Magnetic trap snares methyl radicals

By and large, physicists have succeeded in their quest to tame the atom. These days, atoms can be laser cooled to their ground states, stored in traps for minutes, and switched between internal states virtually at will. (See the article by Ignacio Cirac and Peter Zoller in Physics Today, March 2004, page 38 .)

Molecules, however, are wilder beasts. They are all but impervious to laser cooling, which demands a closed optical loop—that is, a sequence of photoexcitation and decay that can be repeated ad infinitum. Due to the additional degrees of freedom afforded by rotational and vibrational modes, molecules tend to decay unpredictably, often to states that can’t be optically addressed. Inevitably, the loop breaks.

Over the years, experimenters have devised strategies to overcome the optical-loop problem: creating cold molecules in situ from cold, trapped clouds of reactive atoms (see the article by Debbie Jin and Jun Ye, Physics Today, May 2011, page 27 ); cooling molecules “sympathetically” by letting them thermalize with cold atoms; closing optical loops by using RF fields to periodically reset molecules’ internal states (see Physics Today, January 2010, page 9 ). But those methods generally either work only in limited cases or yield gases that are too dilute for investigations of cold-molecule collisions, Bose–Einstein condensation, and other quantum phenomena of interest.

A fourth way to cool molecules into the quantum realm is simply to let them escape from a pressurized container into a vacuum. If the initial pressure is suitably high and the escape orifice suitably small, the temperature of the exiting molecules will fall to well below 1 K, cold enough that they behave more like waves than particles. For the experimenter set on interrogating them, however, there’s a complication: The molecules will shoot from the orifice at roughly the speed of a rifle bullet.

In 2000 Gerard Meijer and his colleagues at the University of Nijmegen in the Netherlands showed that such beams could be slowed to a standstill using pulsed electric fields, provided the molecules had sufficiently strong electric dipoles. ¹ Now researchers led by Takamasa Momose (University of British Columbia, Vancouver, Canada) and David Carty (Durham University, UK) have pulled off an analogous feat on a molecule that has no electric dipole at all: They used pulsed magnetic fields to decelerate and trap a beam of methyl radicals cooled to their rotational ground state. ² The new trapping technique can be applied not only to CH₃ but to any molecule with a magnetic moment—a class that includes essentially the entire family of reactive intermediates known as radicals.

Zeeman deceleration

The concept behind the new decelerator and trap is nearly a decade old, developed by Frédéric Merkt and coworkers at ETH Zürich as a way to corral beams of paramagnetic atoms. When such beams are directed through the magnetic field of a solenoid coil, about half the atoms have their unpaired electron spin aligned antiparallel to the field. Those atoms are weakly repelled by the field due to the Zeeman effect, whereby the energy of an antiparallel state grows in proportion to an external field.

But that repulsion alone doesn’t suffice to slow an atomic beam. A fast-moving atom’s encounter with a localized magnetic field is like a fast-rolling ball’s encounter with a mound: The atom expends kinetic energy climbing the magnetic potential but regains it during the ensuing descent. The trick with Zeeman deceleration is to switch the coil off just as the atoms arrive at the field’s peak, so that the expended kinetic energy is permanently lost. By repeating that process with a succession of a dozen coils, each delivering 1 T pulses, Merkt and his coworkers could stop atoms entirely.

For nearly a decade now, Merkt’s group has been using the approach to trap atomic hydrogen and deuterium. But stopping the heftier CH₃ radicals called for considerably greater braking force. Momose and his colleagues needed coils that could deliver pulses exceeding 4 T, on par with the strongest magnets in laboratory use. And they needed 85 of them.

The team’s instrument, a meter-long cylinder lined with 4-mm-diameter solenoid coils, is partially illustrated in the figure on page 20. (The design is a modified version of an atom decelerator built by a University of Texas at Austin group led by Mark Raizen. ³ ) At the outlet is a pair of opposing permanent magnets that serve as the trap. Near the inlet, a nozzle spouts CH₃ radicals in cold, bunched beams. The appropriate timing for each pulse could be calculated based on the gas’s initial velocity, around 320 m/s. But coordinating the coils to fire with the requisite precision took considerable technical know-how. “We have to send 700 amps to each of the 85 coils for just a few microseconds at a time,” Momose explains. “And we have to do it inside a vacuum. There are always dielectric breakdowns.”

In all, it took six years to get the instrument working—three to decelerate molecules and another three to stop them. In a typical run the team captures some 50 000 molecules in the 1 mm³ magnetic trap, where they can be held for about a second. The trapped gas is sufficiently dense to allow precise measurements of cross sections for collisions between CH₃ and assorted background gases; those measurements are already under way.

“It would have been extremely difficult to trap methyl radicals using any other method,” comments Edvardas Narevicius, whose group at the Weizmann Institute of Science has been developing a magnetic decelerator to simultaneously trap lithium and molecular oxygen. ⁴ “This is really a big step forward expanding the number of species that we can address.”

Interstellar chemistry

On occasion, Momose cadges time at the Nobeyama Radio Observatory’s 45 m telescope in Nagano, Japan. There he combs the space between stars for spectral lines produced by small hydrocarbon molecules, which are puzzlingly abundant in the interstellar medium. A possible explanation is that the rates of hydrocarbon-forming reactions are boosted by quantum tunneling through activation-energy barriers.

That’s one reason Momose is especially excited about the newfound ability to isolate cold CH₃. He previously worked with researchers at Kyoto University in Japan to detect tunneling contributions to the methane-forming reaction CH₃ + H₂ → CH₄ + H in cryogenic hydrogen crystals. Now that CH₃ can be more comprehensively isolated from environmental influences, he hopes to measure those tunneling rates with far greater precision.

The ability to trap CH₃ also presents opportunities for fundamental physics. With the molecule in its rotational ground state, the researchers can make precise measurements of hyperfine transitions and parity-violating interactions. (See the article by David DeMille, Physics Today, December 2015, page 34 .) Ultimately, however, they hope to create a molecular gas that’s cold enough and dense enough to form a Bose–Einstein condensate.

Momose thinks they should be able to cool their gas to submillikelvin temperatures via sympathetic cooling, “and then evaporative cooling should get us much lower, down to 1 microkelvin. Then the only missing part would be the density.”

A BEC requires a phase space density of order 1, which would translate to a volumetric density about three orders of magnitude higher than the 5 × 10⁷ cm⁻³ that Momose and company have achieved so far. “We’d probably need to build another decelerator,” he muses. “So that would mean another three years.”