New “solar thermal fuel” has energy density of lead battery

[rhoenix]

Right now, photovoltaic devices are the cheapest, most efficient way to harvest the energy in sunlight. The problem is that this energy ends up in the form of electricity, which we have difficulty storing in a cost-effective manner. An alternative approach, solar thermal energy, converts solar energy to heat and can use that heat to continue generating power for several hours after the Sun goes down. But that's not enough to make solar an around-the-clock energy source.

Researchers are apparently working on a third option, one that could potentially store energy indefinitely. It goes by the name of "solar thermal fuel," but it's not a fuel in the traditional sense. Rather than breaking apart the fuel molecule through combustion, solar thermal fuels release heat by rearranging bonds within a molecule, leaving all the atoms in place. As a result, they can be recycled repeatedly—in the example that introduced me to solar thermal fuels, a research team ran theirs through more than 2,000 cycles with no loss in performance.

How do you get energy into and out of a molecule without breaking any bonds? In this case, the authors worked with derivatives of a chemical called azobenzene, shown below. The double bond between the two nitrogens forces the remaining bonds into one of two forms: either both of the rings can be on opposite sides of the molecule (top, called the "trans" form) or they can be on the same side (bottom, called "cis").



These two forms, called isomers, are like different energy states of the same molecule. The trans form can be considered the ground state; exciting it with UV light can flip the bonds, converting it to the cis form. The molecule stays locked in the cis form until some additional energy destabilizes it. It will then release energy in the form of heat as it switches back to the trans version. So individual azobenzene molecules can act a bit like a tiny energy storage system.

On its own, though, azobenzene isn't especially efficient at this. It tends to form a disorganized jumble of molecules, making it hard for individual ones to flex bonds. But a team of researchers at MIT did computer modeling that suggested the performance could be improved by up to 30 percent if they could get the azobenzene molecules lined up and stacked in a specific orientation.

To do that, the researchers chemically linked azobenzene to carbon nanotubes. The process wasn't especially efficient, but they could repeat it several times until the nanotubes had a healthy fuzz of azobenzene on their surfaces. The azobenzene could then be packed quite densely into a solid or kept in solution with a simple solvent like acetone.

The azobenzene still worked. Exposure to UV light would "charge" the system, and mild heating (75°C) in the dark was enough to get the chemical to release that energy in the form of heat. And using carbon nanotubes to pack the molecules in succeeded in boosting the amount of energy that could be stored in the system, even though the nanotubes were non-functional and accounted for nearly half of the bulk weight of the storage medium.

The surprise, however, was how much the nanotubes boosted the performance. Rather than the expected 30 percent increase in energy storage, they saw a boost of 200 percent.

With that large of an increase, the numbers on the system became quite good. The authors found that, given UV light, the energy storage efficiency of the material was about 14 percent. The material could be cycled through charge and release of energy at least 2,000 times without any drop in performance, and it could stably store energy for extended periods of time. The energy density was quite good as well, at 44 Watt-hours/kg, making it similar to a lead-acid battery.

The process the researchers used to link the azobenzene to the nanotubes doesn't result in an even coating, and it was possible to detect areas in the bulk material that stored energy more or less efficiently. So figuring out an optimal density (and figuring the chemistry needed to produce it) is probably a priority.

What isn't clear from this work is the sorts of temperatures that this material can produce—specifically, is the heat concentrated enough to boil water? Can it release enough heat to boil water without starting to decompose? Since heat itself can trigger the reverse reaction, it's also worth considering whether there's a danger of random temperature spikes setting off a chain reaction in the storage medium.

Still, the performance of this material seems to make those issues worth looking into. Having an additional option for storing solar power could provide a great deal of flexibility—collectively, the different forms of solar could keep the Sun's energy flowing around the clock.

As the article states, testing isn't close to determining whether this can, for instance, boil water with the heat stored. Even so though, this looks interesting.