Editors Note: Light is going to play a vital role in the future, would like to talk about some of the areas where light will supersede some things that were thought not to be excelled at any time. Who would believe me, isn’t this kind of hard to believe right here Agcat
“I saw two photons holding hands the other day”
Credit: Chris Lee –
All the early quantum computing work was done with light. Light is very easy to manipulate: a few mirrors, crystals, and light detectors and you to can have your very own quantum computer. Over the last two decades, though, that’s changed. Almost all the major developments have used things like ions, rings of superconducting current, or defects in crystals.
This was, in some sense, a reasonable progression. To perform logical operations, you have to modify one quantum state based on the state of another. Light beams, however, tend to pass right through each other without even waving, let alone stopping to chat. Contrast that with ions. Two ions, being charged, cannot avoid talking to each other. That means the quantum state of one ion can strongly influence the state of the other. This makes logical operations much easier.
The flip side is that quantum states that are easily modified are also easily destroyed by the environment. The quantum state of light, on the other hand, is remarkably robust. This has been demonstrated rather spectacularly by performing quantum key distribution between two locations via a satellite.
So light-based quantum states are certainly not out of the picture for quantum computing, though they are mostly considered for information carriers between locations. At each location, the light’s quantum state is transferred to something else to perform the computation. But that transfer may be unnecessary now that researchers have developed material structures that allows light to strongly modify light.
Two photons walk into a glass bar, neither of them notice
So, why are photons so aloof? The problem is that they have to talk to each other via an intermediary: the material in which they travel. When light travels through glass, it slows down because the light fields make all the electrons in the glass jiggle about. This slow down produces the material’s refractive index, which we tend to think of as independent of the light’s brightness. If we ramp up the light intensity, the electrons jiggle a little more vigorously, but it doesn’t change the way the light moves through the medium, so the refractive index hasn’t changed.
For very bright light beams, however, the electrons have to move much further than they are comfortable with—the electrons are bound to atoms, so there is a limit to how far they can move. Once the light makes the electrons uncomfortable, the intensity of the light changes the refractive index. The way the light moves through the material changes, and all sorts of strange things happen. New colors can be generated, the light can focus, or a light pulse might steepen and becomes even shorter and more intense.
If the glass is between two mirrors, it can make these weird effects more apparent. Starting with a very dim light, the distance between the two mirrors tells us what color of light will enter into the space between them (if the front mirror lets light leak in). Light with the right color will reflect back and forth between the mirrors, and build up in brightness as more light leaks into the gap. At the same time, light leaks out of the gap via the second mirror.
After a while, the flow in equals the flow out, and we have reached equilibrium. Remarkably, all of the light that is shone on the front mirror, apparently, passes through the mirror into the gap—none is reflected. The light leaking out of the second mirror has the same brightness as the light shining on the front mirror. And, the light between the mirrors is extremely bright; the more reflective the mirrors, the brighter the light between the mirrors. Effectively, the gap between the mirrors acts as light storage.
If the light is bright enough, it will change the refractive index of the material in between the mirrors. That changes the color of light that can be accepted into the gap between the mirrors. As a result, we never reach the equilibrium described above. Instead, light is, initially, not reflected by the front mirror. As the brightness in the gap increases, however, the front mirror suddenly starts reflecting light. Effectively, light has switched the direction of the flow of light.
This is exactly what we need for our optical quantum computer: light changing the state of light.
The common theme running through these stories is brightness. You need high brightness, and that means lots and lots of photons. But quantum states are stored in single photons, which are the very opposite of bright. That is why optical quantum computers are languishing.
Two photons walk into a pillar, only one walks out
This is where new materials research comes into play. The goal is to create structures that are so sensitive that a single photon can change their properties. This works with a single atom. Let’s imagine that we have a single atom pinned in space, and a squirt gun that fires single photons. We have excellent aim, so every photon will hit the atom. And, because we are clever, we choose the color of our photon so that it matches the energy separation required to excite the atom from its ground state to an excited state.
If we fire a single photon at the atom, it absorbs the photon, and, some time later, spits out a similar photon in a random direction. But, if we fire two photons, one right after the other, then the atom absorbs the first photon and lets the second pass through. One photon controls the passage of another. In contrast to the example above, this only needs a single photon to have an effect. It is exactly what we need.
But atoms don’t stay put, unless you surround them with other atoms. And atoms don’t absorb any old color of light, but only those colors that nature has chosen. Even worse, atoms are very small: the chance of a photon actually hitting an atom and being absorbed is very small. The Universe has let us down, and we need, to put it bluntly, a better class of atoms.
Enter quantum dots. Quantum dots are tiny balls of material. The ball is so small that, if you try and pass a current through it, only a single electron enters the ball at a time. Each electron blocks the entrance of the next, and the single “free” electron contained within a quantum dot behaves like an electron in an atom. But unlike atoms, the spacing between different energy levels is given by the size of the ball of material. In other words, quantum dots are like designer atoms. And since they are much bigger than normal atoms, they’re easier to hit with a photon.
Researchers have now put quantum dots between two mirrors. This is all at a very tiny scale, so the mirrors are tiny pillars, just a few micron in diameter, and the spacing between the mirrors is just a few hundred nanometers. That, however, is not very special—everyone with a cleanroom makes pillars with mirrors and puts quantum dots in between.
There are two developments that make this new work special. First, the researchers have a fabrication technique that allows them to place a single quantum dot close to the center of the gap. Then, because every quantum dot is slightly different, they have electrodes attached that allow them to tune the energy levels of the quantum dot so that the absorption and emission color of the quantum dot matches the color wanted by the gap between the mirrors.
So, picture this: I shine a light on the tiny pillar; it has the right color for the gap. The light leaks through the mirror and reflects back and forth in the gap. Suddenly, the quantum dot grabs the photon and moves to an excited state, changing its refractive index. Suddenly, the mirror is reflective and I see light coming back to me. And, I can do this whole trick with single photons.
A pulse of light containing, on average, less than one photon, can be shone on the mirror. On average, less than a single photon leaks through and is reflected back and forth. Once it is absorbed, the mirror becomes reflective and sends the next photon back in the direction it came.
Single-photon mirrors: What are they good for?
So, what is this good for? Well, it is very good at producing streams of single photons. Let me put it this way: if I take a laser that produces a pulse of light every few nanoseconds and reduce the intensity such that, on average, there is one photon per pulse, I will not have a single photon light source. If I were to measure the number of photons in each pulse, I would find that I am quite likely to find pulses with no photons, many pulses with two or more photons, and very few with just a single photon. Essentially, photons don’t talk to each other, but they like to keep each other company.
To get a stream of single photons from a laser, I have to reduce the intensity of the light beam such that, on average, every tenth pulse should have a photon. And even then, a few percent of the pulses will have two or more photons.
But if I take a light beam that has, on average, about a photon every second pulse in it and shine it on this device, then the reflected light pulses will actually only contain a single photon per pulse. Or, more accurately, after reflection pulses that contained two photons or less will most likely have only a single photon. Pulses that contained more than two photons will contain more than a single photon after reflection. This is because the quantum dot can absorb, at most, a single photon, so subtracting more than one photon from a light pulse is difficult.
But that is just the beginning. This also paves the way for efficient photon gates: one photon changing the state of another. Although not demonstrated in this particular parcel of research, I expect that there will be a paper on that fairly shortly.
What is really exciting about this is that it is built with technology that can, with some work, be integrated into optical integrated circuits. So you can imagine shining a pulsed laser into the entrance, and the first device converting that to a flow of single photons. These can then proceed to a gate that initiates the correct quantum state for computation. Then the photons can be directed to different versions of these little pillars, where the photons modify each other’s states to perform a computation. And all with very high efficiency.
I am quite excited by this development. The fabrication is high quality, so we are not searching through a material for a randomly placed defect. And they are using long lasting photons, not ephemeral, short-lived states in a circulating supercurrent. Photons have always been a great way to transport quantum information; now there may also be a way to efficiently compute using them.
Nature Nanotechnology, 2017, DOI:10.1038/NNANO.2017.85