A perovskite and AlGaN hydrogen economy?

Much research is being done on photovoltaic solar power generation. Silicon, amorphous silicon, thin-film cadmium telluride, thin-film copper indium gallium selenide (CIGS), organic semiconductor, perovskite, dye-sensitized, and others -- so may ways to convert sunlight to electrical power at varying efficiency. Given enough R&D and low enough costs, photovoltaics could form an entire new solar economy, right?

John Wallace 720

Much research is being done on photovoltaic solar power generation. Silicon, amorphous silicon, thin-film cadmium telluride, thin-film copper indium gallium selenide (CIGS), organic semiconductor, perovskite, dye-sensitized, and others -- so may ways to convert sunlight to electrical power at varying efficiency. Given enough R&D and low enough costs, photovoltaics could form an entire new solar economy, right?

Well, we all know it’s not that simple, based on two complicating factors -- night, and bad weather. Unlike fossil-fuel and nuclear plants, sunlight can’t provide continuous power. The electrical grid, as is, can tolerate a certain percentage of its power being generated by solar cells, absorbing the effects of the fluctuations. However, a true solar economy based simply on photovoltaics would not work even with a massive overhaul of our electrical grid.

Although one approach is to try to make the grid itself somehow more resilient, what we really need is a reasonable form of solar-energy storage. Some proposals for storing solar electricity include using massive banks of the same lithium-ion batteries that are used in Tesla electric cars; spinning up huge arrays of high-speed flywheels in vacuum chambers; or pumping water up into lakes high in the mountains for hydroelectric generation. Alternatively, solar heat can be held in reserve by melting salt compounds and storing the heated liquid in tanks for subsequent steam-electric generation.

But we like fuel
The approach that seems to make most sense to me is one that fits best with our existing we-want-power-when-we-want-it economic system: solar hydrogen generation. And while various ways are being developed to use electricity from photovoltaics for hydrolysis of water into hydrogen (and oxygen), I am a fan of more-direct methods to generate solar-based hydrogen – especially, of course, those using photonics technology.

For example, Laser Focus World senior editor Gail Overton recently wrote about research being done at École polytechnique fédérale de Lausanne (EPFL; Lausanne, Switzerland), in which the material perovskite is used, not for photovoltaic electric generation, but instead to produce hydrogen directly from sunlight and water. Even at this early stage, the conversion efficiency from solar energy to hydrogen is 12.3%.

Perov Solar

Hydrogen bubbles escape from perovskite electrodes when the solar cells are exposed to light. (Image: EPFL)

And researchers at National Cheng Kung University (Tainan City, Taiwan) and National Taiwan University (Taipei, Taiwan) have created photoelectrochemical cells that use dodecagon-faceted aluminum gallium nitride (AlGaN/n-GaN) heterostructure electrodes for the same purpose.1 The use of AlGaN/n-GaN rather than n-GaN for the electrodes boosts photocurrent density by a factor of 5.9.

Of course, the production of hydrogen via sunlight does not preclude the use of conventional photovoltaics; it would simply provide the stability that would make a substantial dependence on solar energy possible (if that ever does happen). And just like oil and natural gas, hydrogen can be stored, shipped, and used to fuel cars and trucks.

There are huge impediments: the low energy density of hydrogen, its high diffusivity (meaning tanks and pipes must be well-sealed), the high cost of accessory technologies like fuel cells, the enormous cost of developing infrastructure, safety questions, and the fact that we don’t know if we can make such a system efficient enough for real use.

Well, that’s what research is for.

REFERENCE:

1. W. C. Lai et al., Optics Express (2014); doi:10.1364/OE.22.0A185

More in Commentary