Can Carbon Capture Work?
Well, define "carbon capture". And "work". And "can".
Solving climate change will require amazing new technologies. We’re talking truly futuristic stuff: Self-flying air taxis. Meat grown in a lab. Jet fuel made from pumpkins. A nuclear reactor that fits in your pocket. Fully-electrified house pets. Semi trucks woven from hemp. A personal submarine made from goose shit and melon rinds that can impregnate endangered whales. We’ll basically be living in the “Technofuture” sketch from The Dana Carvey Show.
One of the most interesting futuristic ideas is carbon capture. The technology, which is currently used in a small number of industrial settings and is in the prototype phase elsewhere, sucks carbon dioxide out of the air. The appeal is obvious, especially if you believe, as I do, that the odds are against us hitting the “two degrees Celsius” goal set by the Paris Accord. I think we’re going to blow right past that thing like it’s a white guy in the 200 meters.
The two degree goal is, of course, just a goal. It’s not like everything’s nifty at 1.9 degrees and fiery death commences at 2.1. No matter where we end up, less carbon would be better. And since carbon dioxide stays in the atmosphere for hundreds of years, a long-term carbon removal project seems desirable under virtually any circumstances. A report by the Intergovernmental Panel on Climate Change released in 2018 called for a major push to develop carbon capture, which sparked investment and interest.
But is it feasible? Does the technology show real promise, or is this techno-idealist horseshit? What are the economic and engineering challenges, and how do they relate to policy choices? These seem like good questions for a comedian to try to answer. So here goes nothin’.
The phrase “carbon capture” can refer to several things. It can refer to altering the environment to absorb more carbon through things like reforestation or tweaking the chemical composition of the ocean. People are working on this: Elon Musk recently donated one million dollars to grow trees, an amount chosen because it was how much cash he happened to have in his jacket pocket at the time. Also, my mom planted a dogwood tree in her back yard. So, big progress is being made.
But this strategy is too complex to dive into today. My time at EPA taught me that the answer to every question about land use is “it’s complicated”. Everything depends on what you’re doing, where you’re doing it, what the land was used for, what the land could be used for, what the land nearby is used for…it’s a 20-sided Rubik’s cube. Honestly, Mom planting that dogwood tree probably started us down the road to complete ecological collapse.
“Carbon capture” can also refer to capturing emissions at power plants and other industrial sites. This is highly-controversial; is it green or does it perpetuate the use of fossil fuels? That question deserves its own column. But I’ll say that my position on power involving carbon capture is the same as my position on nuclear power: We’re so desperate for low-cost, low-carbon energy sources that I don’t know why we would preemptively rule anything out. A lot of smart people think that solar will probably be the energy source of the future, and I wouldn’t bet against that. But there’s a “probably” in that sentence. Every energy source has pros and cons, and we should weigh those pros and cons objectively. We need clean, cheap energy the way a starving person needs a sandwich; we can’t be saying “but no mustard, please, and I’m trying to cut out gluten, and everything should be locally-sourced.”
The last thing I’ll say about industrial carbon capture is that one argument in the “pro” column should be that developing the technology could help us in other areas. Even if energy ends up coming from other sources in the long run, what we learn from capturing and sequestering carbon from power plant smokestacks could help similar processes in other places. For example, there would definitely be applications for cement production, which is eight percent of the world’s CO2 emissions all by itself. It could also push us down the learning curve on the technology I want to focus on today: direct air capture (DAC).
Direct air capture captures carbon directly from the air; it’s the rare thing that sounds like what it is. The idea is that…uh…someone would build…uh…something, and that thing would take carbon out of the air and…uh…do stuff with it. As you can tell: We’re basically there. Here’s my highly-scientific schematic of the technology:
A bit vague? Perhaps. But Theranos was valued at $10 billion when they had basically this level of detail, so if I slap on a black turtleneck and pitch this to George Shultz, he might give me a big pile of money. At any rate: On the most basic level, the technology works — you can grab carbon from the air. Forms of carbon capture have been used since the ‘70. Yes, the ‘70s! When we were still smoking in restaurants and heaving our luggage through the airport like a bunch of dumbasses, we were also extracting carbon from the air.
The first engineering challenge is getting enough carbon into contact with whatever you’re using to absorb it. The CO2 concentration from a coal plant smokestack is between 10 and 15 percent, but it’s only around .04 percent in normal air. Hence the Big Fucking Fan: You need to suck in enough air to capture carbon at a reasonable rate. There are ideas for absorbing carbon without the Big Fucking Fan, but the less air you process, the more you exacerbate the second engineering challenge, which is…
Surface area. Again: It’s about getting carbon in contact with the absorbing material. More contact = more absorption. This requires the construction of gigantic facilities. MIT professor Howard Herzog made some calculations based off of one of the most high-profile DAC pilot projects, Carbon Engineering in Canada. CE has designed (though not built) a plant that would absorb a million tonnes of CO2 per year. Wow…a million tonnes! Let me show that graphically — the bar below represents our total CO2 emissions from energy, and one million tons is shaded in RED:
As usual, I’m being a dick. CE doesn’t claim to be changing the world; they’re just building a pilot project. But I wanted to introduce a sense of scale, because scale is a problem. Professor Herzog calculates that the one million ton facility would require a structure ten meters tall, one meter deep, and 2.86 miles long.1 To get to a billion tonnes (one gigaton, about 3 percent of the bar above), you’d need a structure as wide as the entire United States.2
Of course, you wouldn’t do that; you’d break it into pieces. But, putting one unit behind another reduces efficiency, so you’d have to space it out, and that takes a lot of land. Exactly how much land is difficult to calculate, but to get one gigaton, an estimate in the mid-high range would require a land mass the size of Delaware. And I’m sure you’re thinking: “Cool, bulldoze Delaware. What’s the problem?” And I could not be more with you: As a frequent driver on I-95, those fucking First State bridge trolls have dinged me for $15 so many times that I’m ready for some payback. The only problem is: The fucking president is from Delaware. He won’t co-sign a plan to raze the state; he’s too attached to the assemblage of tax-free malls and off-track betting sites that he calls home. The point is: DAC requires a lot of land, and land is difficult to acquire.
These are just some of the major logistical challenges DAC needs to overcome. They’re not necessarily insurmountable, but you have to provide an incentive for someone to surmount them. Which brings us to the economics: DAC is currently very expensive. Right now, it costs in the neighborhood of $600-$1,000 to remove a ton of carbon from the air; most observers think it needs to get down to $100 at the most for DAC to be viable.
There’s one obvious way to drive down costs: Sell the carbon to recoup some money. There’s a market for carbon. Right now, we produce carbon the same way we produce anything: By burning fossil fuels. If you could sell the carbon that’s a byproduct of DAC, then that would be an unambiguous win for the environment. Right?
Well, terrible news: One of the main uses for carbon is…(deep sigh)…oil extraction. Yep: oil extraction. Good one, God! A real knee-slapper. But it’s true: Carbon injected into wells can loosen up oil that can then be pumped, kind of like how if you put some water in a near-empty shampoo bottle and shake it up, you get two extra days of White Rain. So, yes, carbon has uses, and one of those uses it to make the entire exercise of direct air capture damn-near pointless.
Carbon has other uses: synthetic fuels, carbonated beverages, and enhanced cement, to name a few. Some carbon-capture products seem mostly for show: For example, here’s a pair of carbon-capture sunglasses. Now, I’d hate to gain a reputation as a cynic, but I’m skeptical that carbon-capture shades will get us to net zero emissions. I think they’re a niche product that allows a company to say “look at this cool thing!” And the local news will do a five-minute piece on the cool thing because the local news will broadcast any ol’ shit; if you fart on a pound cake, Channel Four Action News will absolutely do five minutes on that.
The market for many carbon products is very small. For example, you can use carbon to make dry ice. How many uses are there for dry ice? I can think of two: Organ preservation and spooky cauldrons at elementary school Halloween parties. How much dry ice does that require? Unless I’m vastly underestimating the number of witches’ cauldrons (or vastly underestimating their spookiness), I can’t imagine dry ice being a major revenue stream.
Selling extracted carbon will probably never provide more than a minor cost offset. That’s not just my opinion: That’s the Intergovernmental Panel on Climate Change’s opinion. They write: “Industrial uses of captured CO2 [except for the oil-extraction method I mentioned] are not expected to contribute to significant abatement of CO2 emissions.”
So, what do you do with the captured CO2? The answer is: You trap it underground and keep it there for a long time. You may recognize this strategy from the Legend of Zelda games, which always end with Ganon being trapped underground. I have never gotten a scientist to admit that this was stolen from Zelda, but it seems like an awfully big coincidence.
The strategy seems to work better in real life than it does in Hyrule (Ganon always escapes when Nintendo needs a flagship video game title). Porous rock formations, which are plentiful (though not evenly distributed around the globe), can trap dense gasses, and surrounding geological features can further reduce the risk of leakage. But it’s not perfect; leakage is possible, and there’s also a small risk of — ahem — earthquakes. Honestly: The politics of this sound extremely dicey to me. It’s too similar to the fracking narrative; fracking also involves injections into rock formations, which entails risk, and the general public is extraordinarily bad at assessing risk. All you need is one leak at one site — which would likely entail one jillionth of the health risk posed by climate change — and you’ll get screaming “GAS LEAK AT INJECTION SITE” headlines. And that will change the politics of DAC forever.
So, add that to the long list of challenges DAC is facing. The truth is, we’re nowhere near DAC being scalable. For DAC to work, it needs to follow a Moore’s Law-type trajectory (Moore’s Law, of course, is the law that states that whenever you buy a computer, you find yourself thinking “why did I buy this piece of shit?” within two years). Really, for DAC to have any chance at all, two things must happen.
First, we need low-cost, low-carbon energy. Every step in the DAC process takes energy, from running the Big Fucking Fan to processing the CO2 to sending it to its final destination, whether that’s deep underground or into some stupid pair of Ray Bans. Without cheap, green energy, the whole thing is pointless. It absolutely won’t scale without the energy component, and it’s difficult to even develop without the energy.
And the second thing we need is — say it with me now! — a price on carbon. God I feel like a dork arguing for a price on carbon for the millionth time. I’ll bet when I have kids, they won’t want to have friends over because they’ll be afraid that I’ll trap them in the kitchen and talk their ear off about a price on carbon. Which I probably will — educate yourself, kids! But I don’t know what else to say; people will only pay to have carbon sucked out of the air when the price is lower than the cost of releasing an equivalent amount of carbon into the atmosphere. There’s just no way that DAC happens without a carbon tax.
So, will DAC happen? It’s a big, fat “who knows?” I sure as fuck don’t. Of course, that puts DAC in a different category from things like solar power and electric cars, which are already competitive and will probably be dominant in the near future. The good news is that a price on carbon, which is the main policy3 that would boost DAC, is the same policy that we need anyway. That is, of course, the main advantage that a carbon tax has over government investment; instead of trying to guess which technology will bear fruit, you let the market sort it all out. But I’ll save the rest of this sermon for my future child’s friend. My personal feeling, after a whole week of research, is that DAC is about as likely as an air-breathing fish or laptop-fresh milk. But that doesn’t mean it’s completely impossible.
I really only scratched the surface of this topic, so here’s some more reading:
Here’s Professor Herzog’s 2019 book.
***POLL FOR TUESDAY***
I'd like the next column to be about...
1. Could diversity, equity, and inclusion efforts be making us more biased? [VOTE FOR THIS]
2. Why are narratives so powerful in shaping how we see the world? [VOTE FOR THIS]
I spoke to Professor Herzog on the phone (he was nice enough to make time to try to get all this through my thick skull), at which point he shared a book chapter with me from a book that hasn’t been published yet. That’s why there’s no link for these stats — they come from an unpublished book.
By the way, this discussion of scale should answer the question: “Could we possibly get so good at DAC that we don’t need to reduce emissions?” The answer is: No fucking way.
I should mention that government investment is currently supporting DAC. Right now, most DAC pilot projects are heavily-dependent on government funding.