Michael, tell us about the Filler Research Group at the Georgia Institute of Technology.  How did it get organized and how did it get started?

We’ve been in existence for about a year now.  I received my training at Stanford and Cal Tech before moving to Georgia Tech.  We work on a number of projects that are related to both near-term and very long-term solutions to different challenges in solar energy conversion.  More specifically, we use a variety of experimental techniques to probe materials and architectures that will hopefully enable highly efficient and extremely low cost devices.

When we first met you gave me the “dummy’s guide” to photovoltaics.  For the benefit of our readers: how exactly does a photovoltaic cell work?

Before I answer let me point out that photovoltaics or “PVs” are not the only way to convert sunlight energy into useful energy that we can us to power a lightbulb or car.  There’s also -- and I’ll talk about these other types briefly before I talk about PVs – so-called concentrating solar power.  Concentrating solar power is very similar to when you were a kid and you took a magnifying glass and you tried to burn a leaf (or if you were really dangerous you tried to burn an ant). 

(Laughing).  You must have had some childhood!

(Laughing).  Well . . . by taking big lenses and focusing sunlight onto a heat transfer fluid we can then use that fluid at a very high temperature to make steam and drive a turbine.  So it’s very similar to what happens in a coal fired power plant but instead of burning coal to create the heat and form the steam we just use heat from the sun.  Concentrating solar power is only possible in areas where there’s a very large amount of sunlight per unit area per day.  This varies from place to place in the world.

And then of course there are other technologies that people don’t often associate with the sun but they are definitely driven by the sun.  One example is wind power.  The wind is driven by heat gradients in the atmosphere which are in part formed from the sun.  And things like biomass -- if you are growing a plant that you’re going to convert into a fuel, it’s getting its energy from the sun and then converting water and carbon dioxide into chemical bonds that you will subsequently use as a fuel.  So there are a lot of “solar” technologies, but photovoltaic is the type of solar power on which my group works.  “Photovoltaic” simply means “photo” (as in photons from the sun) and “voltaic” as in “voltage”.  PV power creates voltage (or electricity) from photons.  Traditionally PV has been accomplished by using a semiconductor material -- the classic example is silicon.  For those who are familiar with the microelectronics industry, silicon is the force behind the rapid growth in that industry over the last 50 years and our ability to incorporate 2 billion [or more] transistors on a single chip the size of your thumbnail.  But that same material, silicon, can also absorb sunlight.  What happens when you put a piece of silicon out in the sun, photons from the sun penetrate deep into the material.  At some point they are absorbed by the material and that absorption process leads to excited high energy electrons.  We then extract this energy from the electrons and we drive it through an external circuit, creating electric power.  At the most basic level, that’s what’s happening in a solar cell.

Of the various technologies you see on the horizon for photovoltaics which do you think are the most promising in the near term for commercial applications?

That’s an excellent question.  I think up until 5 years ago, silicon technologies were the most promising.  Silicon can come in a number of different forms.  It can be either a single crystal where every atom is in a very specific place on a lattice and that allows you to make very efficient devices.  But it’s also very expensive to make that single crystal material.  There are also polycrystalline silicon devices and they are a little bit easier to make but also a little less efficient.  They are not as completely crystalline as the other type (single crystal).  And then there’s so called amorphous silicon where there is no long-range structure.  There are still silicon atoms that enable light absorption, but these are even less efficient.  But potentially you could make these devices very cheaply.

There’s a tradeoff between efficiency and cost.  In an ideal world you would not want to play this game.  You would hope that you could increase the efficiency and bring down the cost.  Up to now, it’s been very challenging to find a way to do that.  So the technologies that I see in the near term are the ones that are already available commercially.  Things like silicon technologies -- you can go to your local distributor, buy solar panels, and build an array on your roof.  Or you can find investors and build a larger scale power plant.

Some new technologies that have been commercialized within the last two to three years are the so called thin film technologies.  Actually, amorphous silicon is part of the thin film category.  But the two technologies you hear a lot about these days are so called cadmium telluride and CIGS which stands for copper indium gallium selenide.  CIGS is simply a complex semiconductor made up of four different atoms. 

I can’t wait until my secretary tries to type that up. (Laughter)

Both of these materials offer reasonable efficiencies at pretty low production costs.  Some people have found ways to manufacture these types of devices on high speed rollers like you would manufacture newspapers.  That’s very different from the traditional way to make solar cells.

How would you use a thin film PV cell?  When I picture rooftop solar I think of a device that that is maybe 3 or 4 inches thick and is mounted on the roof with metal racks.  How would you use a thin film application?

Ok that’s a great question.  What you’re seeing on someone’s roof when you look at a traditional solar cell is mostly the framing.  The actual solar cell (even the old traditional ones) are only about a couple of hundred microns thick.  The actual device that does the absorption of light is only a couple times the thickness of your hair.  All of the rest of what you’re seeing is stuff to connect the solar cells to produce enough voltage to power devices in your home.  One key metric to remember is that a small silicon solar cell, measuring 6 inches by 6 inches only produces about 1 volt of voltage (actually closer to 0.7V).  To make usable power we need to string together a number of these cells to make a panel such that in the end the voltage is at least 120 volts so you can power a circuit in your home.  A nice thing about thin films is potentially the ability to put them on flexible substrates.  Instead of seeing these big panels what a lot of people hope will happen is that we can start making shingles out of these materials.  Your roof would look predominately the same -- you would still see the patchwork of dark blocks that are now made from tar and oil.  But you could just convert them into photoactive materials to absorb sunlight. 

I will say long term one of the things that we have to realize is that if you took the entire area of all the rooftops of every structure in the United States you would only get between one-eighth to  one-tenth the power currently being consumed in the U.S.  There is always going to be a need for large scale power plants out in the middle of nowhere to augment and supplement what we can generate on our own rooftops.

That’s a remarkable metric.  Is it based upon the efficiencies we achieve with current technologies?

It’s important to note that I said “power consumption” and not “electricity consumption.”  So it’s not just electricity.  It would be all the power that we use in the US.  But, we’re not likely to get all our power in the US from PV devices.

When you say all the “power” used in the US as distinct from “electricity;” What other forms of power are you including beyond electricity?