Hold it right there: how (and why) to stop light in its tracks
We are taught in school that the speed of light is a universal constant. Yet we also know that light travels more slowly through materials such as water and glass. Recently, we have even discovered that light can actually be made to stand completely still.
In fact, it was first done a long time ago ... in a galaxy far, far away. In a scene from the latest Star Wars film, Kylo Ren stops a blaster pulse using The Force. The pulse is frozen, shimmering in mid-air. More recently, for our paper published in Nature Physics this week, we stopped a pulse of laser light using a rather different method, by trapping it in a cloud of cold rubidium atoms.
Rubidium and other similar atoms have been used previously to slow down and store light and even to trap it. These systems all work by absorbing and re-emitting laser light from the atoms in a controlled way.
But we found a new way to trap light, by using the light to write a particular "shape" into the atoms. When the light was re-emitted, it became trapped in the atoms. It turned out that once we had picked the right directions and frequencies for our lasers, the experiment was pretty straightforward. The hard part was figuring out the right frequencies and directions!
Why do this? We are interested in trapping light because our ultimate goal is to make individual light particles, or photons, interact with one another. By interacting directly, the photons will become entangled. By scaling this up to many interactions, involving many photons, we could theoretically create the intricate states of information necessary for powerful quantum computing.
Unfortunately, photons interact incredibly weakly with each other, but they can interact more strongly if they can be confined in a particular material long enough to enhance the interaction to a more useful level. In fact, these sorts of interactions have recently been demonstrated by multiple research groups around the world, often by using atom clouds to confine the light. But, as I'll explain below, our new stationary light system may have advantages when it comes to getting photons to interact.
Quantum computing is an exciting and rapidly evolving field of research, and our team is part of the Australian Research Council's Centre for Quantum Computation and Communication Technology. There are many different potential platforms for quantum computing. For example, the centre's UNSW team has demonstrated quantum computing operations using phosphorus atoms embedded in silicon chips.
But our group mainly studies light, not least because it is very likely that light will play some role in quantum computers. It offers a convenient way to send quantum information within or between computers because, unlike atoms or electric currents, it is not vulnerable to stray magnetic or electric fields. It may even be possible to perform quantum computation using light, and this is the idea that motivates our research into stationary light.
Our team has been able to store and retrieve pulses of light in the same system. We have also been able to show that quantum information encoded in these light pulses is preserved - meaning that it can form the basis of computing memory.
However, this is not sufficient to generate the sort of interaction we want, because the light is entirely absorbed into the atoms and it can no longer interact. Instead, we need to trap light in the memory, not just store it.
While researching how to trap light in the atomic memory, I discovered using a computer simulation that a particular kind of shape written into the atomic memory would produce stationary light. By retrieving the light in two directions at once, the light could actually be trapped in the memory. All the light being re-emitted throughout the memory would destructively interfere at the ends of the memory and not escape.
The simulations also predicted other interesting behaviour: if the wrong shape was written, some light would escape, but the memory would rapidly evolve to a shape where the light is trapped. This could be useful for stationary light by making it more robust, but it may also be useful for other optical processing.
We were able to demonstrate all of this behaviour experimentally using our atomic memory. Unlike Kylo Ren's frozen blaster pulse, it was not possible to see the stationary light directly (to see something, photons have to travel from the object to your eyes, and these photons were not going anywhere). Instead, the fact the behaviour of the system matched our predictions so precisely confirmed that the light was indeed stationary.
Light has previously been trapped in a similar system. What makes our system new and interesting is that we believe it is the most convincing demonstration so far, but also that the behaviour of our stationary light is radically different. We believe that this new behaviour, where the light travels more freely through the memory, could allow for stronger nonlinear interactions.
This experiment is only a single step on the long road to optical quantum computing. The next step will be to prove that photons can indeed interact with one another within our system. Looking much further down the road, we hope this will give rise to a device that can use some of our discoveries, among many others, to generate the intricate states of many entangled photons necessary for an optical quantum computer.