SAN FRANCISCO--A novel fabrication technique developed by a University of Connecticut professor could pave the way for vastly improved solar energy systems, according to researchers.
Scientists have for years have studied the potential benefits of a new branch of solar energy technology that relies on nano-sized antenna arrays that are theoretically capable of harvesting more than 70 percent of the sun’s electromagnetic radiation and simultaneously converting it into usable electric power.
Even the best silicon panels collect only about 20 percent of available solar radiation, and separate mechanisms are needed to convert the stored energy to usable electricity for the commercial power grid. Many believe that the panels’ limited efficiency and development costs have been two of the biggest barriers to the widespread adoption of solar power as a practical replacement for fossil fuels.
While nano-sized antennas have shown promise in theory, scientists have lacked the technology required to construct and test them. The fabrication process is immensely challenging. The nano-antennas—known as "rectennas" because of their ability to both absorb and rectify solar energy from alternating current to direct current—must be capable of operating at the speed of visible light and be built in such a way that their core pair of electrodes is a mere 1 or 2 nanometers apart.
The potential breakthrough lies in a novel fabrication process called selective area atomic layer deposition (ALD) that was developed by Brian Willis, an associate professor of chemical and biomolecular engineering at UConn and the previously director of the university's chemical engineering program. Willis developed the ALD process while teaching at the University of Delaware, and patented the technique in 2011.
"This new technology could get us over the hump and make solar energy cost-competitive with fossil fuels," Willis said, in a statement. "This is brand new technology, a whole new train of thought."
Brian Willis, an associate processor at the University of Connecticut, holds a rectenna device.
Photo credit: Sean Flynn/UConn
In a rectenna device, one of the two interior electrodes must have a sharp tip, similar to the point of a triangle. Using ALD, Willis is able to get the tip of that electrode within one or two nanometers of the opposite electrode. Before ALD, existing lithographic fabrication techniques had been unable to create such a small space within a working electrical diode. The closest scientists could get was about 10 times the required separation.
Through ALD, Willis has shown he is able to precisely coat the tip of the rectenna with layers of individual copper atoms until a gap of about 1.5 nanometers is achieved. The process is self-limiting and stops at 1.5 nanometer separation.
The size of the gap is critical because it creates an ultra-fast tunnel junction between the rectenna’s two electrodes, allowing a maximum transfer of electricity. The nanosized gap gives energized electrons on the rectenna just enough time to tunnel to the opposite electrode before their electrical current reverses and they try to go back. The triangular tip of the rectenna makes it hard for the electrons to reverse direction, thus capturing the energy and rectifying it to a unidirectional current.
Willis and a team of scientists from Penn State Altoona along with SciTech Associates Holdings Inc., a private research and development company based in State College, Penn., recently received a $650,000, three-year grant from the National Science Foundation to fabricate rectennas and search for ways to maximize their performance.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.