Harvard University researchers say they’ve created a half-chemical, half-biological system to generate liquid fuel using air, water, and sunlight. And as if that weren’t enough, it takes climate-changing carbon dioxide out of the atmosphere in the process.

A study published Thursday in the journal Science advances research led by Daniel Nocera, a professor of energy at Harvard who has spent years trying to best nature’s original workhorse technology-photosynthesis. Nocera has been developing what he calls an “artificial leaf.”

Trees and other vegetation act as atmospheric filters, sucking in carbon dioxide and locking it in their trunks and branches (a process that’s suddenly reversed by wildfires). The new system announced Thursday is 10 times more energy efficient than natural organisms, according to the paper.

The artificial leaf looks nothing like a leaf, though. Pamela Silver, a Harvard biochemistry professor who co-authored the paper with Nocera, said visitors to her lab seem disappointed when they realize this. “It’s just a jar with wires coming out of it,” she said. “It looks like science.”

The core technology is really a catalyst, cobalt phosphate, used to perform a neat trick: When placed in water and hit by sunlight, it splits water into its component elements – one part oxygen and two parts hydrogen. That’s what plants, trees, and some bacteria, do for a living. It’s the first step in photosynthesis, the process by which plants take CO2 out of the air, add hydrogen taken from water, and emit oxygen.

What the plants make is carbohydrate: the basic natural purpose of photosynthesis that gives plants the energy to grow. It’s biology’s premiere chemical storage system for solar energy, and, under one of its more common names, the bane of well-fed humans. If coal is “buried sunshine,” sugar is “sweet sunshine.”

Nocera’s team builds on the original leaf concept, first published in 2008, which was limited to just splitting up the oxygen and hydrogen atoms of water. In their new work, the team mirrors the making of carbohydrates. They take the hydrogen split off by the artificial leaf and use it to grow an artificial microbe. Ralstonia eutropha breathes in the hydrogen and eats CO2 out of the air, growing and reproducing.

Like everything else, if something goes in, something comes out. In this case, the bio-engineered microbe pumps out PHB, a precursor to biodegradable plastic, or burnable alcohols, like isobutanol and isopentanol. In other words, a cousin of gasoline. Fossil fuels will eventually be exhausted, or too expensive to extract. Cheap gas prices notwithstanding-the artificial leaf offers the promise of fuel as renewable as the sun’s remaining 5 billion years is long.

Step 3 in the artificial leaf process may be to figure out how to use this new fuel source without putting the CO2 right back in the atmosphere, but one thing at a time.

In a sense, the new work comes out of the failure of the rest of the world to catch up with the initial purpose of the artificial leaf.

The first few years of the 21st century were filled with visions of a “hydrogen economy,” in which the most abundant element in the universe fed cars and homes powered by quiet, pollution-free fuel cells. Problem solved? Not quite, since splitting water was until now an energy-intensive process that required a lot of fossil fuels, which kind of defeated the purpose. The clean hydrogen economy could only take off if hydrogen was liberated from water using methods that didn’t create destructive byproducts. The artificial leaf is an end-run around that central problem.

Around 2013, Nocera met Silver, who had been looking for a different path by engineering an organism to take hydrogen from water. If the new research is scalable, oil exploration would theoretically no longer be necessary, Nocera said. Bacteria could just make it.

But there are some problems to making use of this technology. The first is the one faced by climate change activists, namely an entrenched, $8 trillion dirty energy system. Any new energy production method has to compete with something that works and is relatively inexpensive. (Nocera hopes India may be a more receptive market, where 300 million people have no access to modern energy.)

The second problem is that commercialization will be hard and expensive. The last several years have seen liquid-fuel technologies embraced by venture capitalists. A few transited from production to factory-level with great fanfare, only to slip away. Once promising fuel companies, Gevo Inc., in Englewood Colorado, and Kior, in Pasadena, Texas, have faltered.

That doesn’t mean renewable-to-liquid-fuels has no future. It’s just a long game. Nate Lewis is a prominent chemist at the California Institute of Technology who belongs to the Joint Center for Artificial Photosynthesis, a research consortium organized in 2010 with support from the Department of Energy. Lewis calls artificial photosynthesis “an inevitable technology, in that it takes the biggest energy source known to man, the sun, and stores it” in the densest form outside an atomic nucleus-chemical fuels.

The next steps for the technology is to scale up what’s been achieved so far, and to find catalysts (or in Nocera and Silver’s case, engineered genes) that can produce fuels that the economy wants and that take some of the carbon out of the air – until you burn them.