all fly apart: http://arxiv.org/pdf/astro-ph/0501345.pdf

    Chuck S., your namesake Dr. Hailey at Columbia might be faster if his NuSTAR Milky Way survey gets out of peer review: http://i.imgur.com/JMQ3q9X.jpg

    for observations we used to hope for gravitational lensing http://arxiv.org/abs/1301.5067
    but then Wang and Loeb discovered there were lots of ways to look for them: http://arxiv.org/abs/1402.5975 ; eratta: http://mnras.oxfordjournals.org/content/445/2/1507.extract
    and then earlier this year the Japanese found one from the redshift of a gas cloud it consumed:
    http://www.nao.ac.jp/en/news/science/2016/20160115-nro.html

    Note that the reason that’s the first IMBH in the Milky Way is that most of the time, the big ones aren’t consuming anything, and they are likely to be more useful than harmful. Seriously, someone needs to tell EZ-Naut Corporation what gravitational slingshots can do for their fuel bills and travel times: http://jensorensen.com/2014/03/17/corporate-cosmos/

    Anyway, in all seriousness, you can also get early black holes from population III stars, but apparently not fast enough unless you invoke Bose-Einstein condensation of early particulate dark matter in addition to acretion, which replaces the particles: http://iopscience.iop.org/article/10.1088/2041-8205/720/1/L67
    The good news is that also explains dark energy according to http://arxiv.org/abs/1512.08623
    Does anyone understand that?

  21. If it quacks like a duck…
    I’ve been wondering for a while why, given dark matter is dark and black holes (absent accreting material) are also dark, why nobody (afaik) seemed to think dark matter might be black holes.
    I just assumed I was missing something simple – which is easy for a total non-physicist.
    So, the idea that there may be lots of relatively smallish PBH objects in intergalactic (“empty”) space seems fine, but should that not mean LIGO would be practically humming with similar waves arriving from every direction, most of them from high energy events occurring just after the big bang?
    If DM is BHs, an elegantly Occam like solution to the DM “mystery”, would this be seen as rather a let-down by cosmologists? Slightly ironic if so. When I were a lad, BHs were radical, cutting edge SF.

  22. @Soapy It’s still early days in GR wave detections. Many possibilities remain. There may be many medium-sized PBHs around a galaxy, or maybe the supermassive BHs somehow swept up all that were too close to get away. But even if there are relatively many PBHs around, that doesn’t mean they’re merging very often. Two PBHs could orbit each other for billions of years just like planets orbit the sun. In fact most stars aren’t solo like our sun, but are paired into binaries. There isn’t any magical force attracting PBHs to each other, apart from gravity. Orbits decay eventually, but not necessarily soon. Average time to merger of a PBH binary might be many times the age of the universe. As for any events soon after BB, they’re too far for LIGO. The 1 LIGO event so far was toward the high end of both energy and distance away. IOW, unless we’re blessed with unexpectedly energetic events, LIGO can only see up to a couple billion years back. Remember, we can’t see any events that have gone past us already, nor any that won’t reach us in our lifetimes. That’s life, in astronomy. 🙂

  23. @Joel said, “So most of the mass that came out of the Big Bang was in black holes, of a wide variety of sizes. The biggest ones attracted lots of other mass around them and eventually developed into galaxies. Smaller ones drifted off on their own and are still out there: dark matter… Why didn’t anyone think of this before?”

    Good point. I’ve been wanting to repeat your statement of the relationship between PBHs & DM since your clarification beat mine to the punch. Anyway, people had their reasons, but yes, one of them was the “streetlight effect”. When the question is “How might we detect this unseen mass we haven’t been able to detect so far?”, the answer isn’t going to be LIGO if you don’t happen to have a LIGO detector handy.

  24. Who allowed dark matter to become an official phrase used in scientific context anyways?

  25. Dark matter is not a clump of stuff that travels with the matter. Dark matter fills the space unoccupied by particles of matter and is displaced by the particles of matter which exist in it and move through it. Dark matter strongly interacts with matter. Dark matter is displaced by matter.

    [0903.3802] The Milky Way’s dark matter halo appears to be lopsided

    “the emerging picture of the dark matter halo of the Milky Way is dominantly lopsided in nature.”

    The Milky Way’s halo is not a clump of dark matter traveling along with the Milky Way. The Milky Way’s halo is lopsided due to the matter in the Milky Way moving through and displacing the dark matter, analogous to a submarine moving through and displacing the water.

    What physicists mistake for the density of the dark matter is actually the state of displacement of the dark matter. Physicists think they are determining the density of the dark matter by how much it and the matter curve spacetime. What they fail to realize is the state of displacement of the dark matter is curved spacetime.

Comments are closed.

Scroll to Top