From left: Top view SEM micrograph of a tungsten inverse colloidal crystal annealed to 1,000ºC for 12 hours; unprotected tungsten significantly degrades after being heated to 1200ºC; ceramic-coated tungsten retains structural integrity after heating at 1400ºC for an hour. (Credit: Kevin Arpin via Stanford University.)
This is a tale of two materials, either of which could be game-changers in the search for high-efficiency photovoltaics.
Researchers studying solar energy conversion have known for a long time that the efficiency of single-junction photovoltaics is limited simply because they do not capture and convert a broad spectrum of incident light. Energy conversion is most efficient when the incident radiation is similar to the electronic band gap energy of the photovoltaic material. Low-energy photons do not get absorbed, and photons with energies that are too high, including much of the visible spectrum, are wasted as heat.
The problem can be solved by developing a more efficient device or by changing the light itself. The former solution involves “multijunction” photovoltaic materials, which includes the III-V family of semiconducting materials, some organic molecules, and silicon. They do, in fact, convert energy more efficiently than single-junction PVs, however, the “price to pay” is higher cost.
The latter solution involves structuring a device that captures photons across a broad spectrum of energies and converting them to a narrow range of energies that match up well with the band gaps of single-junction PVs. Such a device, known as a solar thermophotovoltaic, couples two components—a broadband absorber and a “spectrally selective” thermal emitter.
According to a new paper in Nature Communications, engineering of broadband absorption of solar radiation is well understood. In contrast, engineering for spectral emissivity is not well understood, because emissivity depends on the intrinsic optical properties of the emitter material. In order to be practical, the emitter needs to operate at 1,000˚C or higher. Even small gains in operating temperature can lead to big increases in efficiency because power output scales as T4.
Besides thermal stability, the emitter area needs to be large for two reasons—power output also scales with emitter area and, the paper says, “fundamental thermodynamics require an emitter that has a substantially larger area as compared with the absorber.”
Photonic crystals with structured 3D architectures can modify photons by enhancing or suppressing certain parts of the spectrum. Tungsten is refractory enough to be stable above 1,000°C and has the right optical properties to work as a photonic crystal. Paul Braun’s group at the University of Illinois at Urbana-Champaign fabricated a 3D, mesoscale tungsten photonic crystal by atomic layer deposition on a self-assembled template of colloidal silica spheres.
The tungsten photonic crystals proved thermally stable after annealing at 1,000ºC for 12 hours without cracking, a result the paper describes as “unprecedented.” Looking for more output by virtue of higher temperatures, the researchers found that the material does not survive beyond 1200ºC because the tungsten sinters and the silica softens. Depositing a protective coating of hafnium dioxide just 20 nm thick added enough protection to enable the tungsten crystal to survive temperatures up to 1,400ºC. According to a press release, Stanford researchers confirmed that the coated emitters did produce the infrared light waves preferred for solar cells.
Wondering whether higher power outputs might be possible if emitters could be run at higher operating temperatures, the group decided to look at other highly refractory materials.
“Borides, nitrides, and carbides have metal-like optical properties, so we looked at hafnium diboride as a substitute for tungsten,” says Kevin Arpin in a phone interview. The paper is based on Arpin’s doctoral research in Braun’s group. These materials are also well known for their thermal stability.
As a photonic crystal emitter, the hafnium diboride performed “better than anything [nonmetallic] published before, but it is not as good as the tungsten—yet,” Arpin says. According to the paper, the HfB2 crystals oxidize above 1,000˚C. However, the paper suggests that better control of oxygen partial pressure during annealing, protective coatings, or special additives could reduce oxidation susceptibility.
Regarding the photonic crystals, Arpin says, “Having these structures survive high temperatures with the fine features intact is a key finding.”
The paper is “Three-dimensional self-assembled photonic crystals with high temperature stability for thermal emission modification,” Nature Communications (DOI: 10.1038/ncomms3630)