The ultimate goal of solar cell technology is to be able to generate electricity at costs lower than sources such as coal, natural gas and nuclear. Key to this is continuing improvements in conversion efficiency, and with the development of the first four-junction, four-terminal stacked solar cell produced using a micro transfer printing process, researchers have taken another step towards this goal by achieving efficiencies of up to 43.9 percent, with the possibility of exceeding 50 percent in the near future.
The multilayer, microscale solar cell was created by North Carolina-based Semprius Inc. and California-based Solar Junction, working in collaboration with a team from the University of Illinois at Urbana-Champaign led by Professor John Rogers. Using Semprius' proprietary high-speed micro transfer printing process high that it says is able to simultaneously produce thousands of stacked microcells with very high yields, the team stacked a triple-junction microcell on top of a single-junction microcell to create a four-junction, four-terminal stacked solar cell.
The team says the use of four junctions allows the stacked cell to capture a broader range of the solar spectrum, while the use of four terminals rather than the standard two increases the yield of the solar cell under normal operation in the field. Additionally, a new interfacial material that is placed between the top and bottom cells helps to optimize overall efficiency by minimizing optical losses within the stack.
The triple-junction solar cell is covered in an anti-reflective coating to ensure efficient transmission of light to the uppermost layers, while the bottom cell is a single-junction germanium microcell. Light with wavelengths between 300 and 1,300 nm is captured by the triple-junction cell, while light with wavelengths from 1,300 to 1,700 nm passes through to the germanium cell. The result is a multilayer, microscale solar cell that the team says outperforms conventional silicon and single-junction solar cell in terms of efficiency.
"The strategy involves high-speed, printing-based manipulation of thin, microscale solar cells and new interface materials to bond them into multilayer stacks,” Rogers said. "Quadruple-junction, four-terminal solar cells that we can build in this way have individually measured efficiencies of 43.9 percent."
However, Semprius, which was co-founded by Rogers in 2006 to commercialize ultrahigh efficiency photovoltaic modules, promises that the same process will be capable of achieving efficiencies exceeding 50 percent in the near future.
Like the 44.4 percent efficiency record reported by Sharp last year, the 43.9 percent efficiency figure was achieved using a lens system to concentrate light onto the solar cells. Using a dual-stage optics system, the incident sunlight was focused more than 1,000 times. Modules created from the microscale solar cells also achieved efficiencies of 36.5 percent under the same conditions.
"This is very nice work," stated Ali Javey, a professor of electrical engineering and computer sciences at the University of California, Berkeley, who wasn't involved in the research. "The results are impressive, and the schemes appear to provide a route to ultra-high efficiency photovoltaics, with strong potential for utility-scale power generation."
Rogers is first author of a paper detailing the new solar cell that is published this week in the journal Nature Materials.
Sources: Semprius, University of Illinois at Urbana-Champaign
Cool optical illusion in the second photo, though. When you smooth scroll down it shifts to the right a bit and bounces back when you stop.
Someone needs to do a study to figure out how much less heat gets sent back into space from an average roof, as compared to one covered in black solar cells making electricity that ends up as heat...
As for the article: I wish the "four terminal" concept was explained; I don't understand why it would increase the yield.
There are two exiting things here: 1) efficiency gains but perhaps more importantly 2) the high speed printing process. In the real world we'd rather have fast automated production than small gains. You can always add panels later as they improve - plenty of space to put them.
at this point the problem for solar is batteries and the fact that you can't control the weather (clouds) ...