NZSM Online

Get TurboNote+ desktop sticky notes

Interclue makes your browsing smarter, faster, more informative

SciTech Daily Review

Webcentre Ltd: Web solutions, Smart software, Quality graphics

Feature

All About Atom Lasers

Physicists are trapping "creatures" from the nuclear zoo.

Howard Wiseman, Matthew Collett, Ana Martins, and Dan Walls

Ever since the birth of modern science in the 16th century, physicists have been arguing about the nature of light. Newton thought that light was composed of particles. In the early 19th century, experimentalists such as Young with his famous two-slit apparatus demonstrated interference effects which are characteristic of waves.

When the peak of one wave coincides with a trough of another wave (destructive interference), no light is seen, whereas when two peaks or two troughs coincide (constructive interference) the light is twice as bright as average. These light waves were shown by Maxwell to be electromagnetic in nature. However new experimental results were found, in particular the photoelectric effect, which seemed to show that light was composed of discrete bundles of energy. This was explained by Einstein using quantum theory, and it is now understood that the quantised version of Maxwell's equations does indeed give particle-like properties to light waves. These particles of light are called photons.

In contrast with this history of light, natural philosophers since Democritus have believed matter to be composed of tiny particles called atoms. In this century, we have discovered that atoms are not fundamental particles, but for many practical purposes they remain intact and discrete lumps of matter.

Is it a Wave? Is it a Particle?

However the same quantum theory which shows that light waves have a particulate nature implies that all particles must have a wave-like nature. The wave-particle duality of matter was observed with electrons in the 1920s, shortly after its prediction by the French physicist Louis de Broglie. Until a few years ago, the wave nature of atoms had never been seen. Just this year, some very beautiful experiments have been performed in Europe and the US which show interference between de Broglie waves of atoms, in exact analogy with the earlier interference of light waves.

Although the dual wave-particle nature of atoms is beyond question, the atomic wave-packets which have been used to show interference are in one respect quite different from typical light wave-packets. A typical wave-packet, or pulse, of light contains a huge number of photons, perhaps 1010. This means that interference effects can be seen using a single pulse of light split into two and then recombined, as in Young's experiment. Where the light waves interfere constructively, many photons will be found, and where they cancel, no photons will be seen.

In contrast, a typical atom wave-packet contains only one atom. Atomic interference effects can be seen only by splitting and recombining atom wave-packets, one at a time, over and over again. This is necessary to build up enough statistics to verify that most atoms will be observed to be where constructive interference of the atomic waves is predicted, and no atoms where destructive interference is predicted.

A Bunch of Atoms

With light it is very difficult to make a wave-packet with just one photon. The reason for this is that photons belong to the class of particles called bosons, named after the Indian physicist Satyendra Bose. Bosons like to be with other bosons of the same sort -- they bunch together. Thus if there is one photon in a wave-packet, it tends to encourage other photons to appear in the same wave-packet. This does not violate conservation of energy as the energy for the extra photons has to come from somewhere. This is the principle by which a laser works -- energy is pumped into the lasing medium and the photons which travel through that medium cause that energy to be converted into more photons in the same wave-packet.

With matter, the situation is inverted in that it is very difficult to get many atoms into a single wave-packet. In fact, it is impossible for certain sorts of atoms. Some atoms are bosons, like photons, but others are fermions. Fermions (named after the Italian physicist Enrico Fermi) behave in the opposite manner to bosons -- they can't stand each other's presence so it is impossible to put two together in the same state. What determines whether an atom is a boson or a fermion is whether the total number of its constituents (protons, neutrons and electrons) is even or odd respectively. So different isotopes of the same element differing by one neutron will behave in a completely different manner if you try to get more than one in the same quantum state (such as having the same position, or the same momentum, or being in the same wave-packet).

Another difficulty with atoms is that, unlike photons, they cannot be created out of pure energy. In order for Bose atoms in a particular quantum state to encourage other such atoms into that same state, there must be some mechanism for transferring atoms from other states into that particular one. The preferred state in which one wishes to collect many Bose atoms is in practice the state of lowest energy (called the ground state) in a trap for atoms.

Atom Traps

An atom trap is a device which confines atoms to a small region of space by a combination of magnetic and optical forces. Many atoms are put into the trap in many different states. Then, by cooling the atoms (reducing their energy), some atoms are forced into the ground state. Being bosons, this encourages more of their fellows to join them in the ground state. After a certain point, this positive feedback process takes off and a large fraction of the atoms are rapidly cooled into the ground state. This phenomenon is known as Bose condensation. Note that the atoms in the ground state do not attract other atoms by some physical force. The condensation is an effect purely of Bose quantum statistics.

Predicted by Bose and Einstein 70 years ago, this ideal condensation of particles had not been observed until this year. The technology for trapping and cooling atoms has been advancing rapidly over the past decade, to the point where the trapped atoms can now be cooled enough to allow condensation to occur. This has been achieved by two groups, one in Colorado using rubidium atoms and one in Texas using lithium atoms.

The method of cooling is conceptually very simple. In any trap there will always be some atoms which are hot enough (i.e. moving fast enough) to escape from the trap. The remaining atoms will therefore be, on average, colder. This is essentially identical with cooling by evaporation, which is how an old-fashioned ice-box works, for example. Over the course of the experiment, the trapping force is decreased until most of the atoms have escaped, and those that are left are very cold. This "evaporative cooling" technique has produced the coldest matter in the world, with a temperature of just 20 billionths of a degree above absolute zero.

Atom Lasers

The cooling of many thousands of atoms into the same quantum state is the first proof that atoms can behave identically to photons in their statistics. However, it is still a long way from making an atom wave which is like a light wave from a laser. The light wave from a laser has the property that it is intense (many photons per state) and coherent. Being coherent means that the fluctuations in its intensity (number of photons) and its phase (the position of the peaks and troughs of its waves) are as small as quantum mechanics allows. A device which could similarly produce an intense and coherent source of atomic waves would, by analogy, be called an atom laser.

At the University of Auckland, we have been investigating various models for an atom laser. At this stage, the work is purely theoretical, but the achievement of condensation for atoms shows that our ideas could soon be realisable by experimentalists.

There are a number of difficulties involved in building an atom laser which go beyond those required for Bose condensation. First, it is necessary to provide a continuous source of atoms into the trap, rather than cooling those present from the start. Second, the condensed atoms have to be removed from the trap in such a way as to form an output beam. (In current experiments, when the trap is turned off the condensed cloud of atoms simply expands in all directions.) Third, in order for the output beam to be coherent, the condensate must attract most of the trapped atoms, which requires very efficient cooling. Fourth, for the same reason, the rate of collisions between the atoms in the condensate must be kept relatively low.

New Applications

If an atom laser is built, it will open up new areas of research in atomic and optical physics, in the same way that optical lasers have done. At this stage it is difficult to speculate on what the technological consequences might be, but one obvious application is in interferometry. By using precisely the interference experiments mentioned above, one can measure distances and forces very accurately. Because they have mass, atoms are much more sensitive to gravitational fields than photons, so atom interferometry could potentially be used for mineral exploration and the like. To do this most efficiently would require an intense wave of atoms with minimal fluctuations, which is just what an atom laser beam would be.

Another property of atoms which is different from photons is that they interact with each other. This leads to interesting behaviour of coherent atomic wave-packets, which other theoretical physicists around the world have been studying. A further application is atom lithography -- a tightly focused atom laser beam could be used for writing microcircuit designs on a crystal substrate.

Clearly the area of atom lasers and atomic Bose condensation is at the forefront of current physics and promises to lead to entirely new physical phenomena, as well as a deeper understanding of more familiar topics.

Dr. Howard Wiseman works in the Physics Department at the University of Auckland.