Atomic telescope brings atoms to standstill

One of the abiding mysteries of physics is how to make the transition between quantum and classical objects. With very few exceptions, we live in a world that is clearly and obviously classical in nature. Quantum mechanics often defies our everyday expectations, which poses a problem. Why is the classical world classical when it is constructed by objects that really don't behave like classical objects?

There are now several ideas about how this transition occurs, and each makes some distinctive predictions. Unfortunately, the experiments necessary to test these predictions are really difficult since you may need to observe the cumulative effect of many small changes. But a new paper, published in Physical Review Letters, shows that these experiments may finally be possible.

What might a quantum to classical transition look like?

If you can get a cloud of atoms cold enough, then the wave-like nature of the individual atoms expands until they overlap and all the atoms start behaving like quantum objects. So for instance, a Bose Einstein condensate (BEC) of atoms will stay together rather than diffuse because the atoms all behave collectively like a single quantum object.

Even when they cannot form a BEC (even if the temperature is low enough, a BEC might not form), the wave-like qualities of the individual atoms will still overlap with other atoms. In this case, the atoms are attracted or repelled from each other through the mixing of these waves rather than colliding like a pair of billiard balls. As an atom slowly meanders through this cloud of atoms, it jumps about a bit depending on the mixing between the wave natures of the surrounding atoms.

These kicks also speed the atom up, increasing the temperature of the cloud. As this happens, the wave nature of the atom starts to shrink and it begins to behave more classically. By carefully examining how a gas of cold atoms expands, these kicks can be evaluated. The transition between quantum and classical can be tracked.

What changed that makes these new experiments possible? Just making the coldest gas in the known Universe, that's all. This work builds on the cooling techniques used to produce BECs, where temperatures in the nanoKelvin range are required. To obtain these temperatures, several different techniques are used. Most approaches have some intrinsic limit that keeps the gas at nanoKelvin temperatures or higher.

How to cool

Cooling atoms down to the nanoKelvin range is usually a multistep process. The first step is called optical molasses. As the name suggests, it uses a set of lasers to slow atoms down. I've described this before (see sidebar).

The issue with optical cooling is that the slower the atoms go, the less efficient the cooling becomes. As the atoms become colder, the more precisely you need to control the color of the lasers, making the experiments technically difficult.

Even if you ultimately overcome these problems, each of the transitions between the ground and excited state have something called a width, which describes the range of colors that can be effectively absorbed to drive the atom from the ground state to the excited state. The very presence of other atoms will broaden that width, making for a wider range of colors that can drive the atomic transition. So once everything's cooled down enough, the laser will start to excite atoms that are standing still and the temperature will not decrease any further.

Once these limits to optical cooling start to dominate, researchers can switch to a sort of evaporative cooling. The cold atoms are held in a magnetic bowl. The coldest atoms will sit at the bottom of the bowl, while the hot atoms spend more time higher up the side of the bowl. To stay in the magnetic bowl, the direction of the atom's internal magnetic field (atoms are rather like bar magnets) has to be anti-aligned to the magnetic field of the bowl. If the magnetic field of an atom were to be flipped, it would be driven out of the trap.

Because the magnetic field of the trap also changes the energy levels of the atom, the energy required to flip the orientation of the atom's magnetic field depends on where the atom is in the trap. This means that the hottest atoms can be selectively removed from the gas by shining a carefully chosen radio frequency on the cloud of atoms. Evaporative cooling is also limited; you can't cool indefinitely because you end up with a smaller and smaller number of atoms.

These two options do not allow researchers to cool a cloud of atoms to the temperatures required to observe quantum to classical transitions as accurately as we need to.

Turning cold up to 11

To obtain atoms in the picoKelvin (10^-12K) range, the researchers used a very nice trick. Once the atoms are cooled down to the nanoKelvin range, the magnetic bowl is removed and the cloud shot upwards through the vacuum. As the cloud moves, it starts to expand. However, instead of just letting the cloud freely expand, the researchers applied a new magnetic field that, from the point of view of the atoms, looks like a lens.

The idea here is to stop the cloud from expanding as it moves. Atoms that hit the lens closest to the outside experience a larger force due to the magnetic field. As a result their outward motion is slowed; atoms traveling right through the center of the lens experience no force at all.

Once the cloud has passed through the lens, it continues to expand along the direction that it is moving, but the transverse expansion (the expansion sideways compared to the direction of travel) is slowed drastically. The cloud can then be described by two temperatures: in the direction of motion, it still has a temperature of about 2nK. In the transverse direction, however, the temperature is only about 50pK—40 times less.

This is only the start though. Like all lenses, a magnetic lens has an intrinsic limit to how well it can focus (or, in this case, collimate) the atoms. Ultimately, this limitation is given by the quantum uncertainty in the atom's momentum and position. If the lensing technique performed at these physical limits, then the cloud's transverse temperature would end up at a few femtoKelvin (10^-15). That would be absolutely incredible.

A really nice side effect is that combinations of lenses can be used like telescopes to compress or expand the cloud while leaving the transverse temperature very cold. It may then be possible to tune how strongly the atoms' waves overlap and control the speed at which the transition from quantum to classical occurs. This would allow the researchers to explore the transition over a large range of conditions and make their findings more general.

Science is filled with missteps and false leads. It makes predictions about the success or failure of a new technique fraught and, almost certainly, doomed to failure. Nevertheless, I am pretty confident that this line of work will be fruitful.

Physical Review Letters, 2015, DOI: 10.1103/PhysRevLett.114.143004