No wonder people who worry about global warming are excited about this breakthrough - but they should get excited slowly. The question was never if you could create a complex hydrocarbon like petrol from just air and water, but how much it costs to do it, compared to just pumping oil out of the ground and refining it.
The answer in the past has been far too much. Splitting water molecules to get hydrogen is expensive in terms of the electricity required. Carbon dioxide is easily available as the by-product of burning coal or oil, but using that CO2 as the feedstock for artificial petrol only postpones the moment when it gets into the atmosphere by a few days or weeks.
If you want a truly carbon-neutral fuel, then the carbon dioxide must come straight from the air. Prototype machines have been built (by Klaus Lackner of Columbia University and David Keith of the University of Calgary) that can extract CO2 from the air in industrial quantities, but the price per tonne at the moment is about $600.
But Lackner points out the cost of any new technology plunges steeply once it goes into volume production. And the cost of getting hydrogen from water may also drop dramatically. Daniel Nocera of the Massachusetts Institute of Technology has developed a catalyst from cobalt and phosphorus that can split water at room temperature.
"As long as you can charge the surface, you can create the catalyst, and it doesn't get any cheaper than that," Nocera said in 2008. So if the hydrogen is cheap, and the cost of extracting carbon dioxide from the air also falls dramatically, how much would it cost to combine them into petrol?
That's what Air Fuel Synthesis is working on: a scalable industrial process that takes CO2 from the air and hydrogen from water, combines them into methanol, then turns that into petrol.
Peter Harrison is cagey about his production cost per litre: at the "proof-of-principle" stage, everything costs a fortune. But as he said recently: "You're in a marketplace where the only way is up for the price of fossil fuel. At some point there will be a crossover where our fuel becomes cheaper."
David Keith sees it the same way. "You're selling this fuel, and they're burning it, putting carbon in the air, but then you're recapturing the same amount of carbon and selling it to them again. That's a business model that could conceivably take a whack at the global transportation market, the hardest part of the climate problem to attack."
Maybe Harrison's process won't win the race to capture that market. Maybe the cheaper option will be to grow green algae in waste water or salt water, crush it to extract the oil from it, and then refine the oil into petrol, diesel and so on. (Exxon-Mobil is currently spending about $100 million a year to develop that process.) But one way or another, the petrol we put in our vehicles in 25 years probably won't come out of the ground.
An entire industry employing millions of people, and the national budgets of entire countries, and much of the military planning by the world's great powers, all rest on the assumption this will never happen. Of course, it will. The pressure to cut greenhouse gas emissions will grow as the temperature rises, and the desire for "energy independence" will only get stronger as oil prices rise.
Back in the 1890s, it was still unclear whether the new horseless carriages would ultimately be powered mainly by petrol, steam or electricity. But it was already clear to those with any understanding of the interactions between markets and technology that the day of the horse and buggy was over, and the smart money was already getting out of buggy whips.
