Say we have a beam of light—maybe we made it with a laser. We’d like be able to change the intensity of the beam so that we can alternately brighten and dim it. Moreover, we’d like to be able to do so quickly. Physically blocking and unblocking the beam just isn’t fast enough. So what do we do?

The solution is to make an electric switch so we can change how the light behaves via electrical signals. This is an electro-optic modulator (EOM). Two weeks ago, I introduced graphene to you all. And last week, I described some of the work I did on graphene as an undergraduate student in the Schibli lab. This time, I’ll explain the ultimate product of that research: a graphene-based electro-optic modulato like the one shown in the figure above. I want to emphasize that I didn’t build this device and I take no credit for its design. I did, however, do some experiments that helped prove that such a device was possible. I’ll talk about all of that today.

(Important side note: one can also make an acousto-optic modulator, which uses sound to change how light behaves.)

The Strange Dance of Light and Matter

Before I can tell you how a graphene EOM works, I need to briefly discuss how a material absorbs light. I’ve given fairly detailed explanations of absorption in the past. But today I want to focus on other things, so I’ll only be presenting a very rough, inaccurate picture of what goes on. Take it with a grain of salt.

We know from James Clerk Maxwell that light is a wave made of electric and magnetic fields. These fields feed into each other and wiggle back and forth, as shown below.

Electrons carry negative charge and are thus affected by electromagnetism. So when light passes through a material, its wiggling pushes the electrons around, accelerating them. But when an electron is accelerated, it leeches energy out of the light. This is absorption.

(Things are much more complicated than this, of course. In insulators, this wiggling causes all sorts of other things to happen to the light. It can bend, slow down, or change direction. We call this refraction, which I’ve written about before here. In a conductor, the light will probably be reflected.  Also, absorption is actually a quantum-mechanical effect.)

The Obstacles

To absorb the light, our electrons must be free to wiggle around. But many electrons aren’t free to wiggle in this way. The material might be too crowded, so that an electron that “wanted” to move would be trapped by all the other electrons in its way. Or an electron might be too tightly bound to the atomic nucleus to be able to wobble. This means that, if we can figure out how to control how mobile electrons are in a material, we can control how absorptive it is!

Chemical Doping

Now let’s talk about graphene. In its natural state, graphene has many electrons that are free to wobble (and thus absorb light). However, let’s say we pour some nitric acid on our graphene. Acid eats things because each acid molecule breaks into two pieces: a negatively charged anion and a single hungry proton (a positive hydrogen ion) that wants to chemically react with anything it can find.

Graphene is too tough for the acid to eat, but the proton still has an effect. It lands on the surface of the graphene, attracts one of the electrons within the graphene, and bonds to it. The electron stays within the graphene, but becomes immobilized so that it can’t wiggle around anymore, as shown below. One less electron to wiggle means one less electron to absorb any light that passes through the graphene, so this causes the whole graphene piece to become less absorptive. This process is called p-doping because we place positive charge on the surface of the graphene.

If we had used a base instead of an acid, we could have played the same game. The base’s negatively charged ion would have latched onto the graphene and bonded with one of its protons, which would free up an electron. If the graphene had already been p-doped, this might make the graphene more absorptive. If the graphene started in its natural state, the newly freed electron might crowd out its peers and make the material less absorptive. This process is called n-doping because we place a negative charge on the surface of the graphene.

(I’m glossing over a lot. The real story of doping involves the band structure of a material. I wrote about that here.)

So if we shine light through a graphene sheet, we can control how much of the light it absorbs by adding protons (p-doping) or removing them (n-doping).

Electrostatic Doping

Now remember, we’re trying to make an electro-optic modulator: a device that lets us quickly control how much light passes through it depending on an electrical signal. For our purposes, adding acids and bases is much too slow. We need a new trick. Is there any way we can mimic chemical doping?

As it turns out, there is! Say we place a sheet of graphene on top of some glass and sandwich our glass between two metal places, as shown below. If we apply a voltage across the plates, we can push electrons onto the graphene and charge it up. The result is that graphene now has more electrons, and they crowd each other out, preventing them from moving or absorbing.This process is called electrostatic doping, and it’s exciting because we can turn its effects on and off as fast as we can turn the voltage on and off.

Electro-Optic Modulator

Now this is an effect fast enough to make an electro-optic modulator. In the Schilbi Lab, where I worked as an undergraduate student, Professor Thomas Schibli and his students Chien-Chung Lee and Seiya Suzuki placed a piece of tantalum oxide—which has nicer properties than glass—on top of an aluminum mirror. On top of that, they put a sheet of graphene. And on top of that, they placed a ring of aluminum, as shown below. (The shape was a ring so that light could pass through the center.) When they applied a voltage between the ring of aluminum and the mirror, they were able to electrostatically dope the graphene and change how much light it absorbed.

It worked beautifully! Below is one of the devices actually made in the lab. On the left is an optical photograph of the device through a microscope. On the right is a two-dimensional color-plot of the “modulation depth,” which is a measure of how much the graphene’s absorption changes over time. The brighter the color, the bigger the modulation depth in that spot.

My Part in All of This

I take no credit for the idea of using graphene for an electro-optic modulator—that was all Professor Schibli. I also take no credit for tackling the engineering challenges of designing and building one—that was all Chien-Chung Lee and Seiya Suzuki. What I did was help prove this device could work before they started building it. I grew graphene samples and then Chien-Chung doped them with acid and measured their absorption. While the samples were still doped, I measured their Raman spectra to prove that the acid was having the expected doping effect. Afterwards, I looked at the Raman spectra again to look for signs of acid damage.

I did have the privilege, however, of seeing a working device. The team managed to get one functional while I was writing my undergraduate honors thesis on what I’d done. Pretty neat, huh?

Further Reading

• This is part three of a three-part series on graphene. If you missed them, you can find the first and second parts of the series here and here.
• I glossed over many, many details here. The true story of how doping works is rooted in the band structure of a material. I talk about that here.
• I also glossed over important aspects of how light and matter interact. I’ve covered these in several articles, but a good place to start is my article on refraction.
• The quantum mechanics of absorption is first detailed in my article on the Bohr model of the atom.
• I also talk about the interaction of light and matter in my article on lasers, my article on mode-locking, and my articles on scattering, which you can find here and here.
• If you’re very brave, you can look at my honors thesis, which explains all of this in extreme detail.
• I am a co-author on a paper detailing the proof-of-concept work. Sadly, it’s behind a paywall, but you can get it here if you have a journal subscription.
• The Schibli group also published a paper on the modulator. It’s behind a paywall too, but if you have a journal subscription, you can find it here.