Skip to main content Skip to main content

Physics for the 21st Century

Dark Matter Interview with Featured Scientist Doug Finkbeiner

Interviewer: What is an unanswered in physics and how does your research address that question?

DOUG: I think one of the biggest questions right now is: what is dark matter? And we’ve been trying to address that in a number of different ways by looking at the microwave signals you could get as dark matter annihilates; looking for gamma ray signals; cosmic rays, all sorts of things.

Interviewer: What is the motivation for your research?

DOUG: It bothered me that five-sixths of the matter in the universe is something we don’t understand, and we should really figure out what it is because it might be important. You lose sleep over this stuff.

Interviewer: How well do we understand dark matter?

DOUG: I think we understand how it gravitates very well because we see the galaxy rotation curves. We see the lensing around clusters of galaxies. We measure things in the cosmic microwave background (CMB). All of these have to do with how it gravitates. And we get consistent answers from all of them. So, it seems reasonable that it’s just some new particle we haven’t discovered yet that interacts only very weakly, if at all, with ordinary matter, but it does interact gravitationally.

Interviewer: Can you give a brief summary of dark matter research?

DOUG: In the 1930s, Fritz Zwicky was observing the radial velocity distribution of galaxies moving in the Coma clusters, and he noticed that the amount of mass required to make the galaxies move that fast was much larger than the amount of mass you get by adding up the visible starlight and other things you know about in the galaxies. So, he claimed that there might be some extra source of gravity there.

Vera Rubin was measuring rotation curves of galaxies, looking at how individual galaxies rotate so that you can measure the radial velocity of the stars as a function of how far they are from the center of the galaxy. And what you see is that the galaxies are rotating faster than they should, based on the amount of mass that you add up or get from the stars and gas.

Interviewer: What is the leading candidate for dark matter?

DOUG: There are two main candidates. One is called the WIMP—the weekly interacting massive particle, and that’s attractive because it has about the right mass to have something to do with what we call the weak nuclear force, which kind of kicks in on scales around a hundred GeV. There’s another particle called the axion, which also deserves attention.

Interviewer: Are massive compact halo objects (MACHOS) a candidate for dark matter?

DOUG: Before the CMB it was still possible that the dark matter was ordinary baryonic matter, like protons and neutrons and electrons, in a dark form. A MACHO is a massive compact halo object that could be something like a really faint brown dwarf or a Jupiter or some kind of massive body that escaped its solar system, and it’s just floating around freely in space. And you can hide a fair amount of mass in that, but you can’t hide nearly enough mass that way. The MACHO project did what’s called gravitational microlensing to look for these small compact objects. And they did find some, but also found that they account for a small fraction of the missing mass. Most of the matter has to be something else.

Interviewer: How did your research on dark matter first begin?

DOUG: I really wasn’t looking for dark matter at all. I was just trying to understand the microwave foregrounds from the cosmic microwave background. We figured we understood those things pretty well, and we could take a linear combination of them and make a map that looked like the WMAPS. And so, this was really a sanity check. And then there was this excess in the center that wouldn’t go away.

Interviewer: What is WMAP?

DOUG: WMAP is a satellite called the Wilkinson Microwave Anisotropy Probe designed to look at the CMB anisotropy, and try to derive cosmological parameters from it because the CMB is not perfectly smooth. It has little ripples in it. And those little ripples are the echoes of vibrations in the gas in the early universe. And by looking at exactly how much power they have on different size scales, we can work out a lot of things about the early universe.

Interviewer: What is the cosmic microwave background radiation (CMBR)?

DOUG: The cosmic microwave background radiation (CMBR) is the leftover radiation from the early universe. So, you start with the Big Bang, and everything is very hot, and all kinds of particles are being created and annihilated all the time. And then as things cool down, one particle species at a time freezes out, and you’re left with just protons, neutrons, and electrons. And then the universe is still opaque. It’s very dense. It’s still ionized. So, light can’t get very far in ionized gas. It just keeps scattering off the electrons. And then at some point, three or four hundred thousand years after the beginning, the universe cools down enough that the gas becomes neutral. Now all of a sudden these photons that have been scattering around all the time, they’re suddenly free. And they just go through the neutral gas for billions of years until we see them today.

Interviewer: What was the purpose of your research with WMAP?

DOUG: Well, to look at these little ripples in the CMB that the WMAP is trying to observe, you have to first get rid of the foreground emission from our galaxy. We live in a galaxy that’s emitting a lot of microwaves through all these different mechanisms. And those have to be subtracted from the maps. I was interested in working on trying to understand the microwave emission from our galaxy and then also working tracking those foregrounds so that we can do cosmology better.

Interviewer: Where you surprised by any of the results of your research?

DOUG: When we first tried to subtract off all the foregrounds that we know about there was always this excess or ‘haze’ left over in the center. And that was surprising to me.

Interviewer: Did you immediately think the haze was an indirect detection of dark matter?

DOUG: No, I certainly didn’t. It had crossed my mind just briefly once. I think one of my friends said in passing ‘it could be dark matter annihilation.’ And I said ‘wouldn’t that be funny.’ And then I dropped it. So, when I published the paper on the haze, I didn’t anything about dark matter. I just left it as a mystery for somebody else to look up. I didn’t have any good explanation about what it could be. Sometimes in science when you find something you don’t expect, people are tempted to just sweep it under the rug. But I think it’s very important to publish the things you don’t understand because then maybe someone else can get an idea on what’s going on.

Interviewer: How did you begin to figure out what the haze is?

DOUG: I think of it kind of like a crossword puzzle. You start with what you know, and you work out from there. But when you see clues you don’t understand, you should follow them and see where they lead. Whenever you sit down to try to explain something, and you think you already understand everything, you’re a little disappointed if it doesn’t work out. So, in that sense, I was a little disappointed at first that the haze was there, and I couldn’t find any way to modify the foregrounds we knew about to explain it. But then, of course, disappointment becomes excitement when you realize there might be something new and interesting that you don’t understand, and you’re about to learn something.

Interviewer: What other explanations are there for the haze, besides dark matter?

DOUG: When people look at a map of the radial and microwave sky, and they see microwaves, they see the synchrotron emission. The first thought is Supernovae because we know that Supernovae can accelerate electrons to high energies that are needed to make the synchrotron emission. But the electron spectrum that you get from a Supernova shock is thought to be too soft, so not enough particles at high energy to explain the haze.

Interviewer: What is synchrotron radiation?

DOUG: Synchrotron radiation should be thought of as a particle spinning around in a magnetic field. So, any time a charge accelerates for any reason, it emits light of some kind. And these charges spinning around the magnetic field, they’re essentially vibrating as they go around in the loop, and those vibrating charges are making photons. To emit synchrotron radiation, you have to have a magnetic field. Now we know around the earth, we have a magnetic field. That’s why your compass works. Everywhere in the galaxy it’s also thought that we have a magnetic field.

Interviewer: When did you decide that the haze might be the result of dark matter?

DOUG: My friend Amber Miller, a professor in the physics department in Columbia, called me up and said ‘Doug, if dark matter were annihilating, what would it look like in the microwaves?’ And I really tried for twenty or thirty minutes to talk her out of this crazy idea because I didn’t know anything about dark matter annihilation. I thought it was really quite fanciful to try to look for that in the microwaves. And then finally, she persisted, and I said ‘well, you know, actually, if you really believe there are WIMPS, and they’re annihilating, and they’re making high energy particles, and those would make microwaves, and you would get a hard microwave spectrum, then maybe I’ve already discovered it’. So, after that, is when I decided to sit down and actually do the calculating and see if it was even in the right ballpark.

Interviewer: What is particle annihilation?

DOUG: Well, every kind of particle we know about either has an anti particle: the electron has the positron; the proton has the anti proton, or it is its own anti particle, which sounds really strange, but the photon, for example, is its own anti particle. When you have a particle and its anti particle, and annihilate, they convert into energy according to the formula E=MC2.

Interviewer: Why did you think the haze could be dark matter particles colliding with each other?

DOUG: One of the funny things about this hazy signal is that it has a hard spectrum. That means there’s more of it at higher energies, higher frequencies than you might expect. So, because this looks harder than the ordinary synchrotron admission from super nova, we were looking for another source of these electrons. One possible source is from dark matter annihilation. Since there is more dark matter in the center of the galaxy and a dark matter collision requires two particles, it goes as the density squared. So, you would expect a lot more dark matter annihilation in the center than you do out – far away from the center. That made more sense to me than another explanation like there’s just more star formation.

The question is just is the dark matter annihilation signal bright enough to see with WMAP or not? So, dark matter particles could annihilate to pions and muons and eventually to electrons and positron. And then, those electrons and positrons are moving very fast. And so, they spiral around in the magnetic field and make this synchrotron emission.

Interviewer: What factors determine if the haze is the result of dark matter annihilation?

DOUG: There are two main ingredients. There are astrophysical ingredients and particle physics ingredients. So, on the particle physics side, you need to know something about the annihilation cross-section and the particle mass. The annihilation cross section comes from what’s called the thermal and relic freeze-out argument because in the early universe, you have all kinds of particles being created and annihilated all the time. And then as the universe cools down, you can still annihilate them, but not create them anymore. And how many you have left over after what’s called freeze out is roughly what the WIMP annihilation cross section is. So, that’s one important ingredient.

The mass scale: it’s probably bigger than a hundred GEV—or we would have discovered it at particle colliders already. And it’s probably less than about ten TEV or ten thousand GEV or else we run into some theoretical problems. On the astrophysics side, we need to know things about the dark matter distribution in our galaxy, which we can get from computer simulations. We need to know things about the magnetic field and the starlight energy density, and we get those from good old-fashioned astronomy observations. So, by putting all these different ingredients together, we can write down an equation for how much annihilation power there should be from dark matter annihilation. So, in other words, how much power is available to make synchrotron radiation that could be observed by WMAP.

Interviewer: What is “freeze out” ?

DOUG: There’s a very important generic argument about thermal relics of the Big Bang. By thermal relic, I just mean a particle that’s left over from the Big Bang. In the early universe, all kinds of particles are being created and destroyed all the time, presumably including our WIMP. And at some point, the universe cools off enough that the number of WIMPS starts to fall because there’s no longer enough energy to create WIMPS, and at that point, the density drops exponentially until something called freeze out occurs. Freeze out occurs when the density is so low that in a Hubble time, the characteristic time scale for expansion of the universe, the two particles cannot find each other to annihilate. And then, they don’t find each other again on average for the rest of the history of the universe. So, the interaction probability for annihilation, the annihilation cross section, is intimately related to how much dark matter we have left over today. And because we know how much dark matter there is, we know roughly what that annihilation cross section has to be.

Interviewer: What is the galactic density profile of dark matter?

DOUG: As galaxies form, they are piles of dark matter and baryons, by baryons, I just mean ordinary matter: protons, neutrons, electrons, hydrogen gas. And these things are collapsing gravitationally, and the ordinary matter can cool down—hydrogen atoms can collide with other hydrogen atoms and get in excited states and then emit light and fall back down. Ordinary matter as it cools, falls in further and further into the center of the gravitational potential well. Dark matter, we think doesn’t cool so effectively. So, what you end up with is this big puffy halo of dark matter with a much smaller pile of ordinary matter near the center of it forming stars, and planets. That’s why the visible part of the galaxy occupies a pretty small fraction of the overall dark matter halo. Nevertheless, the place where you’ll find the highest dark matter density is still in the center of the galaxy because the density goes up rather sharply as you go into the center. So, you expect a much higher density of dark matter toward the center of the galaxy than in the outer parts.

Interviewer: What is the annihilation rate of dark matter in the galactic center?

DOUG: Even in our galaxy, we’re only annihilating about a part in a trillion of the dark matter per Hubble time. So, Hubble time right now is about thirteen billion years. It’s not annihilating very fast. And if we go outside of our galaxy, out into the boondocks, the annihilation time scale is a million times slower.

But even a part in a trillion per Hubble time still produces a signal. There’s so much dark matter. We don’t have to annihilate very much of it to get an observable signal. The total amount of power in the haze is similar to the amount of power in Supernovae popping off in the center of the galaxy, for example. So, it’s not a tiny signal—but it’s not a huge signal either, which is why one worries about it being explained by other things. But yes, this is a very important point to make, when we talk about dark matter annihilation, people might have the idea that we’re burning through all the dark matter, in a million years, it’s all going to be gone. But that’s not true at all. We’re using such a tiny fraction of the dark matter that we’ll never notice the dark matter disappearing from galaxy rotation curves. Even if you waited trillions and trillions of years, you would never notice.

Interviewer: Is the haze the result of dark matter annihilation?

DOUG: Do I actually think it’s dark matter? Well, just between you and me, what we have here is the signal that is consistent with some fairly simple ordinary models of dark matter annihilation. But, this doesn’t have to be dark matter annihilation. It could certainly still be a signal from some other new astrophysical mechanism we don’t understand. Or, it could just be some strange detail about how the electrons propagate around the galaxy after they’re created.

Interviewer: What are your next steps?

DOUG: We need to do more research to get to the bottom of this. There is a new satellite launching this month called Planck that will map the whole sky. And, we’ll get a much better picture of this.

Interviewer: What will the data from Planck satellite tell you?

DOUG: Planck is measuring the microwave background just like WMAP, but is also going on up to higher frequencies—up about a factor of ten higher than WMAP. And that broad range of frequency coverage allows Planck to separate the various microwave foregrounds better.

Interviewer: What will the data from Fermi satellite tell you?

DOUG: Fermi has a gamma ray telescope, and we’re very interested to see what the gamma ray sky will look like. We are predicting, of course, that there will be a lot of inverse Compton scattered gamma rays in the inner few degrees of the galaxy where the haze is bright. And those gamma rays have to be there, or the haze doesn’t come from microwaves.

Interviewer: What do gamma rays tell you about the origin of the haze that microwaves do not?

DOUG: Gamma rays are very helpful. The very same high-energy particles that are making the synchrotron emission are also scattering starlight photons. That’s called inverse Compton scattering: when a fast electron hits a photon and kicks it up to high energy. So, we should see gamma rays from inverse Compton scattering. And if we don’t see those gamma rays at the appropriate level in the center of the galaxy, then that means the haze is not from synchrotron, and we’re back to square one.

Interviewer: What does the Fermi satellite measure?

DOUG: Fermi is surveying the entire sky every two orbits. It is what’s called a pair-conversion telescope. So, a gamma ray photon comes into the telescope, which is a layer of tungsten plates and then a calorimeter at the back. And in one of these tungsten plates, the photon will actually hit something and make a positron electron pair. And then that positron electron pair goes through the telescope and gets measured so you can work out its trajectory and then hits the calorimeter, which measures the total amount of energy in it. And so, by getting this direction and amount of energy, you can figure out their rival direction and energy of the original photon.

Interviewer: What are gamma rays?

DOUG: Gamma rays are just high energy photons. So, they’re just like the light in this room or radio waves or anything, just at an extremely high energy. The light in this room has an energy close to one electron volt; whereas, the gamma rays we’re talking about are about ten billion or a hundred billion times higher energy.

Interviewer: Can you compare gamma rays and microwaves?

DOUG: So, the synchrotron comes from the high-energy electrons spiraling around in the magnetic field and producing synchrotron radiation. And the gamma rays are coming from the high-energy particles hitting starlight photons and scattering them. And so, in as much as we understand the magnetic field well or the starlight well, we can interpret one or both of those. But we don’t understand either perfectly. So, we want to look at the haze from multiple directions and make sure we get agreement.

Interviewer: If your models for dark matter annihilation are confirmed, is there a need for more research?

DOUG: We’re a long way from being sure. Something as slippery as dark matter—you really have to detect in multiple ways. So, that’s why I always like to say there are kind of three legs on the stool. One is direct detection. Another is learning more physics from the LHC and other particle colliders. And then the third is astrophysical—making sure that whatever new particles we discover elsewhere actually are the dark matter.

So, approaching this problem of dark matter from the astrophysics first, is kind of putting the cart ahead of the horse because what you’d really like to do is have the particle colliders teaching you new physics and having direct detection experiments like the Xenon experiments, finding actual scatterings of WIMPS. And then you take the information you learned from that and go to the sky. I’m starting with astrophysics because I’m an astrophysicist, and that’s what I know about. But these other things like LHC and Xenon experiments actually come first in the logical progression.

Interviewer: Is the haze proof of dark matter?

DOUG: The haze is not proof of anything. The correct statement is that a plain vanilla WIMP with very simple assumptions for the mass and the cross section and then simple assumptions about the dark matter distribution in our galaxy and the magnetic field—if you put all of these things together, you get something that looks like the haze for s very reasonable WIMP parameters.

Series Directory

Physics for the 21st Century


Harvard Smithsonian Center for Astrophysics in association with the Harvard University Center for the Environment. 2007.
  • Closed Captioning
  • ISBN: 1-57680-883-1