Efficiencies of current solar cells on the market top out around 20%. Thus, researchers at Kyoto University in Japan are exploring optical technologies to improve energy production converting sunlight. The result could be a game-changing solution.
Higher temperatures emit light at shorter wavelengths, offering more energy, which is why scientists aim to achieve short wavelengths in the design of solar cells. There is a problem, however: Heat dissipates light of all wavelengths, but a solar cell will only work in a narrow range.
A new semiconductor
“Our devise can solve the mismatch between the spectrum of the sunlight and the spectrum of the light that can be efficiently converted to electric power by conventional photovoltaic cells,” explains quantum optoelectronics expert Takashi Asano, associate professor in the Graduate School of Engineering at Kyoto University.
A game-changing solution for better solar cells
Commercially available single-junction photovoltaic cells have a conversion efficiency of about 20%, which is largely determined by the aforementioned spectral mismatch.
The part of the sun’s spectrum where wavelengths are longer than the band gap of a solar cell’s semiconductor are not absorbed, and that energy is lost. The part where wavelengths are shorter than the semiconductor’s band gap wavelength is lost in form of heat. “To overcome this issue, the concept of a thermophotovoltaic power generation system was proposed by RM Swanson more than 35 years ago,” Asano notes. This system uses concentrated mirrors focusing on a thermophotovoltaic (TPV) converter, where the concentrated sunlight heats a refractory radiator. A solar cell then coverts that incandescent radiation into electricity. “However, it has been difficult to develop a thermal emitter whose emission spectrum is well matched to a photovoltaic cell,” Asano says, adding that the “usual materials” show broad thermal emission, known as the black body radiation, when their temperature is raised. Nanostructured refractory metals have also been studied, but the professor says it is difficult to suppress longer wavelength emission components generated from free electrons in metals.
“We have developed emitters based on nanostructured intrinsic silicon, which showed a strong emission peak at near-infrared-to-visible wavelengths and negligible emission at longer wavelengths,” reports Asano. “Our emitter can convert 59% of the input power — sunlight or other heat sources — into emission of wavelengths shorter than 1100 nm, theoretically. Another emitter of a different size can convert 80% of the input power into emission of wavelengths shorter than 1800 nm. These emissions can be efficiently converted to electric power by using silicon and gallium antimonide photovoltaic cells, respectively.”
A projected increase of at least 40%
In reference to his team’s eventual goal for the efficiency improvement this breakthrough technology will enable, the professor tells, “We have demonstrated that our emitters can concentrate input power — from sunlight or other heat sources — into the emissions suitable for photovoltaic cells, as I mentioned.” While Asano and colleagues have not yet carried out experiments to show the actual improvement in a solar cell’s conversion efficiency, he says he does expect a value of 40%.
Achieving selective thermal emission
Asano’s team has achieved near-infrared-to-visible selective thermal emission by controlling — or designing — both electronic states and optical states of the emitter. “First, we selected intrinsic silicon as material of the emitter because above-band-gap electronic transition of silicon emits near-infrared-to-visible light when thermally excited at high temperature, while intrinsic silicon is transparent at longer wavelengths because densities of free-electrons and holes are small and the emission at longer wavelength is expected to be weak,” he explains. “Second, we formed an array of rod-type resonators with silicon to enhance the emission at short wavelengths by resonating effect and to suppress emission at the longer wavelengths by reducing the volume of silicon — compared with, for example, simple slab and thick wafers.”
By optimizing these two controlling methods, the Kyoto researchers have achieved the results that could not be achieved with the previous strategy, which utilized refractory metals or highly doped silicon, according to the expert. “It is also noteworthy that the melting temperature of silicon is as high as 1400 degrees Celsius, so we could operate our device at 1000 degrees Celsius,” he adds.
Moving forward with this research work, the team now has to increase the size of the emitters for photovoltaic applications. “Also, we have to increase the heat resistance of the emitters because higher operating temperature leads to higher efficiency and a higher power density,” Asano concludes.
The paper "Near-infrared-to-visible highly selective thermal emitters based on an intrinsic semiconductor” was published in Science Advances.
Written by Sandra Henderson, research editor Solar Novus Today