Four Direct Air Capture Technologies
Air, water, limestone, and plants are the basis for current working solutions for Direct Air Capture (DAC)
You’re reading Terraform Now, my newsletter on carbon capture and other planet-altering technologies. To support my work, you can subscribe here:
In technological development, a ‘dominant design’ is the widely accepted way of making the best version of something. Best means the most effective, cheapest, and simplest way to deliver on the goals of a project.
The dominant design for aircraft engines was the propellor, then the jet engine came along. Computing evolved from centralized mainframes to distributed data-centers, then back to a more scalable centralized architecture, branded ‘cloud.’
We are so early in Direct Air Capture (DAC) that there is no dominant design — instead there are four designs:
While there are other designs in development, these designs are proven. We have good evidence that each of them have been able to take CO2 out of the sky at a cost of less than $1000 per ton.
Let’s look at each of these, beginning with air-based DAC.
AIR-BASED DAC IS THE MOST POPULAR
It’s tempting to label the air-based1 solution the dominant design, partly because most companies are taking this approach. Climeworks is probably the most well-known air-based player but there are many others like Global Thermostat and Kenya-based Octavia Carbon.
Here’s how air-based capture works:
Large fans blow air through ‘frameworks,’ structures with high surface-area that are coated with materials that do the capturing. These capturing materials are often called solvents or adsorbents2, and are the main focus of technological innovation, since all-in costs need to come down dramatically to make DAC cheap.
Next, the CO2 needs to be released from these materials. This process is called desorption, and usually takes place by heating up the materials and using a vacuum to suck out the CO2.
Climeworks stacks two arrays of fans together and cycles them through the capture (adsorb) and release (desorb) cycles, creating a closed-loop system that preserves heat:
Surprisingly, heat is a bigger energy input than electricity. While it seems like running the fans and compressing the CO2 — both dependent on electrical energy — could be the most important energy drivers, heat is ~80% of the energy used. That matters, because we want to use renewable energy in our DAC design, and neither solar nor wind give off much excess heat.
For this reason, some air-based DAC players are using geothermal energy, which takes advantage of the difference in heat between the surface of the earth and the magma underneath the earth’s crust to produce power. For obvious reasons, geothermal energy can create a lot of excess heat. Those DAC companies without access to geothermal typically use natural gas, since it also creates heat as a by-product.
If you’ve seen pictures of DAC pilot plants, they are usually of the air-based variety. This leads many folks to the intuition that air-based DAC, with its big fan arrays, is the only real DAC technology.
This leads us astray — it’s the equivalent of living in 1940 and believing that propellors are the best way to fly a plane simply because you’ve seen lots of planes with propellors. And even that analogy is not perfect, because there were many more propellor planes than any other type of plane in 1940. That’s not true with DAC, as none of these technologies have jumped out to a lead in real-world scale.
What we need to focus on is whether a technology solves the goals of Direct Air Capture:
Capture: does the technology capture CO2 from the atmosphere in a measurable way?
Speed: can the amount of CO2 captured be measured on the order of weeks or months, not decades?
Storage: does the technology create a carbon-rich solution that can be stored for centuries?
The other DAC solutions also meet these criteria. Next let’s look at liquid-based DAC.
LIQUID BASED DAC IS ON THE RISE
The next design was pioneered by Carbon Engineering, which was acquired for $1.1B by Occidental Petroleum in August 2023. Relative to Climeworks’ air-based DAC, water-based solutions have been subscale for years. But Carbon Engineering’s technology is taking off, and will be used in two DAC plant capable of capturing ~500,000 tons of CO2 per year each. This is more than 100x the size of the largest air-based plant operating today.
At a high level there are five steps:
The liquid-based design uses many intermediary chemicals — water alongside various potassium and calcium molecules — to create a looping reaction that separates CO2 from air. Despite this different approach, it has the same fundamental challenge as air-based solutions. Capturing the CO2 is easy, but separating CO2 from the capturing agent is hard, representing 80% or more of the total energy used in the system. Like the air-based tech, much of the energy needed is heat, which makes geothermal and natural gas more attractive energy sources than solar or wind.
The water-based design is promising for a few reasons:
Affordable materials: Carbon Engineering claims that its materials will be around $50/ton
Scaling potential: In chemical engineering, the ratio of surface area to volume is key for scaling up, and a liquid-based reaction allows engineers to tune this ratio more easily than in a gaseous one
This last bit is very important. Liquid-based DAC resembles typical chemical engineering processes that petrochemical firms are used to. This is one reason that Carbon Engineering is such a good fit with Oxy.
HEIRLOOM’S LIMESTONE APPROACH
Water is not the only medium that chemically pulls CO2 from the atmosphere. Calcium hydroxide, often called slaked lime, incorporates CO2 into its structure over the course of about 3 days. In a very basic chemical reaction it becomes limestone3
The folks at Heirloom Carbon created a neat process that takes advantage of this naturally occurring reaction. It also uses heat to break apart the limestone into it’s component parts, then adds water, which gets back to the slaked lime starting point:
As with air-based and liquid-based solutions, the real hard part is generating enough heat energy that can strip the CO2 away from the quicklime.
PLANT-BASED SOLUTIONS ARE DIFFERENT
A plant-based approach to carbon removal is quite different, in many ways, than the other technologies.
Biological capture: All the other solutions use chemical reactions to pull CO2 from the sky. Plant-based solutions use natural plant respiration and photosynthesis to do the same task.
Storage of something other than pure carbon dioxide: As we’ve seen, air- water- and limestone-based techs all run into the same chemical problem: separating CO2 from another compound is energy intensive. Plant-based solutions attempt to get around that by storing captured CO2 in an organic compound like biochar or bio-oil.
Decentralized: Unlike the large industrial processes, plant-based solutions are spread out over many farms, forests, etc.
Many different forms of this technology, and we will go into depth on one from Charm Industrial in just a minute. But first, a word on what plant-based DAC is not. This is not planting trees and waiting a decade for them to capture CO2.
What Charm Industrial is working on is quite different. It operates on the yearly agricultural cycle, and the outcome of Charm’s process is a quantifiable bio-oil, all of which has been stored underground to date. This is much more like DAC than a nebulous and easily falsifiable nature-based solution.
Here’s how Charm’s solution works:
A farm grow crops, like corn
Corn is removed from the plant, leaving behind the stalks
Stalks are collected in one place on the farm
A Charm Industrial truck with gassifier trailer drives to the farm
Stalks are fed into the gassifier at high heat
Out comes carbon-rich bio-oil and carbon-poor biochar
Biochar can be used in farming or other processes
Bio-oil is buried underground
Intuitively, you’d think that there’s a lot of excess orgranic material from farms, and Charm is (smartly) taking that waste off the farmers’ hands. But that’s not quite right. Corn stalks and other agri-waste, are often re-used as a key mixture in compost or as the actual structure of irrigation walls.
Charm may be having trouble getting a reliable supply of feedstock for the gassifier. They also have a lot of coordination costs, plus the logistics overhead of driving out to every farm, and transporting all the bio-oil via truck at a cost of greater than $10 per ton of CO24. All of this could mean an overly complex and potentially expensive business model.
It’s also worth noting that Charm’s bio-oil may change based on what crop(s) they are gassifying. This could be problematic for storage. Even if the company is confident that they can maintain a ‘good enough’ standard across all bio-oil batches, regulators may not agree, which would make permitting harder. This may already be happening.
Charm and others are definitely onto something, because the natural carbon cycle of plants is so powerful. Eventually someone will create a business model that cuts out excess complexity. Perhaps land dedicated to growing plants for storage, or even using vertical farming near a storage site.
FINAL THOUGHTS
We’re already seeing that individual DAC plants will leverage multiple technologies, like the Cypress plant that was just funded by the American Department of Energy.
This is an early indication that it’s unlikely that there is a ‘winner-take-most’ in Direct Air Capture. The job to be done by DAC is big enough for many companies, technologies, and business models to thrive.
Today all of these technologies cost more than ~$500 per ton of CO2 captured, and every DAC company is aiming to get below $100 per ton. Each tech will be subject to different scaling laws and experience curves, and these are mysteries today. And they all have room for a lot of innovation and process improvement.
With this much uncertainty, countries and VCs should imitate the American Department of Energy and bet on as many techs as possible.
Air-based designs are often referred to as ‘solid sorbent’ by DAC insiders, since the capturing material (sorbent) is a solid rather than a liquid
You read that right — air-based solutions adsorb, while liquid-based solutions absorb
And evaporated water
The biofuel is not actually CO2, so transportation costs would be ‘leveled’ to the bio-oil equivalent to CO2, or CO2e as it is sometimes shown in the scientific literature
The opinions in this article are my own and do not reflect the views of my employer, Bain & Company