WIMPS ON THE LOOSE

Around the world scientists are setting traps to find out what makes up 90 per cent of the universe

If Eddie George announced that he could not account for 90 per cent of the gold held by the Bank of England, he would have some explaining to do. The world's astronomers have a roughly equivalent problem, only theirs is about mass, not money. For the past 20 years, astronomers have had compelling evidence that there is more to the universe than meets the eye. About 90 per cent of the calculated mass of the universe seems to be invisible.

This "dark matter" (apparent only through its gravitational attraction) could consist of dim stars, not bright enough to be seen from Earth. It could simply be the great profusion of infinitesimal elementary particles (neutrinos) that zip around the universe, rarely interacting with matter in any way at all. But there is a more exotic alternative, harder to prove but increasingly appealing: that the dark matter is made up of massive particles yet to be discovered and collectively known as WIMPs (weakly interacting massive particles).

Introduced by particle physicists, WIMPs became popular with cosmologists and astronomers in the mid-80s, and they are not just the stuff of debate. Astronomers are scouring the perimeter of our galaxy for unseen stars and scrutinising the neutrino to see whether it could make up the mass deficit. But as such searches look less and less likely to offer a full answer to the dark-matter puzzle, researchers are also setting traps for WIMPs, which might be drifting continuously through Earth - as a beam of light might shine through a pane of glass. This year, what was a trickle of WIMP-hunting experiments will swell to a stream as a new generation of high-technology WIMP detectors joins the effort.

WIMP-hunting devices are appearing all over the world. A US-based experiment - that monitors crystals at close to absolute zero temperature for the tiny pulse of heat that might signal a WIMP - was expanded recently, and this month a more sensitive competing experiment will debut at Italy's underground Gran Sasso laboratory. Some 20 other WIMP searches are on their way, aiming to snare these phantoms. The hunting technique varies but each basically employs very sensitive materials to register a WIMP's energy as it passes through Earth. Researchers are looking for the amount and regularity of energy a WIMP dumps into a detector.

Success is a long shot, the researchers concede that WIMPs are, as their name implies, loath to interact with ordinary matter, and their properties - and thus the best way to trap them - are largely unknown, all of which makes WIMP-hunting "a very difficult business", says Oxford University physicist Susan Cooper, a leader of the Gran Sasso project. "[A WIMP] doesn't have a clear signature," she laments. "It's not the kind of experiment that you really like to do, because it is so hard, but I feel we are forced to try because the question is so big."

Dark matter may be elusive, but the evidence that it exists is strong. It's about 20 years since astronomers first began measuring the speeds at which hydrogen clouds orbit the centres of spiral galaxies and realised that the speeds they were finding were too high for the gravitational pull of the visible stars in the galaxies to hold the clouds in their orbits. Astronomers concluded that a large but invisible source of gravitational pull in the outer reaches of these galaxies must also be at work.

Gravitational lensing provides even more persuasive evidence. Light reaching Earth from distant galaxies is sometimes distorted by the gravitational pull of a cluster of galaxies that lies in the path of the light. Analysis of the distorted image allows astronomers to "weigh" the cluster, and "that shows that there's lots and lots of dark matter in the cluster," says Oxford astronomer Will Sutherland.

At least some of that dark matter, says Suther-land, could consist of the so-called massive compact halo objects (MACHOs), a catch-all term embracing any kind of star, approximately the size of Jupiter and bigger, that does not emit light and therefore cannot be seen as well as any other massive and invisibly dull stars. But most researchers, even those hunting MACHOs, expect that these objects will only make a modest contribution to the total of dark matter. According to the type of universe cosmologists currently favour, theories of how matter formed after the big bang limit the amount of normal or "baryonic" matter - the protons and neutrons that make up both normal stars and MACHOs - to about 10 per cent of the mass required to gradually slow the universe's expansion. If the cosmologists are to be believed, this means that the rest of the matter must be something more exotic.

Neutrinos are a possibility for this non-baryonic matter. Some theorists speculate they are endowed with a minute mass, tens of thousands of times smaller than the mass of an electron. Yet, despite decades of trying, experimenters have yet to pin a mass on the neutrino. Cosmological models, based on neutrino dark matter, predict large groupings of galaxies not seen in the real universe, casting doubt on the fine details of neutrino theory. So many are turning to WIMPs, which may match particles predicted by some exotic theories in particle physics.

So far, WIMPs have fewer strikes against them than the other dark-matter candidates - except that they are entirely hypothetical. They also provide a unique challenge for experimentalists: building a detector for a particle that you do not know for certain exists, with unknown properties. To make matters worse, WIMP detectors risk being flooded with known particles, such as cosmic rays (they are often located underground to minimise this risk), and debris from radioactive decays of atoms around the experiment and in the apparatus itself. According to Cooper: "Any signal that is seen is liable to be a bit shaky, it's very important to have the possibility of confirming it with different techniques."

One technique has already been tested over the last several years in an earlier generation of WIMP searches. It relies on scintillation: the tiny pulse of light given off when an incoming particle strikes an atom in certain crystals, such as sodium iodide. The UK Dark Matter Collaboration (UKDMC) runs a scintillation detector in Europe's deepest mine, the Boulby salt and potash mine in northern England, where spurious signals from cosmic rays are at a minimum. The experiment, led by Peter Smith of the Rutherford Appleton Laboratory, consists of a 6kg crystal of sodium iodide watched by a pair of light-detecting photomultipliers (which detect and amplify light from the very faint source). Running since 1994, the detector has established upper limits on the frequency with which WIMPs of various masses knock into the detector.

Meanwhile, in the Gran Sasso laboratory beneath the Apennine mountains of central Italy, the Dama group, led by the University of Rome's Rita Bern-abei, has a similar 115.5kg sodium iodide detector. To improve their chances of detecting a WIMP, the team is planning to increase this to one ton of sodium iodide. And both Dama and the UK group are currently working on new detectors using liquid xenon which is expected to improve sensitivity.

Many of the new detectors, however, are based on a different strategy which should raise the odds of catching a WIMP: detecting an incoming WIMP not by scintillation but by the energy it deposits in the detector material. "The amount of energy, by room temperature thermal standards, is absolutely insignificant," explains Tom Shutt of the University of California, Berkeley. But if you cool your crystal to a temperature of 200th of a degree above absolute zero, a single particle depositing a few kiloelectron volts will cause a millionth-of-a-degree temperature rise. This measurement, which according to Shutt "turns out, is quite measurable", is about equivalent to detecting the amount of energy you would get if you burnt a drop of petrol so small it could not be seen with an optical microscope.

One set of these detectors, or bolometers, is already keeping watch for WIMPS. The Cryogenic Dark Matter Search (CDMS) experiment, headed by Bernard Sadoulet at Berkeley, has had two detectors stationed a few meters underground at Stan-ford University, California, since last September and recently added a third detector. Running at a temperature of a few tens of millikelvins, they consist of between 100g and 200g of semi-conductor crystals. An incoming WIMP knocks into a crystal nucleus, which recoils and creates heat.

To distinguish between WIMPs and cosmic rays, the CDMS detectors can also detect the charged particles - electrons or nuclei - dislodged in the crystal by a particle impact. "That's very important, because essentially all the radioactive background interacts with the electrons, and the signal that we are looking for interacts with nuclei," says Sadoulet. "In the end, we identify the amount of charge produced by each particle that strikes our detector, and thus distinguish WIMPs from background photons [light radiation]," adds Shutt. Soon the experiment will move to the Soudan mine in Minnesota, where it will be better shielded from cosmic rays. By that time a host of other cryogenic experiments should be underway. Makoto Minowa and his colleagues at the University of Tokyo are building a WIMP detector based on measuring the temperature rise in ultracold lithium fluoride. And the French Edelweiss (Experience pour Detecter les Wimps en Site Souterrain) collaboration, is building a germanium bolometer in the Frejus underground laboratory in the Alps, which, like the Berkeley device, will be sensitive to both the charge signal and the temperature rise from an impinging particle.

Lighter, slower WIMPs would deposit less heat in a bolometer, a possibility that has encouraged some groups to develop new, more sensitive thermometers. One is the closest competitor to the Berkeley experiments, the Cresst (Cryogenic Rare Event Search with Super conducting Thermometers) project run by Cooper in Oxford, and her collaborators from Munich. The detector, at Gran Sasso, should be able to sense just 500 electron volts, a tiny amount of energy, equivalent to just one single soft X-ray photon. "We're specifically most sensitive to low-mass WIMPs" says Cooper, "complementing many other experiments."

Physicists from Paris and Saragossa, Spain are building a smaller version of Cresst, dubbed Rose-bud (Rare Object Searches with Bolometers Under- ground), in the Canfranc underground laboratory in the Spanish Pyrenees, with data taking expected to commence next year. But there are other ways of exploiting superconductivity to catch a WIMP. The Swiss Orpheus team is building a detector containing billions of tin granules, each just 30 micrometers in diameter (about half the size of the smallest grain of sand). More than a kilogram of these grains are cooled to the superconducting transition temperature for tin and swathed in a magnetic field. An incoming WIMP will strike a nucleus in a granule and warm it up. "When the granule is heated up, it then makes a transition from the superconducting phase to the normal conducting phase ... [creating a magnetic signal which] can be detected with a magnetometer," says Orpheus leader Klaus Pretzl. The team expects to install the detector beneath the city of Bern at the end of the year.

One advantage of the magnetic "thermometer" strategy, says Tom Girard of the University of Lisbon, is its ability to reject spurious signals from natural radioactivity. A WIMP will strike only a single grain of tin, which radioactive background will cause transitions in a streak of grains, giving a larger blip in the magnetometer, he says. Girard and other researchers from Saragossa, Lisbon and Paris are mounting a magnetic experiment resembling Orpheus, called Salopard, which will be installed in the Canfranc laboratory in a year or so. "We estimate the ability to reject 97 per cent of the background contribution," says Girard.

Two other groups hope to eliminate spurious signals by exploiting difference phase transitions (eg water to steam, liquid to bubble) at higher temperatures. Both teams, one in Canada and one a Cern-Lisbon-Paris collaboration, use droplets of liquid Freon (a coolant), a few tens of micrometers in diameter, entombed in a clear gel. At room temperature, above Freon's boiling point, the droplets are confined in an unstable superheated state. If an impacting WIMP deposits enough energy in a Freon droplet, "it is vaporized and expands into a bubble of Freon gas of about 1mm in diameter, which is contained at its location in the gel," says Viktor Zacek of the University of Montreal, spokesperson for the Canadian group. The result, says Juan Collar of Cern, "is a characteristic audible sound emission that can be picked up when this happens."

Such detectors "are totally insensitive to low- energy photons, the main source of background in dark-matter searches," Collar adds. Most background radiation does not deposit enough energy in a sufficiently short distance for the superheated bubbles to notice. "So life is much easier for the WIMP hunter," says Collar. Last month, the Canadian team started running a prototype system based on just a few grams of Freon droplets. Larger systems are in the pipeline, to be installed in a mine in Ontario. The European team, says Collar, plans to install its prototype in a shallow tunnel near Paris this year.

In spite of the impressive technology being deployed, there remains the possibility that all these searches may draw a blank. WIMPs could simply be a fix conjured up by astronomers and cosmologists to get their theories to match what they see in the universe around them. But Sadoulet says the history of physics shows that what appears at the time to be a "fix" can later look like prescience. He points to the difficulties faced by Danish physicist Niels Bohr and his contemporaries in the early 1930s as they struggled to understand radioactive beta decay, in which some energy seemed to simply vanish. Bohr proposed dumping the principle of energy conservation, while Wolfgang Pauli proposed that the energy was fleeing in the form of a ghostly new particle, purely hypothetical at the time - the neutrino. The rest is history.

Originally published in 'Science' magazine