A very hot area in solar R&D is screening-engineered field-effect photovoltaics (SFPV) – it has the potential to enable low-cost, high-efficiency solar cells to be made from virtually any semiconductor material, according to researchers from the US DOE's Berkeley Lab and the University of California (UC) Berkeley.
While still in its infancy, SFPV technology opens the door to the use of plentiful, relatively inexpensive semiconductors, such as promising metal oxides, sulfides and phosphides, according to research conducted by physicist Alex Zettl and colleague Feng Wang.
There are many kinds of semiconductors that absorb light effectively, explains Will Regan, part of the Zettl Research Group and coauthor of a paper on SFPV by Zettl and Wang. “The tricky part is being able to pull out those excited electrons,” he says.
People in the industry have been doing this by creating what is known as a p/n junction. “Two ways that industry typically makes these are to either take two different materials, one with an absence of electrons (p type) and one with excess electrons (n type) and stick them together, or take silicon and change the carrier type using chemical doping,” explains Regan. “Doping works pretty well, but the problem is that chemical doping does not work for all materials.”
Zettle, Wang and Regan have found a third way to make a p/n junction that eliminates this problem. “We do not need chemical dopants and we do not need perfectly compatible p-and n-doped materials which match nicely to each other. We just use a gate to apply an electric field that moves charges around to create a p or n region,” Regan says.
Under the SFPV system, he explains, the architecture of the top electrode is structured so at least one of the electrode's dimensions is confined. In one configuration, working with copper oxide, the Berkeley researchers shaped the electrode contact into narrow fingers. In another, working with silicon, they made the top contact ultra-thin (single layer graphene) across the surface. With sufficiently narrow fingers, the gate field creates a low electrical resistance inversion layer between the fingers and a potential barrier beneath them. A uniformly thin top contact allows gate fields to penetrate and deplete/invert the underlying semiconductor. The result in both configurations, he adds, is high quality p/n junctions.
“We developed a theoretical model that allows us to put in any material properties that you want and any semiconductor and it will tell us how well it will perform to a decent approximation.” With that model, the research team has designed prototypes with silicon, the most commonly used photovoltaic material.
Significantly, the Berkeley team has also designed prototypes using cuprous oxide – a material nobody else is using at present because it is not compatible with any current technology. “Now we are trying to figure out what the other most promising materials are, both current and those not yet in use, to make the highest efficiency and lowest cost solar cells,” Regan says.
“My specific research group and the one we partner with, work on carbon nano materials and graphene. We have been working on using graphene in solar cells just as an electrode material because it is very transparent and will let the light into the solar cell,” he says. “We have found that we can actually improve the junction between the graphene and the silicon by applying that electrical field because the graphene is so thin and has so few charge carriers.”
Regan says the first goal for SFPV is to use it to further improve the performance of existing high efficiency solar cells. Long-term goals are to find new materials that are cheap, abundant, perform well and will last the required lifetime of a solar cell.
Another intriguing near-term prospect being worked on by NREL is a SolarWindow that opens up a new niche market for organic solar cells (OSC). The work is being conducted Creative Research & Development Agreement (CRADA) with New Energy Technology Inc (NET).
Dr. Scott Hammond was doing his post doctorate at NREL, working with OSC technology, when he was hired by NET as a principal scientist. “Industry opinion is that organic solar cells are a low-lifetime technology largely because the traditional organic structure uses low-work function metals that are easily oxidised like calcium and aluminum,” he says. “Our inverted organic solar cell technology is much more stable because the metal materials used are not low-work function so they are not prone to oxidation.”
As he notes: “OSC is a technology that will not go anywhere if it can't last.” According to Hammond, in adopting the inverted architecture approach, “the lifetime can be tens of thousands of hours”.
If that proves to be, the market is potentially huge, says John Conklin, President & CEO of NET. “New commercial building construction is the prime market for this because most tall commercial buildings do not have the roof space for standard PV and yet all have many windows that can act as power generation units,” he says. The refurb market will also be important. “They take out the old window and just slip in the solar window into the same casing for retrofit.”
Hammond is in the process of fine tuning NET's mini-modules in preparation for a battery of tests. “We need to look at the integrity and degradation testing before we go into real time prototype testing on an existing buildings,” Conklin notes.
Check out our other Solar PV articles here.
In part 1. George Marsh: introduction to PV research
In part 2. George Marsh looks at some other novel technologies under development.
In part 3. Joyce Laird takes up the story, talking to some of the leaders of global solar power R&D
In part 4. Joyce Laird looks at 2 innovative areas of PV technology - Screening Engineered Field Effect PV, and Inverted Organic Solar Cells...
About: Joyce Laird has an extensive background writing about the electronics industry; semiconductor development, R&D, wafer/foundry/IP and device integration into high density circuit designs.