Tunable bandgaps, lattice matching and the making of high-efficiency CPV cells: an interview with Solar Junction Co-founder Craig Stauffer

Craig Stauffer
Craig Stauffer

Craig Stauffer is a co-founder of Solar Junction, which on February 15th 2011 announced that it has achieved 41.4% efficiency on a production-sized concentrating solar photovoltaic (CPV) cell, as verified by the U.S. Department of Energy's National Renewable Energy Laboratories. At Solar Junction, Mr. Stauffer directs strategic initiatives, and previously served as the Company's CFO.

Mr. Stauffer's industry background spans positions in technology, marketing and sales, operations, finance, and corporate management as the former CEO of Genesis, including five years in mergers and acquisitions. He also held management positions with GCA, Bendix, Plessey, General Signal and Applied Materials.


Solar Server: Congratulations on your 41.4% efficient production-sized cell. What are the main elements of the cell design that allowed Solar Junction to reach 41.4% efficiency at a production scale?

Craig Stauffer: A multi-junction solar cell is essentially a number of electrically interconnected sub-cells, each of which take a portion of the solar spectra and convert it. So instead of having a single bandgap cell that converts in a mediocre manner, you have three different bandgap cells each which are more optimized to the available sunlight and a portion of it.

The current or incumbent cell providers use a cell developed for space that has a germanium bottom junction at about .7 eV, and a middle junction about 1.4 and an upper about 1.9. That's not idealized to the spectrum on earth, it’s idealized to the spectrum in space. If you wanted to idealize it to the spectrum on earth, that .7 would be swapped out for a 1. And that would give you four or five percent more real efficiency in the cell.

The problem is that in order to do that, they have to leave their design paradigm where the layers are lattice-matched; the inter-atomic spacing, layer-to-layer is exactly the same. You leave that paradigm, you create efficiency and reliability problems.

Solar Junction's technology allows the ability to stay in that lattice-matched paradigm and create that 1 eV junction that throws another real five percent on the efficiency. So we have more productively sliced up the solar spectrum in our three sub-cells.


Solar Server: Without getting too technical, can you describe material bandgap tunability for our readers? How was this developed?

Craig Stauffer: This came out of a development at Stanford University. And so we just talked about each of those sub-cells having a particular bandwidth. And those bandwidths were chosen based on an idealized or typical solar spectra on earth. And so if that sunlight were the same everywhere and the optics that transferred it were the same everywhere, those are exactly the bandgaps those cells should have for their best efficiency.

Solar Junction CPV wafer
Solar Junction CPV wafer

We also talked about the fact that moving bandgaps around throws you out of the lattice match. Solar Junction has the ability in our A-SLAM layer to adjust the bandgap to be optimized to changes in the solar spectra as a function of the system supplier's optics and as a function of changes in the spectra in different places around the world.

So we can squeeze even more extra efficiency out by matching up our cells' response to the optics and the solar spectra of the supplier and the place on the earth where they put their system - the unique capability within our technology.


Solar Server: On your website and in press statements, Solar Junction emphasizes the potential for CPV to compete with PV. Even with higher efficiencies, given the additional cost of the optics, can CPV deliver a comparative levelized cost of energy? If so, how?

Craig Stauffer: The answer is absolutely equivalent and better. These solar cells are dramatically more efficient than any other solar technology.

Technologies for solar cells range on the low end from maybe 10% up through the best monocrystalline cells made at about 24%. The multi-junction cells we are making now are, of course, right around forty-one and a half percent. Over the next five years that number will rise to 50%.

CPV systems as a system basis also have two axis-tracking, so they are at all times perfectly pointed towards the sun.

The net of that is that you get higher energy harvested throughout the course of the day, and on top of that you get this efficiency adder. Essentially CPV system panels are about 50% more efficient than any other base panel. If you add tracking to them, which is integral to the CPV system cost, you add at least another 50%, in terms of energy harvested.

So you have a dramatically better energy collector and converter in CPV systems. The cells are expensive, but the optics is relatively cheap. So the combination of the optics and the cell make the difference.

In a CPV system, an individual cell might be about 5mm on the side. That is going to take the area collected by an 8" square glass lens. So you are literally reducing from an 8" x 8" area, down to a 5mm square cell, and that cell is going to output on the order of 15 watts.

So you get the collection area, and you get just a huge efficiency and increase in harvest of energy.


Solar Server: Can you tell us a little about the role of both the U.S. Department of Energy (DOE) and your investors in making the commercialization of this technology possible?

Craig Stauffer: We've had a wonderful set of investors, the first investor in this company was New Enterprise Associates, who put us in business. They were subsequently joined by Advanced Technology Ventures and Draper Fisher Jurvetson.

Making high efficiency cells is a capital intensive business, and they've given us the money that has allowed us to put together a development and demonstration line here of the highest quality, and cycle through iterative engineering improvements at the fastest rate of anybody in the CPV cell business.

Once you've done that, the next question is the capital required for expansion, and as I'm sure you are aware, for smaller companies, arranging debt, or borrowing money, is not necessarily easy, particularly smaller emerging technology companies.

The DOE created this loan guarantee program to help with that difficulty. In this process, the DOE, after scrupulously diligence-ing the company, makes available to the company a loan guarantee so that the company can afford reasonably priced capital at an attractive debt-to-equity ratio, that frankly likely would not otherwise might not be available.

So it's just an outstanding program. It creates American jobs, it enhances American technology, and it helps solve the clean energy dilemma.


Solar Server: Anything else about the cells, Solar Junction's process or your investors that you feel like sharing with our readers?

Craig Stauffer: We've just been a very fortunate company to have an extremely supportive investor base, who has shown the willingness to supply the type of money required to make these breakthroughs, and to set the company up on a path that has another decade of technology improvement the whole way through 50% or more cell efficiency.

We have an executable, and that's just a very gratifying situation.


Interview conducted by Solar Server International Correspondent Christian Roselund

on February 18th, 2011