Skip to main content
rob magazine

With Carbon Engineering, physicist David Keith has developed a simple, scalable method for capturing CO2 from the air

The company built a pilot plant in Squamish, British Columbia, a town better known as a climbing mecca than a clean-tech hub.Alana Paterson

David Keith would like you to know that his technology for sucking carbon dioxide directly out of the atmosphere is not a single, spectacular solution for climate change.

He especially discourages any suggestion that Carbon Engineering, the company he founded in 2009 to develop a simple, scalable method for direct air capture of CO2, is engaged in some kind of climate-mitigating “moon shot.” In a conversation from Harvard University, where Keith has appointments in both the Paulson School of Engineering and Applied Sciences, and the Kennedy School of Government, he responds with a sharp, “That's ridiculous. It's a stupid thing to say.”

First, given the breadth of the problem—the amount of climate-changing CO2 already in the atmosphere and the additional greenhouse gases humans are adding to it every day—Keith says, “No single thing is going to take that much of the pie.” Beyond which, he says, it would be a serious mistake to imagine that an after-the-fact carbon cleanup technology will allow everyone to continue, say, driving their cars, eating their burgers, flying whenever they want and burning vast amounts of Amazonian rainforest. Specifically, he says, it's not sensible to rely on Carbon Engineering's technology to take CO2 out of the atmosphere while we're still adding massive amounts by doing things like burning coal: “That's just silly.”

So, we know that David Keith—lean, hungry-looking, an avid mountain climber, as well as a brilliant scientist—ranks low in his capacity to suffer fools. But it’s the “moon shot” analogy that apparently rankles most. That, he says, is because the race to put a human on the moon was “a large government project in which cost was no object” with “a high technical risk.” Carbon Engineering, on the other hand, is small and uncomplicated. (“This is not particularly hard engineering,” says Keith.) It has been developed to its current proof point on a relative shoestring, and it can ultimately only succeed if the CE team can prove that its carbon capture method is cost-effective at scale.

But here's the thing: Carbon Engineering is already giving tours of a functional direct air capture facility that may, indeed, provide a spectacular—if not singular—solution to the hardest part of the climate change problem. Even if we stop adding to the greenhouse gases that are warming the world at a dangerous and unprecedented pace, we still have to do something about legacy emissions—the billions of tonnes of CO2 we have poured into the atmosphere since the beginning of the Industrial Revolution. And some pretty savvy investors, including Bill Gates and Canadian Natural Resources Ltd. (CNRL) cofounder Murray Edwards, are betting CE has the key.

What began 10 years ago as a pilot with about $3.5 million in seed money is now a proven technology, capturing CO2 from the air. So far this year, the company has attracted more than $115 million in next-step financing—$25 million from government, but most from industry heavyweights, including Occidental Petroleum, Chevron and mining giant BHP. Occidental and CE are collaborating on an industrial-scale plant designed to capture 500 kilotonnes a year for use in Occidental's enhanced oil recovery operations in Texas. This might be liftoff.

In the first step of Carbon Engineering’s direct air capture process, this fan—which sits atop CE’s pilot plant’s air contactor—pulls atmospheric air into the structure.Alana Paterson

The process of sifting the atmosphere for carbon dioxide seemed, until very recently, more than a long shot. As a matter of practice, we’ve had CO2-harvesting technology since the 1950s. NASA perfected some excellent (and expensive) processes in the run-up to the actual moon shot so that astronauts, cooped up in a small capsule, wouldn’t asphyxiate on their own exhalations. And energy companies have been scrubbing CO2 from coal flues and oil refinery stacks for decades, reusing it in industrial applications. But the concentration of CO2 in the atmosphere is both too large and too small to make this process easy on a grand scale. It’s too large because human activity has inflated the atmospheric content of the heat-trapping gas from 280 parts per million (ppm) in the late 1700s to about 410 ppm today. That’s enough to raise the global temperature to dangerous levels, but when you’re trying to reach into thin air to pick out specific molecules you can’t see or smell, it’s still a vanishingly small proportion—just 0.04%.

Apparently, David Keith found the prospect less daunting. He was raised in Ottawa and educated at the University of Toronto (B.Sc. in physics) and the Massachusetts Institute of Technology (PhD in experimental physics), where he won MIT’s Prize for Excellence in Experimental Physics for developing the first interferometer for studying atoms. His followon successes—which included building a high-accuracy infrared spectrometer for NASA and becoming one of the leading innovators and commentators on geo-engineering—were enough to land him on Time’s Heroes of the Environment list in 2009. Five years earlier, Keith had relocated to the University of Calgary as the Canada Research Chair in Energy and the Environment, and if he didn’t get much love from the Alberta fossil fuel establishment, his work caught the eye of Microsoft founder Bill Gates. Keith was recruited as one of the billionaire’s advisers on energy issues.

There followed a moment when Keith’s tendency to understatement came in handy. He says he wrote the first business case for Carbon Engineering “in a very Bill-oriented way.” It was “low key—not hypey.” And Gates responded with what, in his budget, must have seemed like an understated investment of $1.5 million. That enabled Keith to raise as much again in seed funding from, in his words, “small family investors,” including Edwards, the oil sands magnate who, in addition to having co-founded CNRL, is also a major shareholder and chair of Ensign Energy, one of Canada’s biggest energy services companies. “Murray wanted to support the University of Calgary, and he gets that the climate problem is real,” says Keith of the press-shy billionaire. “He wants a win.”

The Alberta government was less engaged, so Keith found himself drifting over the mountains to B.C., where there was greater enthusiasm for supporting clean-tech startups. He found a cheap lease on some waterfront property in Squamish, British Columbia, that had undergone remediation for mercury contamination.

Company lore has it that he then put up a “no entry"—style sign that said, “No Science.” Geoff Holmes, one of CE’s original employees, says Keith assembled the early development team—a revolving group of eight or 10 engineers and scientists—from among the students and research associates at the University of Calgary. Holmes, for example, had just completed his master’s of science with a major project on direct air capture. But Keith made it clear he wasn’t looking for something new and complicated. Rather, his vision was to source every component in the CE process from some other industry, using off-the-shelf parts that could be manufactured at a low cost and supplied at massive scale. Keith also decreed that CE’s process should require minimal consumables—that is, inputs other than CO2 and energy that would have to be constantly renewed or replaced. That meant the process had to recover and regenerate all chemicals for reuse. After six years of testing and fine-tuning his original model, that’s what Keith got. By 2015, CE had settled on a three-step process to strip the CO2 from ambient air, concentrate it into a solid and then turn it back into a gas that could either be used or buried.

It functions much like a cooling tower on top of any large building.Alana Paterson

The fan sucks air through a big plastic honeycomb that’s constantly flooded with fluid.Alana Paterson

In the first step, a fan sucks air through a big plastic honeycomb that’s constantly flooded with fluid—much like a cooling tower on top of any large building. But instead of the fluid being used to shed heat, the liquid in CE’s variation contains potassium hydroxide, which reacts with CO2 in the atmosphere and bonds to it. The result is a fluid with a CO2 concentration of 2% (up from 0.04% in the atmosphere).

In the second step, CE uses a pellet reactor common to most water treatment plants. Water flows into the reactor, where it’s seeded with calcium carbonate pellets, which bond with CO2 until they’re heavy enough to fall to the bottom. What you’re left with is a material that is chemically similar to seashells. These little pellets now have a CO2 concentration of 45%, so you could simply bury the product, effectively sequestering the carbon in a solid state. But then you’d have to maintain a permanent mining operation to source the component chemicals consumed to this point.

Instead, CE goes to the third step. Using a calciner—a steel cylinder in an airtight furnace commonly used to process mineral ore—the “seashells” are heated to 900 C, liberating 100% of the CO2 and leaving behind calcium that can now be reused.

The next challenge is figuring out what to do with the CO2, which, unattended, will simply disappear, literally, into thin air.

Inside the pellet reactor, water is seeded with calcium carbonate pellets, which bond to carbon dioxide and create a material similar to seashells.Alana Paterson

One choice is to pressurize it into a liquid state, pump it into geologically stable formations in the ground (deep saline aquifers or recently emptied oil wells) and seal it off, one hopes, forever. The alternative is to mix the CO2 with hydrogen to form something very similar to the hydrocarbon fuel that got us into this trouble in the first place. This might seem perverse at first blush, but the CE fuel would be endlessly renewable, perfectly clean-burning (unsullied by any of the sulphur or heavy metals that stick to fossil fuels even after refining) and carbon neutral—it wouldn’t create a net addition of CO2 to the atmosphere because CE’s feedstock is atmospheric CO2.

So, CE set up a fourth step, also reliant on traditional industrial processes. The company uses an electricity-based system to separate water into oxygen and hydrogen, and then combines the hydrogen and CO2 under pressure to form liquid fuel—which, as long as the electricity used to make it comes from a renewable source, is nearly or fully carbon neutral.

The pellets have a CO2 concentration of 45 per cent.Alana Paterson

In the usual history of academic-driven startups, this is the perfect moment for an entrepreneurial conflagration. It’s so common for startups to fail at this point that people refer to the transition phase as “the valley of death.” Having proved their concept, most entrepreneurs don’t have the skill or discipline to move to the next stage, and they don’t know when to give up the reins. Here again, however, CE appears to have tapped into a different kind of industrial precedent.

The functional grandfather of all things high-tech in British Columbia is MacDonald Dettwiler and Associates (MDA)—maker of everything from the Canadarm to broadcast satellites. The company was born in 1969, when University of British Columbia electrical engineering professor John MacDonald (also an MIT PhD) linked up with computer scientist Vern Dettwiler, in large part to create local opportunities. MacDonald was tired of seeing hot talent training at Western Canada’s premier research university and immediately leaving Vancouver for work. In building a company that now has more than 5,000 employees worldwide, one of MacDonald’s greatest joys was the energy, opportunity and economic activity MDA generated in the B.C. market. The evidence is everywhere, but it’s most obvious on a map posted on the wall at the BC Tech Association. The map shows thick bundles emanating from UBC and Simon Fraser University—the most common points of origin for the innovations and people that have created some of B.C.'s biggest tech successes. But the next-thickest bundle spreads from MDA to local companies like Creo and Mobile Data International, and further afield, to firms like BlackBerry. MDA’s technical and managerial talent has been infectious. And CE appears to be suffering a fortuitous outbreak.

It started with Denis Connor, an MDA veteran, a founder of QuestAir, a serial supporter of alternative energy technology, and an early CE investor and director. In 2017, Connor recruited a former protégé, Dan Friedmann, who was MDA's CEO for 20 years, beginning in 1995. Friedmann soon took over as the chair of CE's board (Keith still sits as a director) and recruited Steve Oldham, who had been MDA's senior vice-president for business development. As CEO, Oldham has since hired two other MDA business development specialists, Doug Rae and Lori Guetre. (MDA is now called Maxar Technologies and has shifted its headquarters to Colorado.)

As Friedmann says, these are all people who are accustomed to solving big problems and, often, getting startups quite literally off the ground. As a satellite business, he says, pretty much everything MDA built under his leadership cost hundreds of millions of dollars, and the hardware didn't pay off until you designed it, sold it, launched it into space and, in a moment that could be thrilling or chilling, turned it on. (Oldham says this speaks to one of the things he likes about CE: Unlike radar satellites, “our plants stay on the ground; we can fix them.”) So when you talk to MDA veterans, they turn quickly to practical questions of financing and insurance—of raising capital, reassuring investors and allying with great partners.

The calciner superheats the pellets, separating them into pure carbon dioxide and calcium, which can then be reused.Alana Paterson

In regards to that last point, CE’s leading partner is Houstonbased fossil fuel giant Occidental Petroleum. As Oldham says: “Who better to put CO2 back in the ground than the people who have been digging it up?” But Occidental has a particular advantage. According to Richard Jackson, president of subsidiary Oxy Low Carbon Ventures, over the past several decades, Occidental has sequestered roughly 500 megatonnes of CO2, without a single safety violation and with a 99.9% success rate. It has generated that CO2 from 13 plants, built 4,000 kilometres of pipelines and created 34 CO2 injection sites (or “floods”) in an area covering 1.4 million acres.

To be clear, Occidental hasn't been doing this for carbon credits. The company is the world leader in enhanced oil recovery, in which it pumps CO2 down into mature reservoirs in order to extract oil and, because it's sequestering CO2, to provide a product with a lower carbon density. And given rising global consciousness of the risks of global warming—and the policy changes already coming into effect—Jackson says Occidental is committed to continuing to lower the direct emissions of CO2 per barrel of oil, increase the energy efficiency of all its operations and expand its capacity for carbon capture, utilization and storage.

The proposed CE plant, which could cost up to $1 billion (and for which they are still gathering financing), is perfect for all three because it means Occidental doesn’t have to rely on nearby sources of industry-generated CO2 or expensive pipeline infrastructure. A Carbon Engineering plant can be put anywhere that has a ready source of electricity—and Texas is a perfect place to build large-scale renewable power. And, as Keith says, a 100-megawatt solar array is now the cheapest form of solar energy on the planet.

Once refined, CE’s synthetic crude—made from air, water and renewable electricity—can power existing vehicles without the need to modify the engines.Alana Paterson

That gets, finally, to the question of why anyone would bother with all these fuel plays when solar is so cheap. Why not just crack water for hydrogen and build a fuel-cell world? Friedmann says it’s all about infrastructure. Unlike hydrogen, which is highly explosive and complicated to handle, CE’s fuel can flow seamlessly into all current fossil fuel applications. It’s easy to refine in conventional facilities and cleaner, as well. Unlike biofuels, it’s blendable up to 100%; you can pour it straight into your car’s gas tank. It’s also portable. Oldham says liquid fuel is 30 times more efficient for carrying energy than batteries in terms of energy content per kilogram of weight. And it works easily in all those hard-to-electrify applications—not just automobiles, but ships and planes too. As Friedmann puts it: “It’s a world-changing product that doesn’t require the world to change.”

We are, of course, in transition, and Keith is especially alert to the question of “moral hazard.” There is a concern that this not—a—moon shot might make people and policymakers complacent, as if there is suddenly less need to reduce emissions because CE will be able to draw them out of the atmosphere later—no need to switch infrastructure, because a carbonneutral fuel is coming to a plant near you. Ultimately, the latter might be true. Carbon Engineering hopes to build proprietary plants to prove their effectiveness at scale and then to license the technology, so its plants can be replicated quickly and globally. By peer-reviewed calculation, a one-megatonne plant should ultimately cost about US$700 million, at which price, you could harvest CO2 profitably using tax credits, such as those already instituted by California and the U.S. federal government, that will pay companies for the CO2 they sequester.

Which leads back to the impatient David Keith and his passionate position that direct air capture is but one small tool at a time when we have to be using everything we can to address the gathering climate crisis. And really, with Keith in the room, you shouldn't suggest otherwise. Pretty sure he would tell you, “That's a stupid thing to say.

Your Globe

Build your personal news feed

Follow topics related to this article:

Check Following for new articles