quantum computing #2

Some of the other research Prof. Kimble's group is working on involves cavity QED. Though it may sound like a highly specialized field to some, it's basically the study of how light and matter interact in confined regions (as opposed to outer space). In the classical limit, it becomes Maxwell's equations for waveguides or resonant cavities (like a home microwave!). Technology for creating high quality factor (high-Q) resonant cavities plus improvements in magnetic trapping of atoms has created the possibility of seeing the interactions between photons and electronic levels on a quantum level.

An alternative approach is to use photonic crystals to create a high-Q cavity, and embed either one or a few atoms or quantum dots in the middle of the cavity. Photonic crystals have two attributes that are of great relevance: (1) they can effectively confine light to small regions with low losses and (2) they can be used to convey photons over long distances (one might even replace fiber optic cables connecting the internet between the US and Europe with photonic crystals).

The other piece of the puzzle is nonlinearity. Single photons interact very weakly with one another in vacuum (an order \alpha^2 process in quantum electrodynamics), but one can introduce matter that can mediate a strong interaction between the same two photons. Most of these nonlinearities come from electromagnetically induced transparency (EIT), a remarkable physical phenomena whereby two absorption processes in a 3-level atom "cancel" such that light at a probe frequency can pass through without absorption in the presence of strong nonlinearities (normally this would be impossible). That's where the name comes from. An additional wrinkle is that if you introduce a fourth atomic level, one can change where this cancellation takes place with a different frequency (the "gating" frequency). In our system, we envision sending in a single gating photon, which will then change the property of the atom(s) inside the resonant cavity such that a single probe photon will behave differently than it did before, e.g., it would be transmitted instead of reflected. As a result, one obtains an entangled state between two photons. This idea could be used as the basis for a quantum network. For more details, click here for our preprint.

quantum computing

Another interesting topic I've been thinking about recently is quantum computing. Professor Jeff Kimble from Caltech came to give a talk at MIT two days ago. He's one of the people responsible for quantum teleportation. The way that works is pretty funny: the first person, Alice, takes some unknown quantum state and entangles it with some quantum source; then she measures it, which destroys the unknown quantum state, and then sends the classical result over a normal communication channel; then the recipient, Bob, "subtracts" the result from the same quantum source and viola! he regenerates the original quantum state. The original proposal in in this paper: Bennett et al., Phys. Rev. Lett. 70, 1895 (1993). Obviously it's a cool application. Note that it has to obey special relativity, i.e., if Bob gets impatient and tries to guess what Alice is sending him, then the state that he constructs is a random mixture of all four possibilities, which conveys no information. It's also interesting to note the way in which the information is transmitted: clearly, one can't encode a 2-state quantum system with two classical bits of information. However, on the other hand, the quantum source in this example doesn't know anything about the unknown quantum state Alice was using. So what you've done here is split up the information into a completely classical and a completely quantum-mechanical piece, which can have all kinds of implications for quantum communication, calculations, information storage, etc.

On a lighter note, this work was featured on the Daily Show a number of years ago (back in the days of Craig Kilbourne). They said that it was a really good application if you're a beam of light and you need to get from one side of the optical bench to the other. I guess they must have been pissed about that since they called Prof. Kimble a "quantum creep". :-)

In the next post, I'll talk about some of his recent work and how it ties into one of my current research projects.

Sunlight Photonics website

For more information on Sunlight Photonics, click here.

Sunlight Photonics business concept

Here is the business concept behind Sunlight Photonics, a semi-finalist in the MIT $100k competition (the recently renamed version of the $50k competition):

The electricity generated by silicon-based solar cells is directly proportional to the amount of light they absorb (C. Kittel, 1986). However, silicon displays low absorption within the critical near-infrared region of the solar spectrum. As a result, one can either make a thin cell that absorbs very little of this light, decreasing efficiency, or a thick cell that absorbs more of this light, but requires more raw material (driving up costs).

We plan to build an R&D oriented firm that sells intellectual property which improves the efficiency of existing silicon-based solar cells. Its revenues will be based on licensing our IP to existing solar cell manufacturers. The IP includes device technology and expertise in the implementation of the device technology into standard solar cell manufacturing processes. We have performed a detailed calculation which shows that our current device technology can improve the efficiency of solar power generation by up to a factor of five in some parts of the near-infrared spectrum, and at least 20% overall compared to those cells which lack photonic crystals. Alternatively, we can use this improved efficiency to decrease the thickness of the solar cell dramatically, cutting down the amount of silicon used by as much as a factor of ten (cutting the biggest component of materials costs, silicon, by 90%), while retaining a similar level of efficiency as seen in current technology. This grating is projected to integrate well with preexisting photolithographic processing steps and should only lead to small increases in overall processing costs.

solar cells

One idea I've been researching in the last year is using photonic crystals to improve solar cells. The biggest problem with solar cells right now is that they're too expensive compared to other sources of energy, even if they work perfectly for 20 years. The root of this problem is a lack of solar cell efficiency. In silicon solar cells made by the industry, only 12-16% of light coming from the sun is converted into electricity; in laboratory silicon cells, about 30% efficiency can be achieved.

Our idea is to introduce photonic crystals into the solar cells to improve the light trapping properties. This takes place via two mechanisms. First, the photonic crystal diffracts incoming light into modes that are internally reflected within the silicon, thereby converted the solar cell into a waveguide. Second, the photonic crystal refracts light into crystal modes that are confined to bounce back and forth inside the photonic crystal until they're absorbed. When you combine these two effects, we estimate that you could improve the efficiency for 20% for normal incidence. For off-angle incidence, we're currently investigating what should happen. We've already filed for a patent and have started a business idea related to licensing this technology. In the next posts I'll go into more detail about the details of the light trapping as well as the business model.

welcome

This is the first entry for my blog on photonic crystals. My name is Peter Bermel, and I'm a grad student in the physics department at MIT. In this blog, I plan to cover issues arising from my research in photonic crystals under John Joannopoulos, and also other issues of interest in physics (particularly optical or quantum physics). To give you an idea of the research I've done so far, have a look at my publications page.

In future installments of this blog, I will plan on explaining individual projects and their development in detail, while avoiding technical information or jargon as much as possible.

Also, I welcome participation from my readers. Feel free to post comments or questions and I'll do my best to answer them!