It's not especially efficient, but we haven't had time to optimize the reaction.

The carbon dioxide we're currently dumping into the atmosphere started out as atmospheric carbon dioxide hundreds of millions of years ago. It took lots of plants and millions of years of geological activity to convert it to fossil fuels. One obvious way of dealing with our atmospheric carbon is to shorten that cycle, finding a way to quickly convert carbon dioxide into a usable fuel.

Unfortunately, carbon dioxide is a very stable molecule, so it takes a lot of energy to split it. Most reactions that do so end up producing carbon monoxide, which is more reactive and a useful starting material, but it's far from a fuel. Now, though, researchers have discovered a catalyst that, with a little help from light, can take CO2 and make methane, the primary fuel in natural gas. While the reaction is slow and inefficient, there are a number of ways it could be optimized.

Unexpected methane

The work started out with a catalyst that converts carbon dioxide to carbon monoxide when supplied with a source of electrons. The catalyst is a complex ring of carbon-based molecules that latch on to an iron atom at the center. The iron interacts with carbon dioxide, allowing hydrogen atoms from water to break one of the carbon-oxygen bonds, liberating water. The iron loses some electrons in the process, and these have to be re-supplied for the cycle to start again. Typically, that supply comes in the form of a separate chemical that readily gives up some electrons.

The team involved in the new work appears to have been looking to optimize this reaction. Their efforts focused on the reaction that transfers electrons to the iron, so they tried to enhance this process using light and experimented with different chemical electron sources. The reaction sped up, but carbon monoxide was no longer the only product. Instead, the reaction also liberated some hydrogen and methane.

This was surprising enough that the team used a specific isotope of carbon to confirm that the methane that was being produced got its carbon from carbon dioxide. The researchers seem to have stumbled onto a methane-producing reaction almost by accident.

The majority of the carbon that reacted still ended up as carbon monoxide (82 percent vs. 18 percent for methane). And the reaction was pretty slow, producing only about 12 grams of methane an hour for each gram of catalyst. But the researchers noticed that methane production didn't even start until significant amounts of carbon monoxide had been produced first. So they suspected that the methane-producing reaction worked with carbon monoxide and not directly from CO2. The catalyst was accelerating both reactions separately rather than accelerating one reaction that could go down different pathways.

To figure out whether this was the case, the team supplied the catalyst with carbon monoxide instead of CO2. Methane production went up dramatically: 83 percent of the reaction products were methane, and the remaining 17 percent was hydrogen. Now, each gram of catalyst was producing about 30 grams of methane an hour.

Room for improvement

One thing that's clear is that this is an utterly abysmal way of using light energy, with a quantum yield of about 0.18 percent, meaning that only a tiny fraction of the photons supplied are used as energy to produce methane. By contrast, a lot of catalysts that split water using light produce hydrogen with a quantum efficiency over five percent.

But it's worth noting what the light is used for: inducing the electron donor to hand the electron over to the iron at the center of it all. While the catalyst itself is stable and remained active for days in these tests, the reaction can't go forward without a steady supply of electrons delivered to the iron.

That's a key step you would want to optimize anyway. Ideally, it would be much simpler to feed electrons to the iron rather than constantly refresh the supply of a chemical. And it might be possible to chemically link the ring system that holds the iron in place to something that would supply it with electrons that would then get passed to the iron. If that can be arranged, the low-efficiency use of light wouldn't matter, as you could hook the whole system up to an efficient solar panel.

The authors also suggest setting the whole thing up as a two-stage process: one to produce carbon monoxide, the second to convert it to methane. Hydrogen would also be a useful byproduct that can be harvested from the second stage.

Overall, the work isn't about developing a technology that's ready to use. Instead, it's more about developing an option for further development. And when it comes to moving off fossil fuels and even potentially pulling some carbon dioxide back out of the atmosphere, it's nice to have options.