Science 1663

The Hunt for Dark Matter

Cosmologists believe that most of the matter in the universe is dark and hardly interacts with the ordinary matter found in planets, stars and galaxies.

1663:What is dark matter and why is it interesting?

Habib: When we look up at the night sky, we see bright matter—stars and galaxies, comets and asteroids, all made of the same chemical elements we have on Earth. They radiate and absorb light over a range of wavelengths, and that's how we know they're there.

Andrew Hime

Hime: Dark matter is very different. It neither absorbs nor emits light, so there's no direct way of knowing it's there. But in the 1930s a fellow named Zwicky was studying the motions of galaxies in the Coma cluster, a giant galaxy cluster close to our galaxy, and he noticed that the bright matter in the cluster didn't have enough mass to keep the fastest-moving galaxies from flying off into space.

Habib: Some missing mass, invisible to us, had to be holding the galaxies together through its gravitational pull.

Hime: Zwicky dubbed this mysterious matter "dark"; matter. Evidence for it has accumulated over decades, but in the last 5 to 10 years, precision measurements of the cosmic microwave background (CMB), the primordial radiation left over from the Big Bang, clinched both the existence of dark matter and its great abundance.

1663: The Big Bang is the explosion that started the expansion of the universe?

Habib: That's right. The CMB is important. It fills all space and contains an imprint of the hot gaseous matter that existed about 100,000 years after the Big Bang. Although that matter appears to be nearly homogenous, very small temperature fluctuations in the CMB reflect its tendency to condense into separate clumps under the attractive force of gravity. Cosmologists have calculated that ordinary matter alone could not have driven the gravitational instability needed to condense the nearly smooth matter distribution imprinted in the CMB into what we see some 13 billion years later—huge walls of galaxies and galaxy clusters surrounding giant regions devoid of bright galaxies. Instead, the total amount of matter in the universe must be much larger, with ordinary visible matter composing only 15 percent and dark matter, which we cannot see, composing the other 85 percent.

1663: What might dark matter be made of?

Hime: We don't really know, but there are compelling reasons to think it's made of subatomic particles yet to be discovered—particles that interact very weakly with ordinary matter.

1663: Could they be neutrinos?

Hime: We do live in a sea of primordial neutrinos, also leftovers from the Big Bang, and we discovered at SNO that neutrinos are particles with mass. However, they are too light and too energetic—they move on average at nearly the speed of light. We need a massive, slow-moving particle to form the cosmological dark matter.

Map of the cosmic microwave background (CMB) radiation, showing temperature fluctuations (yellow and blue) that vary by only a thousandth of a percent. These fluctuations represent an imprint of the variations in matter density in the early universe. Image courtesy of NASA/WMAP Science Team.

Habib: If the dark matter particles move too quickly, they would counteract the tendency for gas clouds to condense into galaxies. With slow dark matter particles, structure in the universe would form naturally in a hierarchical fashion, small structures forming first and then merging to create larger ones. And that is what we see in the universe. First, galaxies formed, then galaxy clusters, and finally a network of galactic superclusters encircling our cosmic neighborhood.

Hime: Remarkably, a new theory of elementary particle physics, called supersymmetry, predicts the existence of elementary particles with just the right properties to explain what dark matter is and where it came from. These new particles have no electric charge, but they do interact with ordinary matter through what's called the weak nuclear force, the force that causes nuclei to undergo radioactive decay. They are called Weakly Interacting Massive Particles, or WIMPs, and the strict symmetries in this new supersymmetric theory ensure that they remain stable and do not decay. They also have the right masses and the right level of interaction with ordinary matter to be yet another relic of the Big Bang.

Salman Habib

Habib: Andrew means that as the universe expanded after the Big Bang, it cooled to a temperature at which the WIMPs froze out of the cosmic soup, forming a background of matter with the right mass density (and hence gravitational influence) to explain the buildup of the structures we observe in the universe.

Hime: Just a back-of-the-envelope estimate based on the strength of the weak nuclear force and the expansion rate of the universe gives you the right mass density for a primordial background of WIMPs. Now we want to make a direct search for those particles to see if they're really out there.

1663: How can you detect a WIMP if it hardly interacts with ordinary matter?

Hime: Imagine our galaxy permeated by a cloud of dark matter WIMPs, and imagine the solar system, and thus Earth, moving through the cloud as it circles around the center of the galaxy. The WIMPs would be moving at several hundred kilometers per second relative to us. We want to record the small wake of energy left behind as a WIMP collides with and recoils from an atomic nucleus in our experimental apparatus. In the experiments we now have planned, a big spherical container filled with an ultra-pure, liquefied noble [inert] gas—either argon or neon—will be surrounded by light detectors. When a WIMP hits an argon or neon nucleus, that nucleus will recoil, causing the energy of its motion to be translated into a flash of light (scintillation light) as it slows to a stop. The WIMP has the same mass as 100 protons, and if you work out the math, you find that the nucleus it collides with will recoil with a very tiny energy, some thousand times smaller than the energies we detect when a neutrino from the sun collides with a heavy water molecule (D2O) in the huge detector at SNO.

This map shows the positions of over a million galaxies (dots) relative to us (the origin—center, bottom). As time progressed, galaxies coalesced into walls and clusters surrounding large voids—the lumpy large-scale structure of the universe in our cosmic neighborhood. image courtesy of: Sloan Digital Sky Survey (SDSS.org)

1663: Can you detect such a low-energy collision?

Hime: Yes, we can, and we have successfully demonstrated that in the laboratory. Even though a WIMP collision will produce only a few photons [light particles], all of them will reach the light detectors because these noble liquids are transparent to their own scintillation light. The wavelength emitted by the ionized diatomic molecules created by the WIMP collision is longer than single neutral atoms can absorb (see “WIMP Detection”) From the pattern of light hitting the detectors, we can reconstruct the position of all these events, and we consider only those that are distant from the container walls to eliminate events caused by radioactive impurities or other nuclear reactions in the walls.

Habib: The real question is, how will we know the photons we detect are from a WIMP and not some other particle going through the tank?

The proposed 100-ton CLEAN detector (for Cryogenic Low-Energy Astrophysics with Nobel gases) is designed to detect dark matter particles (WIMPs). Buried deep underground, it will consist of a giant bottle filled with liquid argon (or liquid neon) and lined with photomultiplier tubes that convert scintillation light to electrical signals. CLEAN is expected to detect upwards of 100 WIMPS per year.

Hime: That's the difficult challenge of this experiment, to reduce the background signals, and that's where the properties of argon and neon become so important. First, because these noble gases liquefy at very low temperatures, all the radioactive contaminants that would produce spurious signals condense out of the starting gas well before it liquefies. Second, because these liquids are transparent to their own scintillation light, we can make a very big spherical container and still detect the very few photons produced in a WIMP-nucleus collision. Third, these liquid noble gases scintillate very brightly in the extreme ultraviolet—enough to allow detection of even the faint signal induced by the recoiling WIMP of interest. Now, the pièce de résistance for both neon and argon is that the scintillation light has two different components: a fast component, “early light,” emitted in nanoseconds and a slower “late light” component emitted over several microseconds. This allows us to distinguish a nuclear recoil, which produces hardly any of this late light, from an electron or a gamma ray from radioactive decay, which produces a lot of late light. So if you don't see this late light, you know it's a WIMP, provided the detector is deep underground where there are no other neutral particles—no cosmic-ray neutrons, for example—to fake the nuclear recoil signal of a WIMP. This power to discriminate between background events and the nuclear recoil from a WIMP collision is so large that it increases our sensitivity a thousand times over present experiments.

WIMP Detection A WIMP (black) can pass through the earth and into the liquid argon-filled CLEAN detector, where it collides with an argon nucleus. In the liquid, the recoiling nucleus collides with and ionizes atoms, which pair up and form excited diatomic molecules. These decay and emit light either quickly (“early light”) or slowly (“late light“). In the WIMP event shown here, there’s hardly any late light. But in background events (electrons stemming from radioactive decay), most of the light is late light.

A WIMP (black) can pass through the earth and into the liquid argon-filled CLEAN detector, where it collides with an argon nucleus. In the liquid, the recoiling nucleus collides with and ionizes atoms, which pair up and form excited diatomic molecules. These decay and emit light either quickly (“early light”) or slowly (“late light“). In the WIMP event shown here, there’s hardly any late light. But in background events (electrons stemming from radioactive decay), most of the light is late light.

1663: How many WIMP events do you expect to see?

Hime: It could be as little as one per year per ton of noble liquid. We're working with argon first because its nucleus is twice the size of neon's and has four times the chance of colliding with a WIMP. The downside is that argon contains a radioactive isotope, but since the decays will produce late light, they will be distinguished as background. We've already tested and demonstrated the concepts in small prototype experiments of 1 to 10 kilograms. Obviously, we don't have a source of WIMPs, so we used neutrons. They produce nuclear recoils in the liquid just as WIMPs would do. Now we're building a 100-kilogram detector, which should be about 50 times more sensitive to WIMPs than the current state of the art. If it works, it's easy to take it up to the metric ton scale, which adds an additional factor of 10 in greater sensitivity. If we detect a WIMP, it will be very exciting. If we don't, it will be equally important because it will rule out WIMPs as dark matter candidates. In any case, these novel detector concepts will be paramount to resolving the dark matter problem.

Supersymmetry may not be exactly the right theory of elementary particles. But there's got to be a simple principle like it that solves fundamental issues on the microscopic scale and, for a completely unpredicted reason, solves this dark matter mystery for you. The possible existence of dark matter particles like WIMPS is so compelling that it just invites you to go out and directly look for them.