Marking 20 years
of bold journalism,
reader supported.
Science + Tech

David Keith’s Carbon Removal Moonshot

The Canadian scientist built a world-class emissions capture project in Squamish. But don’t misunderstand his aim.

Arno Kopecky 16 Aug 2019Alberta Views

Arno Kopecky is an environmental journalist and the author of The Oil Man and the Sea. He lives in Vancouver. This article first appeared in Alberta Views.

In 1958, when NASA initiated the Mercury Project that would send the first American into space, one of the many questions facing scientists was what to do about the carbon dioxide emissions of their astronauts. Our bodies exude about a kilogram per day of the gas, which is lethal in high concentrations; anyone living inside a sealed container quickly poisons their own air supply. NASA’s solution: disposable lithium hydroxide canisters that looked like metal speakers, whose contents chemically scrubbed and stored the spaceship’s ambient CO2. Once the mission was complete the canisters could be thrown away and forgotten along with the exhalations they held inside.

By coincidence, 1958 happened also to be the year a geochemist named Charles Keeling began measuring the atmospheric concentration of carbon dioxide at Hawaii’s Mauna Loa Observatory. Keeling’s first entry at Mauna Loa was 313 parts per million. Atmospheric CO2 concentrations fluctuate slightly over the course of a year, as the planet’s biosphere effectively breathes in and out with seasonal plant growth and decay. But Keeling soon observed what we all now know: The annual mean concentration of CO2 has marched resolutely upward every single year since 1958. The rate of increase is itself increasing, from an average of 0.8 ppm/year over the first decade of measurements to 2.3 ppm/year in the decade since 2009 — a velocity that strongly correlates with mass extinction in the geological record. Plot this on a graph and you get what’s called the Keeling Curve. We now know where it begins — with the industrial revolution about 200 years ago, until which time atmospheric CO2 was stable at roughly 280 ppm — but nobody knows where it will end.

Mauna Loa averaged 377.52 ppm in 2004, the year the University of Calgary hired a rising star in climate academe named David Keith “to help build a top-notch research centre that would inform the hard energy choices faced by Alberta, Canada and the larger world,” as Keith later described his ill-fated role as Canada Research Chair in Energy and the Environment.

Climate change had caught the attention of governments and institutions around the world, not least those whose budgets seemed irrevocably bound to the Keeling Curve. Keith, then 41, was a big score for the university. The son of an environmental scientist who worked for Canada’s wildlife service and contributed to the effort to ban DDT, Keith had a multidisciplinary skill set rooted in experimental physics and hardware engineering; he’d earned his Ph.D. from MIT by building the world’s first interferometer to study atoms, and later created a high-accuracy infrared spectrometer for NASA. He’d held research positions at Colorado’s National Center for Atmospheric Research and at Harvard and Carnegie Mellon universities. By the time he got to Calgary, he’d come to view climate change from an astonishing array of perspectives. Over the course of his seven years at U of C, Keith taught in the departments of chemical and petroleum engineering, economics, environmental design and physics and astronomy. He was the kind of person who kept people like Bill Gates up to date on the latest climate research.

“Dave’s this tremendously smart, very funny, highly opinionated guy who cares passionately,” says Jason Switzer, who has collaborated with Keith through his work as managing director of the Pembina Institute’s industrial decarbonization program. “He believes there’s just not enough advocacy, not enough concern, not enough action that’s anywhere close to the scale of the problem.”

Keith is also known for his willingness to confront adversaries. “You better have done your homework before you get into a fencing match with Dave,” says Switzer. Seven years after arriving in Calgary, Keith publicly lambasted his employer for, as he saw it, allowing oil money to corrupt the university’s academic integrity. “A group of players from the government and the energy business worked to muzzle a debate about energy and climate change,” he wrote in the Toronto Star. “[Science] that threatens industry’s interests does not get funded.”

582px version of DavidKeith.jpeg
David Keith is acting chief scientist of Carbon Engineering, the company he founded in 2009 with $3 million in seed money. Photo courtesy of Carbon Engineering.

Keith quit the University of Calgary and went back to Harvard, where he remains today, teaching applied physics and public policy. But Keith, an avid mountain climber, keeps a home in Canmore and still spends a third of his time in Alberta. When I reached him by phone in January, he was at a climbing gym.

“Lots of old Alberta families and conservatives care deeply about the environment,” he insisted. “It’s easy for people in, say, Cambridge, Massachusetts, to think people in the oil patch are stupid or evil. Neither of course is remotely true.”

Which is to say, whatever you think of the work Keith did in Calgary, or the company he founded there, he’s probably thought it too.

Keith founded Carbon Engineering in 2009 (Mauna Loa: 387.43 ppm) with $3 million in seed money, two thirds of it from Bill Gates and Murray Edwards, the oil sands billionaire and co-owner of the Calgary Flames.

Keith had begun studying direct air capture (DAC), “an industrial process that captures CO2 from ambient air, producing a pure CO2 stream for use or disposal,” even before his arrival in Calgary. That research accelerated in Alberta, and by 2009 his research had outgrown the lab. A private company would allow him to build a pilot plant and start engineering machinery that until then had only existed on paper. (Responding to criticism that he was profiteering off taxpayer-funded research, Keith has written: “My view is that universities... are too willing to accept a professor’s involvement in companies that are tightly tied to their academic research... I try and keep the division sharp. I ended all my academic work on DAC soon after forming Carbon Engineering. I have no research grants on DAC and no students or research staff working on it or any similar technology.”)

Direct air capture was first proposed in the 1990s as a process for achieving “negative emissions” — the point at which humanity takes more carbon dioxide out of the atmosphere than we put in. The idea was, and remains, pretty far-fetched. Beyond a few academic papers and a handful of startups, no one had taken the idea seriously until Keith showed up in Calgary. At that time, Keith estimated global R&D on direct air capture came to less than US$3 million. Universities didn’t have the money to pursue what amounted to a moonshot, and in the mid-2000s government and industry alike were far more excited about DAC’s big brother, carbon capture and sequestration (CCS).

The crucial difference between these two technologies is that CCS captures CO2 exclusively from the exhaust streams of major industrial sources — for example, coal-fired power plants — where carbon dioxide is up to 300 times more concentrated. That makes it a far more efficient place to gather CO2 than ambient air. But even as Alberta and Saskatchewan were gearing up to spend billions of public dollars on the world’s first generation of major CCS projects, climate scientists were starting to worry that the need to capture CO2 would long outlast the need for coal power.

For direct air capture, Carbon Engineering would have to move orders of magnitude more air to process the same amount of carbon. But an enormous advantage for DAC was that you were free to install it anywhere on earth so long as a clean power source were available. “You could build a bunch of these air capture projects in the middle of the desert, or in an off-grid area in Alberta that has relatively near-surface big, deep saline aquifers, and you could store your CO2 there,” said Switzer. CCS, by contrast, called for attaching monumental complex infrastructure onto the already monumental complex infrastructure of a power plant, which was in turn often surrounded by the infrastructure of a city.

This logistical challenge is a big reason why CCS failed to live up to the previous decade’s hype. Even if those logistics were overcome, its dependence on industrial emissions acts as an inherent constraint; CCS, by definition, can’t touch what are known as “legacy emissions” — the 750 billion tonnes or so of carbon dioxide that humanity has already put in the atmosphere. But direct air capture would draw carbon directly from the sky. Its theoretical potential was limitless.

The technology breakthroughs required to make DAC work in practice were more a matter of engineering than chemistry. Industry had developed a variety of means to separate carbon dioxide from ambient air over the past half century — but it was one thing to extract a few kilograms from a spaceship or submarine, quite another to harvest billions of tonnes from the sky. What kinds of machinery would you build, exactly, and how would the parts fit together? How much would they cost to build and operate? Most important, how much energy would it require? Energy, of course, was what got us into this mess in the first place — but by 2009, solar and wind power were finally becoming cost-competitive with fossil fuels, and Keith suspected that the high-energy requirements might no longer be an insurmountable factor.

In an article he published in the journal Science that same year, Keith made the case for Carbon Engineering: “Even if we could halt human carbon emissions today, the climate risks they pose would persist for millennia — assuming that we must rely only on natural processes… it is therefore in our interest to have a means to reduce atmospheric CO2 concentrations in order to manage the long-run risks of climate change.” Reducing emissions, he stressed, should remain the top priority for human civilization. “The global energy system is marvellously diverse, however, and in that diversity there are important niches where air capture might play a role in reducing emissions in the near term, long before we are able to bring emissions near zero.”

By that he meant what Carbon Engineering now calls air-to-fuels (ATF), a complementary process of adding hydrogen to CO2 to create carbon-neutral synthetic fuels that could complement, or outright replace, the gasoline, diesel and jet fuel being burned today.

Direct air capture, Keith concluded, “is neither a silver bullet nor a hopeless dream. It is simply another chemical engineering technology.”

The environmental community has tended to view carbon capture in all its forms as neither bullet nor dream, but rather as an insidious permission slip for the oil and gas industry to keep pumping till the very last drop. Given that we are still dumping more CO2 into the atmosphere with every passing year, and are currently at over 37 billion tonnes per year, that argument is hard to ignore.

Keith is not ignoring it. “The first rule of holes,” he told me, “is you stop digging the hole before you try and fill it.”

“That isn’t to say we shouldn’t develop the technology for carbon removal,” he added. “I absolutely think we should, and development means doing some deployment. But I don’t think there should be large-scale carbon removal until emissions are really headed down towards zero.”

At the same time, he considers it delusional to hope that humanity will bring emissions under control before it’s too late. By the time he founded Carbon Engineering, it was already too late.

That’s because there’s a 10- to 15-year time lag between the rise of carbon dioxide emissions and the subsequent rise in air surface temperatures. The climate impacts we’re seeing today — the forest fires, hurricanes, droughts and floods — were locked in around the time Keith moved to Calgary. As Keith knew then, most of those changes would be irreversible. The earth’s natural carbon sinks — primarily the ocean and our forests, though the global spike in forest fires is already turning those into sources rather than a sink — can only absorb about half the CO2 we put into the sky; the rest will stay in the atmosphere until the end of this millennium and probably much longer. Surface air temperatures are even less reversible: As was known by 2009, fully two-thirds of the maximum warming we impose on this planet will persist for 10,000 years.

Over the last few years, a resigned consensus has coalesced in climate circles. Kirsten Zickfeld, a researcher at Simon Fraser University and one of the lead authors of the latest Intergovernmental Panel on Climate Change (IPCC) report, told me, “It’s almost impossible to achieve [a 1.5-degree warming target] without negative-emission technologies. This implies carbon capture coming into the game in the next few decades and offsetting some of the emissions, but then towards the second half of this century they dominate.” That is what the IPCC reports now call an “overshoot” scenario. The only good news is that the models suggest we can reverse at least some warming, if we learn how to reverse the Keeling Curve.

In 2015 Mauna Loa surpassed 400 ppm, the first time CO2 concentrations had been that high in about three million years. To much less fanfare, Carbon Engineering’s pilot plant finally started running. Not in Calgary, though.

“We just didn’t ever get much love,” Keith told me. “The Alberta government never really stepped up to help in terms of grants, and neither did industry.” British Columbia, on the other hand, had not only a welcoming political environment for clean-tech industries but also a vast supply of carbon-free hydroelectricity.

The company settled in Squamish, B.C., an hour’s drive north of Vancouver. It’s not exactly Cape Canaveral, but the drab patch of gravel on which the unassuming fans-in-a-warehouse pilot plant has been humming away for the past three years does command a stunning view.

Carbon Engineering extracted one tonne of carbon dioxide per day, with only slight breaks, for three straight years; toward the end of that period, they added a barrel per day of synthetic fuel production to the process, deriving the necessary hydrogen by splitting water with electricity.

851px version of CarbonEngineeringPlantWorkers.jpeg
On site at the Squamish pilot plant, which in its first three years extracted one tonne of carbon dioxide from the air each day. Photo courtesy of Carbon Engineering.

Combining carbon and hydrogen to make a liquid synthetic fuel that can run a car or plane is nothing new. The oil and gas industry has been doing it for decades, but the inputs have always come from fossil fuels. Making it at scale from air and water would be unprecedented.

In June 2018 (Mauna Loa: 408.52), Keith and his colleagues published a highly technical paper in the energy journal Joule explaining how they’d brought the cost of CO2 capture down to $100 a tonne — one sixth the cost of their closest competitor, the Zürich-based Climeworks. At that price, Carbon Engineering says it can make carbon-neutral synthetic fuel for between US$1 and $2 per litre, depending on the electricity source (splitting water for hydrogen accounts for 70 per cent of the cost of the ATF process). That’s still too expensive to compete with gasoline in an unfettered market, but markets around the world are increasingly being fettered by carbon pricing, a trend that Carbon Engineering is betting will continue.

Steve Oldham, Carbon Engineering’s new CEO, told me “the value proposition for us is, if you believe that a low carbon economy is going to happen, however that manifests itself, then the capture of large amounts of negative-emission CO2 is an asset that will have considerable value.”

The Joule paper created a media explosion for Carbon Engineering. Go ahead and google it yourself. A second media bump followed in January 2019 (Mauna Loa: 413.20 ppm), when the company announced that Chevron and Oxy Low Carbon Ventures — a subsidiary of Occidental — had bought equity in the company, becoming the first oil and gas companies in history to invest in a direct air capture company (the size of their investment remained private). Carbon Engineering had doubled in size over the previous year, from 25 employees to 50, and Oldham expects to hire another 25 before the year is out. They spent the winter expanding the pilot plant in order to integrate the DAC and ATF processes, which until now had been running separately; the new facility will produce both powdered CO2 and liquid fuels in a continuous, automated loop at slightly higher volumes than the previous three years.

But that’s almost a formality, Oldham said. “We’re already starting the design phase of large commercial plants. Then we’re going out to the market to finance those plants, and also to arrange customers who will take the products of those plants.” Two kinds of commercial facilities are in the works.

The first is a pure DAC plant on 150 to 300 acres that captures one million tonnes per year of CO2, which “would do the work of about 40 million trees,” as Oldham put it. Oxy’s parent company, Occidental, is the world’s biggest consumer of pure CO2, using the equivalent of 50 DAC plants’ worth of output for its enhanced oil recovery operations today. They inject the gas into depleted oil fields, a common practice that can boost production by more than 50 per cent while leaving more CO2 below ground than is “liberated” in the oil.

The second is Carbon Engineering’s commercial ATF plant. It would occupy roughly 30 acres, and produce 2,000 barrels per day of fuel. As a loose thought experiment, consider that the world’s biggest oil refinery, in Jamnagar, India, occupies 7,500 acres and has a capacity of 1.2 million barrels per day; that same area could house 250 cookie-cutter ATF plants, which collectively could produce half a million barrels a day. Without even factoring in the footprint of oil extraction, that would put ATF in roughly the same ballpark as conventional fuel for land requirements. This is another critical selling point for Carbon Engineering, setting it apart from ethanol and other biofuels, whose gargantuan need for land has emerged as a crippling factor in recent years.

The company hopes to have its first generation of commercial plants running by 2021. The plan is to license their technology. That’s the only way it can work, from both a business and a climate perspective, given that thousands of commercial plants would have to be built to make any serious difference.

Jason Switzer, from the Pembina Institute, described the magnitude of Carbon Engineering’s scaling challenge this way: “We’re looking at a CO2 capture, use and storage industry that’s going to be moving somewhere between three and 10 gigatons of molecules. The only industry that moves that much stuff around right now is the oil and gas industry, and it took 100 years to build.”

I don’t know about your friend circle, but in mine a brief description of Carbon Engineering has proved to be a reliable generator of skeptical facial expressions. The relevant term here is “moral hazard”: the situation that arises when someone transfers the risk of their behaviour onto someone else. In this case, we get the money and the cheap flights to Costa Rica; our grandkids get dystopia.

At one point in our conversation, Oldham said his company was offering a “non-disruptive disruptive technology. We’re disruptive because we can make a massive difference in the amount of CO2 in the atmosphere, but we’re non-disruptive because building a plant to offset the emissions of a pipeline doesn’t affect the economics of that pipeline.” Like Keith, Oldham took pains to emphasize that reducing emissions was civilizational priority number one — but now, said Oldham, there’s a way to embrace that priority “without impacting critical things like economics, transportation, people’s lifestyles.”

That argument runs directly against a deep current of environmental thought which holds our economic systems, transportation habits and daily lifestyles to be very much in need of disruption — not just for the climate, but for our own psychological well-being.

And yet, when I started calling professionals to ask their opinion of what Carbon Engineering was up to, I discovered a surprising lack of disdain. The nearly universal attitude was, yep, we’re watching. “Carbon Engineering is a globally significant play,” said Switzer. Noting the oil and gas terminology, I asked him if Oxy and Chevron might not be using their investment to greenwash a corporate strategy that has done to the earth’s climate what the tobacco industry did to lung health. “If we’re going to bring these technologies to market,” he replied, “we’re going to need carbon pricing for sure, but we’re also going to need investment by first adopters. That’s going to involve the oil and gas sector.” (He also asked me if I’d heard of Pascal’s wager.)

Kirsten Zickfeld, the IPCC lead author, said: “These technologies need to be scrutinized carefully because they all have a number of side effects. But direct air capture, at least to my knowledge, is one of the ones with the fewest side effects.”

Zickfeld did emphasize the issue of moral hazard that came with any carbon capture technology. “I think the reason people are so excited by them is that they give us time. That obviously is a risky game, because we’re betting on something which does not exist yet.”

The week before I spoke to David Keith, Danielle Smith, former leader of the Wildrose Party and current Calgary radio talk-show host, published an op-ed in the Calgary Herald under the headline, “Let’s celebrate that Canada is likely a net-zero polluter.”

“Rather than absurdly trying to reduce emissions from 700 megatonnes down to 200 megatonnes by 2050,” Smith argued, Canada should “develop strategies to capture or offset 700 megatonnes.” Citing Carbon Engineering as an example of how we could get there, Smith concluded that the country should “get on with building pipelines.”

Keith was delighted to hear about this. “Oh, I love it,” he exclaimed. “She really said that?” He asked me to send him a link to the story and carried on: “So, it’s utter bullshit. It’s similar to a piece in Forbes that said this is ‘Capitalism Versus Climate Change’ — of course, we’re a proud company and we’re capitalists, but the only reason Carbon Engineering is able to raise money, and the only reason we’ll ever be able to be commercially successful, is government regulations that prevent using the atmosphere as a waste dump for carbon, period.” A world in which carbon is priced high enough for air-to-fuels to be competitive, Keith insisted, will be a world in which “the oil sands get hammered and eventually go out of business.”

A few days after that, I reached Smith to ask what she thought of Keith’s take on her article.

“I haven’t said I disagree with a carbon price,” she said. What she disagrees with is anything over Alberta’s price of $30 per tonne. “There are people you could push into energy poverty if you’re not careful about your greenhouse gas pricing policy.” Raising the spectre of poor seniors who might not be able to pay their home heating bills in the winter, Smith suggested, “if you go up to $200 a tonne, you’re going to kill people.”

So how about levelling off production in the oil sands, I asked — neither shrinking nor growing, but just staying even.

Smith declined the compromise. “As long as we continue to have the demand for oil products go up, Canada should continue to produce what the world demands.” Other, less environmentally friendly oil producers, she said, would simply fill whatever demand Canada decided to forgo.

That argument ignores both the moral imperative of climate change and the strategic opportunity of it. Canada wasn’t a globally significant emitter of the chlorofluorocarbons that were burning a hole in the ozone layer when I was growing up, either, but by hosting the Montreal Protocol we played an outsized role in forging the global agreement to stop producing them.

But I wasn’t expecting to make a convert of Danielle Smith, and didn’t try. What I did attempt earlier was suggesting to David Keith that he wouldn’t be history’s first visionary to have his message subverted by opposing forces.

He emphatically disagreed that Carbon Engineering ran any such risk. “Is there a moral hazard to battery-powered cars?” Keith asked. “Because fundamentally it’s the same kind of thing. And the only way those count as moral hazards is if you really believe the right thing is just to reduce consumption. So maybe from Naomi Klein’s point of view — and you really think the right thing is to reduce consumption and show that industrial capitalism is unsustainable — then anything that gets [us] out of the problem is in some sense a problem.

”The day after we spoke, Keith tweeted a link to Smith’s article with the comment: “Want a taste of Alberta’s right-wing group-think? see @ABDanielleSmith argue Canada is carbon neutral. Her shout-out to Carbon Engineering misses [the] mark — low carbon solutions depend on emissions laws.”

He got three likes, and zero retweets.  [Tyee]

  • Share:

Get The Tyee's Daily Catch, our free daily newsletter.

Tyee Commenting Guidelines

Comments that violate guidelines risk being deleted, and violations may result in a temporary or permanent user ban. Maintain the spirit of good conversation to stay in the discussion.
*Please note The Tyee is not a forum for spreading misinformation about COVID-19, denying its existence or minimizing its risk to public health.


  • Be thoughtful about how your words may affect the communities you are addressing. Language matters
  • Challenge arguments, not commenters
  • Flag trolls and guideline violations
  • Treat all with respect and curiosity, learn from differences of opinion
  • Verify facts, debunk rumours, point out logical fallacies
  • Add context and background
  • Note typos and reporting blind spots
  • Stay on topic

Do not:

  • Use sexist, classist, racist, homophobic or transphobic language
  • Ridicule, misgender, bully, threaten, name call, troll or wish harm on others
  • Personally attack authors or contributors
  • Spread misinformation or perpetuate conspiracies
  • Libel, defame or publish falsehoods
  • Attempt to guess other commenters’ real-life identities
  • Post links without providing context

Most Popular

Most Commented

Most Emailed


The Barometer

Are You Concerned about Your Municipality’s Water Security?

Take this week's poll