Carbon dioxide levels in the atmosphere have touched 400 parts per million for the first time in at least 800,000 years. The jagged saw-toothed line of the Keelings' father-and-son measurements climbed above that milestone briefly this month before the budding growth of the Northern Hemisphere's spring began sucking CO2 back out of the sky. But human greenhouse gas pollution looks set to continue to rise—and photosynthetic plants on land or at sea can only do so much. As greenhouse gas levels increase further, could machines help wash the skies of the excess CO2?

On an early spring day touched with the promise of warmth plus the threat of rain, I headed up to Columbia University's Mudd Building and the lab of physicist Klaus Lackner, formerly of Los Alamos National Laboratory. His hearty chuckle belies his formal German diction and physicist's habit of obfuscating with numbers. Girding myself for potentially indecipherable jokes, I'm here to see Lackner’s potentially world-saving technology: a plastic resin that can capture carbon dioxide directly from the air.

The resin rests outside a clear greenhouse bearing basil plants, bamboo, a houseplant and cucumbers that glow an eerie purple-red under ultraviolet light. The plants' leaves rustle in the breeze from a Dyson bladeless fan. Next to the big tank, a computer monitor charts CO2 levels and a tube on one side separates the environment within the greenhouse from the outside world. With the UV light on, the plants are busily sucking in CO2 to make leaves, roots and vegetables. "The cucumber got fat on the CO2," Lackner notes and chuckles.

A pale beige polypropylene plastic embedded with 25-micrometer particles of the resin is inserted into the tube in the form of a long-haired shag carpet sample and, almost immediately, CO2 levels inside the greenhouse begin a steady march downward as the resin binds CO2 to form bicarbonate, a kind of salt produced. This type of salt, more familiar perhaps as baking soda when there's a sodium atom involved, holds the CO2. The resin sucks in CO2 even more powerfully than the plants do, as a function of the relative humidity of the material. That makes the process reversible; just add water to get the CO2 back out again.

This is no joke. A polycarbonate plastic bottle used to store some of the resin ended up scarified. "They broke the plastic," Lacker says of his lab co-conspirators, showing me the streaked, cloudy, hard plastic bottle. The resin pulled CO2 out of the polycarbonate in its vigorous quest for chemical equilibrium.

Lackner calculates that more than 700 kilograms of CO2 passes through an opening the size of the door to this lab over a 24-hour period when the wind is up, courtesy of another Dyson or just a windy building top. That's how much a sheet of this material might pull from the air. Or it could be refashioned into a brushlike or folded checker configuration, exposing more of the resin.

Of course, 700 kilograms of CO2 only equals the breath of 13 people for one day and night. There would need to be a lot of these resin machines to make a significant impact on pulling this trace greenhouse gas out of the atmosphere to lower atmospheric concentrations; Lackner estimates 10 million such artificial “trees” would be required to drop atmospheric concentrations by 0.5 ppm per year. Each machine would require roughly 1.1 megajoule of electricity for pumping and compressing per kilogram of CO2 captured. That's not to mention all the water required to wet the filters (and evaporate) in order to get the CO2 back out again so the resin can be re-used to capture yet more CO2. The compressed and captured CO2 can then either be used for industrial purposes, like enhanced oil recovery to improve the economics of all this, or buried deep beneath the surface of the planet. In other words, a vast industrial infrastructure of air-capture machines would be required to remedy the effects of our vast, industrial infrastructure for fossil fuels.

Just how the resin operates is the focus of the other experiment in this lab. Hidden inside a Styrofoam cooler—with a dark blue Columbia necktie as de facto latch—the resin is exposed to water and CO2 and precisely weighed while temperature is kept constant. The idea is to keep CO2 steady at 400 ppm with no temperature variation and then change the conditions to determine how well the resin works.

In fact, it may well prove that another similar material works better to directly capture CO2 from the air. TerraLeaf is working on using chlorophyllin—the chlorophyll from a plant turned into a salt by adding sodium ions and copper—paired with an electrically conducting polymer to form a membrane that can pull CO2 (or other greenhouse gases) from air and form carbon-based chemicals, and potentially even fuels. Harvard University scientist David Keith and his team are working on making a machine capable of pulling CO2 from the air using liquid sodium hydroxide, also known as lye, and then reheating it to release the CO2, allowing for continuous operation. This is how breathable air is re-created on submarines and spaceships, after all. And there are yet more options from groups such as Climeworks, the Georgia Institute of Technology, and the University of Southern California.

But Lackner's polymer may be hard to beat on price. The resin, made by Dow and known as Marathon MSA, finds use in food processing and water purification, among other applications, and costs just $2.50 per kilogram, according to Lackner. They are still working on their first roll of the material after nearly a decade of experimentation.

Yet, therein lies what may prove the ultimate challenge of such direct air capture: cost. On the one end, no one is willing to pay to suck CO2 out of the air at present and, on the other, few are willing to pay for CO2 either to be stored or used. An estimate from the American Physical Society (pdf) suggested that such air capture might cost roughly $600 per metric ton of CO2 captured, which makes even Lackner's bulky resin “trees” into the Tesla Roadster of emission reductions. Capturing a gas that makes up just 0.04 percent of the air may prove too energy intensive and too expensive to sustain.

Then again, given the climate change impacts already seen at present greenhouse gas concentrations, the world may not have much choice. Lackner and colleagues argued in a paper published last July in Proceedings of the National Academy of Sciences that air capture may prove the only way to deal with greenhouse gas emissions from transportation, all those tailpipes and engine exhausts on the world's millions of cars, airplanes and other vehicles. "Given the enormity of the global climate challenge, we think this [air capture research and development] needs to be scaled up urgently," the research team wrote. Already, the $25-million Virgin Earth Challenge Prize for CO2-reduction technologies has identified Lackner's work, along with four other air capture schemes, for award consideration. The other possibilities include biochar and biofuels with CO2 capture as well as efforts to enhance natural processes that capture CO2 like rock weathering and vegetation regrowth.

Human civilization shows little inclination to reduce CO2 the cheapest industrial way: cut greenhouse gas pollution from the biggest, most concentrated sources, such as coal-fired power plants or oil refineries. Placing carbon capture at these industrial facilities is significantly cheaper than general air capture and reduces emissions from the single largest sources of CO2. As Princeton University mechanical engineer Robert Socolow argues: it makes little sense to capture CO2 from the air until these sources of pollution have been eliminated.

The ultimate problem may be that the technology cannot work overnight. Even if Lackner was able to deploy his millions of artificial trees employing this resin, it would take decades at least to restore pre-industrial atmospheric concentrations of CO2. And then the question becomes: What is the appropriate concentration of CO2 in the atmosphere, and who decides? In the end it might be easier to just stop pumping CO2 into the air in the first place and let photosynthetic plants handle the rest.