It is the stuff that both science fiction and a good part of author Dan Brown's fortune are made of—antimatter.

A research group at CERN, the European lab for particle physics in Geneva, has managed for the first time to confine atoms of the stuff. Fleeting antimatter atoms have been produced in the lab for years, but until now the ability to trap the elusive atoms for detailed study has been out of reach. (The confined amounts of antimatter are many orders of magnitude smaller than that swiped from CERN by insidious plotters in Brown's Angels & Demons.)

The new advance, published online November 17 in Nature by the ALPHA Collaboration experiment at CERN, is only a proof of principle—the anti-atoms have only been confined for less than two tenths of a second—but the research could set the stage for a new round of fundamental physics tests. (Scientific American is part of Nature Publishing Group.)

The ALPHA group mixed antiparticles in a vacuum trap to create atoms of antihydrogen, then held on to them briefly in the trap before turning them loose. Antimatter annihilates on contact with ordinary matter, so the anti-atoms disappear in a shower of secondary particles, known as pions, when they hit the walls of the trap. By tracking those annihilation products, the physicists conclude that they succeeded in producing, trapping and then releasing a few dozen atoms of antihydrogen.

Neutral hydrogen is made up of one proton and one electron; antihydrogen is composed of the corresponding antiparticles, the antiproton and the antielectron. The component antiparticles that make up an anti-atom are not on their own terribly exotic. Antielectrons, also known as positrons, are in wide use in PET (positron emission tomography) scanners. And antiprotons have been produced and accelerated to high energies for smashups in particle colliders for decades.

But marrying an antielectron to an antiproton to form a bound antimatter counterpart to the hydrogen atom was not achieved until the mid-1990s. And those early anti-atoms, produced at CERN and at Fermi National Accelerator Laboratory in the U.S., were "hot," zipping along near the speed of light. The difficulty of corralling anti-atoms with such potent kinetic energies led groups to pursue "cold" antimatter that could more easily be confined and studied.

"I was never too worried about producing antihydrogen, but holding on to it is another thing entirely," says ALPHA spokesperson Jeffrey Hangst, a physicist at the University of Aarhus in Denmark. "We're kind of overjoyed, to put it mildly, that this is working so well."

The challenge in confining antihydrogen, besides the fact that it annihilates on contact, is that it is electrically neutral, so the same traps that can be used to steer and confine the charged antiparticles are useless once those antiparticles bind together into an atom. On the bright side, the physicists can sweep the trap with applied electric fields after mixing the antiparticles to clear out any antiprotons and antielectrons that have not been bound into atoms of antihydrogen. "It's neutral, and so it's very difficult to influence in any way, but it still has a magnetic moment," Hangst says. "You can think of it as a little compass needle that responds to external magnetic fields."

With superconducting magnets, Hangst's group was able to manipulate the neutral anti-atoms, trapping them—however briefly—before switching off the magnets to let the antimatter wander off and reveal itself through annihilation. Key to detecting those confined anti-atoms was the development of superconducting magnets that can be shut off almost instantaneously, allowing the researchers to look for pions from matter–antimatter annihilations in a span of just 30 milliseconds. The detectors are struck regularly by cosmic rays, which can mimic the annihilation signal, so narrowing the window of time in which matter–antimatter annihilations should be taking place significantly reduces the background noise the physicists must sift through to identify genuine annihilation events.

The 38 annihilation signals detected by ALPHA were well above the expected background of 1.4 occurrences, strongly indicating that antihydrogen atoms were indeed slamming into the trap walls after being released from confinement. "I'm convinced that they have succeeded in trapping some antihydrogen atoms," says Fermilab physicist David Christian. "That's a big milestone in their experimental program."

If the proof of principle leads to more robust trapping of anti-atoms, researchers could test a number of long-standing theories for how antimatter should behave. For instance, all indications are that gravity should act on antihydrogen just as it acts on hydrogen, but empirical tests are not yet feasible. "There are lots of arguments why it should behave exactly as matter, but they are just arguments," says theoretical physicist Michael Nieto of Los Alamos National Laboratory.

Physicists would also like to study anti-atoms with laser spectroscopy to probe their energy level structure; according to fundamental physics theories antihydrogen should have the same spectrum as ordinary hydrogen. Any detected deviation "would be a huge thing," Christian says. "Whether or not they can do the spectroscopy is still a few steps off, but they've come a long way." As a ballpark figure, Hangst says researchers might need to confine 100 anti-atoms on a timescale of seconds to probe their structure. He notes that ongoing work to improve the trapped lifetime of the anti-atoms is coming along very well.

Meanwhile, a competing group at CERN, known as ATRAP, has been proceeding apace with their own cold antihydrogen program. Harvard University physicist Gerald Gabrielse, the ATRAP spokesperson, says he is "delighted" by the new announcement but that his group is taking a slightly different tack. "We have been focusing almost entirely of late on producing much colder antiproton plasmas that contain many more particles," Gabrielse says. "The hope is that with these we can make many more antihydrogen atoms that are cold enough to be trapped for the longer times needed for laser spectroscopy."