return to the BubbaSoft Knowledge Base Page
	Physics with  Bubba Drew
Home · Recent Articles · Visit Forums · Chat Live · Contact Guide · Free Newsletter

Bringing Light to a Standstill

The dictum that light can never be trapped or brought to a standstill has been challenged by two independent experiments. Both experiments were able to take a pulse of light, stop it inside a cloud of atoms and then release it at will.

The experimental results are being published in forthcoming issues of Nature and Physical Review Letters, both prestigious research journals that have each taken the rare action of publishing the papers online before the due date for publication because of the immense interest in the results. On top of this, the competitive element of research publishing has played a hand, with both journals keen to demonstrate that they attract the best of the best to their pages.

In this article, we'll walk you through the physics behind the experiments and show how this seemingly impossible feat has been achieved. Interestingly, the two experiments take a somewhat different approach. That shows the concept has the potential to be used more widely than from just a single proof-of-principle experiment, which may have depended on fortuitous coincidences of nature to work.

Also interestingly, both teams are from Cambridge, Massachusetts and have probably been aware of each others work for some time. We are seeing the end result of a long-term battle to achieve the goal of stationary light. 

The two teams are

1. Chien Liu, Zachary Dutton, Cyrus H. Behroozi and Lene Vestergaard Hau from the Rowland Institute for Science and Harvard University, publishing in Nature 409, 490 (25 January 2001) and who we shall call the sodium team.
2. D. F. Phillips, A. Fleischhauer, A. Mair, R. L. Walsworth and M. D. Lukin from the Harvard-Smithsonian Center for Astrophysics, publishing in Physical Review Letters 86, 783 (29 January 2001) and who we shall call the rubidium team.

Our discussion of the experiments will be broken into three parts. Firstly, the similarities and general idea behind the experiments will be explained. Then the specific details of the sodium and rubidium experiments will be given.

Next page > How Light Was Halted > Page 1, 2, 3, 4

Bringing Light to a Standstill

Past experiments have successfully slowed light down to a crawl. However none had achieved a complete stop of a light pulse. The new result of stopping light follows on from the slow light and uses the same technique - called electromagnetically induced transparency (EIT).

In EIT, signal light pulses are directed into a dense gas which is normally opaque to the light. However, a second light beam (the control beam) is shone on the gas and this changes the internal electronic state of the atoms in the gas. This change is enough to make the gas transparent to the signal pulse.

The basic idea behind stopping light is to use the control beam to allow the signal pulse enter the gas and then turn off the control beam. This essentially traps the signal pulse inside the atomic gas. When the control beam is turned on again, the signal pulse is released and free to travel out of the gas.

The amazing thing about this process is that the quantum state of the signal light is unaffected by this stopping and releasing. The physicists conducting the experiments have determined that the information about the photons becomes imprinted on the atoms in the gas. When the control beam is turned on this information allows the signal photons to be regenerated exactly as they were before.

The experimental groups mention possible applications of this technology in quantum information processing. The ability to trap and hold information contained in photons may allow all sorts of new ways of manipulating information. Combined with the recent advances in optical quantum computing, we could be seeing some new designs for quantum computers fairly soon. (See Optical Quantum Computing) 

Each experiment was done slightly differently and the details of each are discussed in the next sections.

Next page > The sodium experiment > Page 1, 2, 3, 4

Bringing Light to a Standstill

The experiments with sodium occurred in a cooled gas. The typical cloud of atoms in these experiments contained about 11 million atoms at a temperature of 0.9 mK, just above the critical temperature for transition to a Bose-Einstein condensate. The cloud was cigar-shaped, about 339 mm long and 55 mm across.

A signal photons are about 3.4 km long in free space. However, the density of the gas meant that the light was slowed to approximately 28 ms^-1, at which speed the pulse compacts to less than the size of the sodium cloud. Not all of the energy in the signal pulse can be contained in a specified area but the excess outside the cloud was equivalent to less than 1/400 of a photon.

The trapping phase in this experiment, where the control beam is turned off ranged from a few microseconds up to hundreds of microseconds. In each case, the pulse of light was re-emitted when the control beam was turned on again. The difference between the output signal pulses at different trapping durations was the intensity of the pulse. The longer the pulse was trapped, the smaller the output pulse. However, it is important to realise that even though the intensity of the pulse was reduced (i.e. the number of photons), the state of the photons remained the same as at input.

The experimentalists determined that the intensity of the output pulse related to trapping time dies off according to an exponential decay. This suggests that the pulse information, when stored in the atoms, is gradually destroyed. This is expected and is called loss of coherence. It happens because the atoms still have some thermal motion and that motion destroys the stored information. The 1/e decay time for the coherence was measured as 0.9 ms. That means the output signal will drop to 1/e (about 28%) of its intensity after a trapping duration of 0.9 ms.

What is interesting about this partial transmission of information is that the pulse given out does not carry all the possible information away with it. The experimentalists showed they could regenerate a second and sometimes third pulse that had far fewer photons but those photons contained the information of photons that had come in with the original signal pulse.

The sodium group believe that by conducting this experiment in a Bose-Einstein condensate, much of the loss due to coherence decay could be reduced (by the definition of a Bose-Einstein condensate). That advance could allow for quantum information processing during the storage time by controlling the atom-atom interactions.

Next page > The rubidium experiment > Page 1, 2, 3, 4

Bringing Light to a Standstill

The rubidium experiments were quite different to the sodium experiments in a number of ways. Whereas the sodium experiments occurred in an ultra-cold gas, the rubidium experiments occurred at ~70-90°C. The rubidium vapour was contained in a cell about 4 cm long.

Photons from the signal pulse slowed to about 1 km·s^-1 in the vapour cell. Note that this is a factor of more than 5 orders of magnitude (100 000 times) slower than light in a vacuum.

The experimentalists conducted a series of experiments that had similar results to the sodium group. They could also trap photons for hundreds of microseconds.

The rubidium group goes on to examine the theory of this process in their Physical Review Letters paper. They describe how the information from the photons is stored in the spin states of atomic electrons. Their theoretical simulations of the experiment match well with the experimental results indicating that their theoretical analysis is probably quite accurate.

They point out that although perfect storage and retrieval of the light pulses should be possible but their experiment contained a number of factors that inhibited perfect results. They included the fact that atoms in the cloud would wander out of the region where the light beam was present and at high atomic densities, the atoms would collide together in what are called spin-exchange interactions in which the coherence of the spin states would decay.

Overall, these techniques of trapping light pulses and their information is looking very promising for quantum information processing. As the techniques have been performed over a wide range of temperatures, the practicality of the techniques seems quite reasonable. There is still a long way to go before these techniques would form the basis of quantum computers but the possibilities the experiments raise open up many new ideas for the potential of this fledgling field.