I am a research associate at the University of Bristol and mainly work on “photonic crystals” and placing “resonance cavities” and “waveguides” in them, with the hope that they might one day be used to make “single-photon sources” and “spin-photon entanglement gates” for future “quantum computers” among other things.

Well, let’s start from the “quantum computers”. You may have heard of them.

They are a new way of computing, exploiting the “superposition” (a bit can be in multiple states at the same time) and “entanglement” (multiple bits can be linked together, so that if you check one you know the values of the others) properties of quantum mechanics to solve problems more efficiently than a classical computer can.

In practice, they could lead to another computing revolution and allow us to factor large numbers very quickly, simulate molecules, find information faster in large databases, etc.

Unfortunately, they are really hard to make!

Mainly because “quantum states” (i.e. the information stored in “quantum bits”) tend to “decohere” very quickly. For example an electron prepared so it stores a 0 or a 1, will tend to end up in a random state after a while, due to interacting with all the matter around him.

Photons (the smallest unit of light) are a little bit better for storing information over long distances, but unfortunately they also have problems:

• * It is hard to make a single one on demand. We do not want millions of photons, like coming from lightbulbs or lasers. And we need to be able to make them at the right time, and quickly enough, so we can process a lot of them in the right way to “compute” things.
• * They are hard to measure. A lot of photons can generate enough current in a solar cell so you can detect them. But this is much harder when you want to measure just one of them!
• * They don’t interact with each other! Unlike Star Wars and Ghostbusters, nothing happens when you cross two beams of light. So if we want to pass information from one photon to another, or combine information from two photons, we need to go through matter.
• * They can easily be absorbed or scattered. Any bit of dust or other impurities can cause the photon to not make it through the whole circuit. And you do not want to lose data when computing something.

But they have advantages:

• * They travel faster than anything else (at the speed of light).
• * They are easy to set to 0 or 1 and measure if they are 0 or 1. The 0 and 1 for photons are usually defined as being horizontal and vertical polarizations. And polarizers have been around a long time and are easy to use. Similarly, when we want to check if they are horizontal or vertical, we pass them through a special crystal (“polarizing beamsplitter”) that sens them one way or the other, based on their polarization state. By placing single photon detectors at each possible output, we can then know if it is 0 or 1. (in comparison controlling the “electron spin” requires freezing temperatures and electromagnets).

The two main problems I try to help solve are:

• * Making single photons on demand (i.e. preparing the input data)
• * Making “transistors” for future quantum computers (i.e. something to process the data with)

It turns out that one possible solution is to trap the light in a small box with mirrors on all 6 sides and add one or more atoms in there, which are made so that they can only absorb or emit one photon of a specific wavelength/frequency (think color).

Unfortunately, a mirror, like the one you might find in bathrooms, is not good enough, because they tend to absorb too much light. Remember: we are working with the smallest unit of light here, a single photon.

So instead, we use so-called dielectric materials, which don’t absorb light much, but can still reflect some of it.

If we want to reflect a lot more, just need to add more layers of it.

Then we exploit a quantum property of light: the wave-particle duality!

If we look at the photon as a wave instead of a particle, we can align the layers so that all the tiny reflections from each layer have their peaks perfectly aligned.

To get the overall reflection, you simply add up all the waves. If their peaks are aligned, you can get a very strong reflection (theoretically 100% for infinite layers with no absorption).

Such a stack of layers is called a “photonic crystal” and is how we make our “light traps”.

Here are some pictures of the tiny structures we are able to make:

An example of a 3D photonic crystal: The Rod-Connected Diamond structure. The whole block is about 20 times smaller than the width of a human hair! The figure was taken from: https://doi.org/10.1021/acsphotonics.9b00184

Examples of structures made for fun with the Nanoscribe, our 3D printer. Taken from my thesis: https://research-information.bris.ac.uk/en/studentTheses/modelling-and-fabrication-of-nanophotonics-devices

I mostly sit in front of a computer (especially during the lockdown) doing coding, simulations, data analysis, writing papers (to publish my research results) and proposals (documents to request funding for new research).

But sometimes I need to got to the lab. And there I either work with lasers and light in a dark room or I put on a complete protective suit to protect myself from chemicals, but also to protect my lab from all the dirt on me!

In one of the labs, we have a machine to write really tiny structures. We made a 3D model of the Clifton suspension bridge in Bristol that is smaller than a human hair for example!