A New State of the Fifth State

If a group of JQI researchers is right, one of the strangest phenomena in nature has an even stranger side that could lead to previously unseen kinds of atomic behavior, and possibly to a means of robustly fault-tolerant quantum computation.

Their goal is a radical variation on a condition called Bose-Einstein condensation, often deemed the “fifth state” of matter after solids, liquids, gases and plasmas. Predicted in 1925, but not achieved in the laboratory until 1995, it occurs when a cluster of atoms is cooled nearly to absolute zero and all of the atoms finally condense into exactly the same quantum state. At that point, they are as indistinguishable as photons in a laser beam, and collectively behave like a single “super atom.”

Not all particles can do this. Only those with whole-number values of a quantum angularmomentum property called “spin” can condense. That group, called bosons, includes photons and atoms such as helium-4 (but not its rarer isotope, helium-3) whose individual components have spins that add up to an integer.

The other class of particles -- fermions, with halfinteger spins -- includes electrons and atoms whose component spins add up to half-integer values, often called “spin-1/2.” Laws of physics prevent fermions from occupying identical quantum states, and it is that prohibition that gives rise to the periodic table of elements. So when two electrons are in the same orbital around an atom, and all their other properties are the same, their spins will “point” in different directions, conventionally designated “up” and “down.” As a result, fermions cannot condense into the undifferentiated uniform condition of a BEC.

JQI Fellow Victor Galitski and research associate Tudor Stanescu, however, have determined a way to produce condensed bosonic atoms that display spin-1/2 properties. It requires several delicately controlled steps. First, researchers amass a population of atoms with multiple, slightly different (“hyperfine”) ground states -- that is, different possibilities for the atom’s minimum-energy conditions caused by tiny perturbations in the way the atoms’ electrons interact magnetically with their nuclei. The atoms are then irradiated by light of different frequencies, boosting them into an excited state.

As the atoms shed that absorbed energy by giving off photons, they drop back to one of the slightly different ground states. (See illustration below.) Because an atom can only leave its ground state by absorbing a photon of precisely the right frequency -- that is, the right energy content -- to raise it to an excited state, it will remain in the ground state unless such a photon arrives. By carefully manipulating the way laser light interacts with the atoms, researchers can arrange conditions so that atoms in those ground states never encounter photons they can absorb, in effect trapping the atoms into one of the different “dark” ground states. The result is a fragmented population of BEC atoms.

Because it is impossible to know which atom is in which state, they are in a “superposition” of all three states simultaneously. The superposedatoms can be in one of two conditions because the system offers two minimum-energy possibilities. These alternatives correspond to a degree of freedom which is, in effect, the same as a spin-1/2 particle has under ordinary circumstances, and they behave like fermions in ways that are, remarkably, mathematically describable.

In particular, electrons are affected by a phenomenon called “spin-orbit” coupling, which is a result of the fact that an orbiting particle experiences a constantly changing interaction between its intrinsic spin and the magnetic field it generates as it orbits.

The “pseudo-spin-1/2” bosonic atoms do the same sort of thing. In this case, the spin-orbit coupling -- produced by the action of the laser fields -- causes atoms in each of the two ground states to have opposite momentum vectors: If one is “left” the other is “right.” Changing the laser parameters alters the number and distribution in each group.

If this theoretical condition can be achieved and controlled, as seems very likely, it could have significant potential for quantum computing. For one thing, the two dark ground states serve as a two-level system that could form a “qubit” -- the quantum analogue of a binary digit, or bit. Again, because it is impossible to tell which of the two states an atom is in, the atom exists in a superposition of both states at once until a measurement is made, destroying the superposition and forcing the atom to take on one state or the other.

In this case, of course, the qubit value is the collective state of the entire ensemble of BEC atoms confined in a electromagnetic trap: the fraction of all atoms that was in each of the two momentum states. But the computer outcome could be the same. Instead of “reading out” a photon from a single atom to determine its state, researchers could measure the overall state of the new kind of BEC by simply turning off the trap and the lasers and letting the confined atoms expand. Because the atoms in each state have different momenta, they would leave the trap in different ways. (See illustration.) So measuring the
density of atoms in each condition would constitute reading the value of the qubit.

There is another potential benefit to using the new BEC form as a qubit. Quantum systems are notably vulnerable to random “noise” and to errors introduced by decoherence when delicate superpositions interact with their environment.

The new type of BEC qubits, however, would be protected by one of nature’s most inviolable principles: conservation of momentum. Although the momentum of any single component is not conserved, the total momentum of the entire system cannot change.

The new BEC model is only part of a comprehensive JQI effort to study cold atoms, magnetic fields and the burgeoning science of “spintronics.” The story will continue next month with another aspect of the research.